Methodology document for the WHO e-atlas of disaster risk. Volume 1. Exposure to natural hazards Version 2.0
Preparation of the datasets
Dr Zine El Abidine El Morjani
Taroudant poly-disciplinary faculty of the Ibn Zohr University of Agadir, Morocco
Last Update: January 2011
Cataloguing-in-Publication Data Methodology document for the WHO e-atlas of disaster risk. Volume 1. Exposure to natural hazards, Version 2.0: Preparation of the datasets 1. Disasters. 2. Geographic Information Systems. 3. Risk Management. ISBN: 978-9954-0-5397-3
© Ibn Zohr University, 2011 All rights reserved. Publications of the Ibn Zohr University can be obtained from Ibn Zohr University, BP 32/S Agadir, Morocco (tel.: +212 528 22 71 25; fax: +212 528 22 72 60; e-mail:
[email protected]). Requests for permission to reproduce or translate Ibn Zohr publications – whether for sale or for noncommercial distribution – should be addressed to Ibn Zohr University, at the above contact details. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the Ibn Zohr University concerning the legal status of any country, territory, city of area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. All reasonable precautions have been taken by the Ibn Zohr University to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the Ibn Zohr University be liable for damages arising from its use.
2
Acknowledgements The development of the methodology for the preparation of the datasets used to generate the different hazard distribution maps is the product of contributions by several institutions and individuals. The funding to conduct the research leading to the development of the protocol has been provided by the World Health organization (WHO). The elaboration of the models was carried out by Zine El Abidine El Morjani, Taroudant poly-disciplinary faculty of the Ibn Zohr University of Agadir, Morocco and Steeve Ebener, WHO Mediterranean Centre for Health Risk Reduction, Tunisia. The data collection and analysis, implementation of the models, and documentation of the methodology were carried out by Zine El Abidine El Morjani.
3
Contents Acknowledgements.....................................................................................................................3 Preface
...................................................................................................................................7
1. Introduction.............................................................................................................................9 2. Required software and hardware ............................................................................................9 3. Vector layers .........................................................................................................................11 3.1 International boundaries and coastlines .............................................................................11 3.2 Road network .....................................................................................................................12 3.3 Hydrographic network .......................................................................................................13 3.4 Surface geology .................................................................................................................14 3.5 Tectonic..............................................................................................................................17 3.6 Location of weather stations ..............................................................................................20 3.7 Number of previous flood events.......................................................................................23 4. Raster layers..........................................................................................................................25 4.1 Digital Elevation Model (DEM) ........................................................................................25 4.2 Slope ..................................................................................................................................27 4.3 Aspect ................................................................................................................................29 4.4 Flow accumulation.............................................................................................................30 4.5 Land cover .........................................................................................................................32 4.6 Soil type .............................................................................................................................34 4.7 Soil texture .........................................................................................................................38 4.8 Annual maximum total precipitations over 3 consecutive days ........................................42 4.8.1 Extraction, preparation and pre-processing of the daily precipitation data ............42 4.8.1.1 Extraction and preparation of the daily precipitation data...............................43 4.8.1.2 Pre-processing of the daily precipitation data..................................................44 4.8.2 Calculation of the total precipitations for a given period of consecutive days and annual maximum total precipitation over 3 consecutive days for each weather station and year of observation....................................................................................................49 4.8.3 Calculation of the annual maximum total precipitations over 3 days for a two, five, height and ten years return period....................................................................................52 4.8.3.1 Creation of weather station specific files.........................................................52 4
4.8.3.2 Application of the Gumbel frequency analysis................................................53 4.8.4 Identification of the independent variables and selection of the regression model 60 4.8.4.1 Preparation of the GIS layers containing the spatial distribution of the causal factors and dependant variable.....................................................................................62 4.8.4.2 Integration of the annual maximum total precipitations over 3 consecutive days figures into the weather stations location GIS layer............................................66 4.8.4.3 Preparation of the stepwise regression analysis...............................................67 4.8.4. 4 Application of the stepwise regression analysis .............................................68 4.8.5 Spatialization of the predicted annual maximum total precipitations over 3 consecutive days ..............................................................................................................69 References and further reading .................................................................................................72 Annex 1. Classes observed for the geology of the region covered by this version of the e-atlas74 Annex 2. Procedure to clip a shapefile dataset to the borders of the region covered by this version of the e-atlas using the international boundary template (extracted from the SALB editing protocol).....................................................................................76 Annex 3. Procedure to assign the “no data” value to islands without data coverage within the region covered by this version of the e-atlas ..........................................................78 Annex 4. List of soil grouping types present in the region covered by this version of the eatlas .........................................................................................................................79 Annex 5: Procedure for converting a shapefile into a grid .......................................................80 Annex 6 - Description of the NCDC daily meteorological elements dataset ...........................81 Annex 7. Projection of a GIS layers into the metric projection system....................................84 Annex 8. Creation of a 300 km buffer around each climatic zone and clipping of the different layers for the regression analysis ............................................................................85 Annex 9. Final regression for a five year return period by climatic zone ................................87 Annex 10. Metadata for the e-atlas region datasets ..................................................................89 Annex 10.1. Metadata for the international boundaries layer..................................................89 Annex 10.2. Metadata for the road network distribution layer................................................91 Annex 10.3. Metadata for the hydrographic network distribution layer..................................93 Annex 10.4. Metadata for the surface geology distribution layer............................................95 Annex 10.5. Metadata for the tectonic layer............................................................................98 Annex 10.6. Metadata for the location of the weather stations .............................................100 Annex 10.7. Metadata for the number of previous flood events distribution layer ...............102 Annex 10.8. Metadata for the Digital Elevation Model layer................................................105 Annex 10.9. Metadata for the slope distribution layer...........................................................107 5
Annex 10.10. Metadata for the aspect distribution layer .......................................................109 Annex 10.11. Metadata for the flow accumulation distribution layer ...................................111 Annex 10.12. Metadata for the land cover distribution layer ................................................113 Annex 10.13. Metadata for the soil type distribution layer ...................................................115 Annex 10.14. Metadata for the soil texture distribution layer ...............................................118 Annex 10.15. Metadata for the annual maximum total precipitation over 3 consecutive days distribution layer for five year return periods ........................................................................120
6
Preface Being able to conduct geographically based risk assessment at the sub national level requires being in a position to spatially distribute all the elements reported in following conceptual formula1:
This process being very much driven by the type of hazard faced by the population and/or the key infrastructures in a given country the World Health Organization has been working, since 2006 on the development and improvement of an electronic atlas which could stimulate ministries of health and other health stakeholders to improve their disaster management capacity as well as serve as the entry point for conducting sub national geographically based risk assessments. The WHO e-atlas of disaster risk models the distribution of natural hazards and population’s exposure and provides baseline data and maps needed to advocate for resources to improve disaster preparedness; aid emergency response measures; and assist in identifying, planning and prioritizing areas for mitigation activities. The first version of the e-atlas published in 2008 covered the WHO Eastern Mediterranean Region (22 countries) and five natural hazards (flood, seismic [earthquake], landslide, heat and wind speed) and was distributed to more than 500 users. Encouraged by this success, working in close collaboration with the WHO Regions and taking advantage of the establishment of the Vulnerability and Risk Analysis and Mapping programme (VRAM), it was decided to publish a second version of the e-atlas that would, this time, also the 46 countries forming the WHO African Region as well as 32 countries of the WHO European Region (due to limited resources, this version of the e-atlas focuses on Central Europe only). Building on the successful collaboration established between the Taroudant polydisciplinary faculty of Ibn Zohr University, Agadir, Morocco and the VRAM, most of the models used in the first version of the e-atlas have been improved and heat replaced by heat wave, a current preoccupation of many ministries of health. In order to allow for any other region or country to also apply the models on their own it has been decided to document not only the research behind the models but also provide users with a protocol that would allow them to generate the final hazard distributions maps. The present series of methodology document is the result of this documentation.
1
Modified from: Office of the United Nations Disaster Relief Co-ordinator (UNDRO). Mitigating natural disasters. phenomena,
effects and options. A manual for policy makers and planners. New York, United Nations, 1991.
7
It is important to underline that the hazard distribution maps resulting from the application of these models are nevertheless only the first step of a process allowing countries to assess their risk at the sub national level. Analysing vulnerability and capacity require a process which is difficult to be applied at the level of a region for the following reasons: - availability of desegregated data - incompatibility of indicators from one country to an other - important differences in terms of health context between one country and another. WHO has therefore been looking at having the vulnerability, capacity and therefore indirectly risk analysis, conducted on a country by country basis. In this context, the VRAM is supporting Member States and partners to strengthen their capacity in order to conduct such analysis and have it presented in a manner such as the figure below.
The result of such analysis is then to be integrated in the country Disaster Risk Reduction (DRR) and Health Emergency Preparedness and Response Programmes (HEPRP) and serve, among other things, to build safer hospital, improve mass casualties’ management and help specialized units within health Organizations (including MoH) for public health planning. At the same time, the baseline data, information and maps collected or produced through the process can be used by health authorities and partners to take informed decisions in times of crises. 8
1. Introduction This document describes the source data used and the process applied to them in order to generate the dataset used to generate the different hazard distribution maps for the WHO eatlas of disaster risk, volume 1: exposure to natural hazards, Version 2.0. The primary objective of this work has been to produce a homogenous dataset covering the 46 countries forming the WHO African Region and the 22 forming the Eastern Mediterranean Regions as well as 32 countries of the WHO European Region (due to limited resources, this version of the e-atlas focuses on Central Europe only). The dataset has the following characteristics: • the scale of all vector datasets is 1:1 000 000; • the resolution of all raster datasets is 1 kilometre; • unprojected (geographic projection); • Esri shapefile or grid format; • all layers providing complete coverage of the study area (above mentioned 100 countries) to which a buffer of 300 km has been added in order to ensure a good result for the countries located at the border of the study area. In order to be able to distribute this dataset to third parties, efforts have been made to use as much as possible data in the public domain. The methods outlined in this document could be applied to other geographic areas than those already covered in this new version of the e-atlas.
2. Required software and hardware The implementation of the methods and processes described in this document requires the following software under Windows NT 4 (Service Pack 5, or 6a), Windows 2000 (Service Pack 3 or 4) and Windows XP (all versions) and all newer Windows version: ¾ ArcView 3.x with the Spatial Analyst 1.1 extension; both developed by the Environmental Systems Research Institute, (ESRI) Inc., for the extraction of the data columns to be processed and the geospatial operations. ¾ The following publicly available scripts and extensions which are accessible directly in the e-atlas DVD (in the tools section) have also been used: • • • • • • •
Grid Utilities v1.1 (file: Grid01.avx) Grid Analyst (GridAnalyst.avx) XTools (Xtoolsmh.avx) ImageWarp (ImageWarp.avx) Grid and Theme Projector v.2 (grid_theme_prj.avx) Hydrological Modeling v1.1 (hydrov11.avx) MapInfo conversion (convertMif.avx). 9
• Convert overlapping polygons (Theme.OverlappingPolys_2_grid_revised.ave) • Compiled_Table_Tools.avx
to
grid
script
These scripts should be saved on the computer in the C:\ESRI\AV_GIS30\ARCVIEW\EXT32 and then uploaded as extensions in ArcView before starting the process presented in the following sections. ¾ Matlab 6.0 (or higher), developed by MathWorks. Matlab has been used for the development of the EatlasClimMod 1.0 because software such as Excel can’t handle the large files used in the context of the present work (tables of approximately 160 000 lines) and there was a need for a platform for programming. ¾ The EatlasClimMod 1.0 application. This application has been developed under Matlab 6.0 to calculate and estimate the different climatic variables, including annual maximum precipitations for a five years return period, used in the context of this version of the eatlas. The codes of this application and the instruction file are all available in a zip file named EatlasClimMod.zip located in the tools section of the e-atlas DVD. ¾ A text editor such as WordPad to display the results tables. ¾ S-Plus 6.0, developed by Insightful Corporation, to explore and identify statistically significant parameters and their relative contribution to the spatialization of the precipitations using a stepwise multiple regression. The minimum and recommended hardware requirements for running Matlab and the EatlasClimMod 1.0 application are as follows: • • • •
Processor: Intel Pentium 3 and above 256 MB of RAM (512 MB or more is recommended) 600 MB of free hard drive space (1 GB is recommended) A colour graphics card and monitor (SVGA is recommended)
As a reference, EatlasClimMod 1.0 application has been used on the following computer configuration in the context of the WHO e-atlas project: • • •
processor Pentium Intel 4 (1.4 GHZ) 512 MB of RAM 2 Go of free hard drive space
It is therefore recommended, if possible to use a computer presenting these characteristics or better.
10
3. Vector layers 3.1 International boundaries and coastlines The 1:1 000 000 scale International Boundaries Dataset (IBD), produced by the international and administrative boundaries task group of the United Nations Geographic Information Working Group (UNGIWG), has been used as the source of reference for international boundaries and coastlines. This dataset is used for two reasons: • It allows the seamless integration of the international boundaries of countries as per the standards and practices used in the UN. This dataset is also the one which has been used in the context of the WHO e-atlas to delimitate climatic regions. • it allows the seamless integration of the delimitation of the subnational administrative boundaries produced in the context of the Second Administrative Level Boundaries data set project (SALB). Access to the IBD dataset is restricted to the UN. This dataset is therefore not available in the atlas. For more information and to get access to this dataset please consult: http://boundaries.ungiwg.org/ [Accessed December 15, 2010]. Access to the SALB dataset is public. The validated information can be downloaded directly from the SALB Project webpage at: http://www.unsalb.org/ [Accessed December 15, 2010]. The IBD dataset, st_int_bord.shp, is unprojected (geographic projection) and therefore requires a little work to allow its use in the context of the atlas, namely: • the merging of some polygons in order to make sure that each country is stored as one unique record in the attribute table • the preparation of two masks (polygon and line) in order to allow the correct representation of disputed areas and borders in question. The resulting international boundaries layer for the European Region is reported in Figure 2. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.1.
11
Figure 1. International boundaries for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk
3.2 Road network The extension of the road network over the study area has been extracted from the 1:1 000 000 scale Global Insight Plus dataset 2009 produced and distributed by Europa Technologies (http://www.europa-tech.com, [Accessed December 15, 2010]). Even if this dataset is not complete and up-to-date, it represents certainly the most recent and consistent dataset currently available on the market for this region. In this dataset, road segments are classified into the following categories: • motorway/highway • major trunk road • primary road • secondary road • city road • track, trail or footpath. The fact that some of these categories might overlap are due to differences in the nomenclature used by different countries. These data are not in the public domain and hence not part of the atlas. Refer to the Europa Technologies website mentioned above for information regarding the terms of use as well as the cost of this dataset. Since the data are unprojected, the global road network layer needed only to be clipped to the extent of the Region before its use. This was done as follows. 12
Make sure that the XTools extension is active in ArcView. Display both the global roads shapefile roads.shp and the international boundary layer of the study area in the active view st_int_bord.shp. Open the XTools>Clip with Polygon(s) function. Select the global roads shapefile as the theme that contains features that you wish to clip. Select the study area borders shapefile as being the theme that contains the polygons on which to clip the road layer. Specify the name for the new shapefile to be created, such as st_roads.shp. The resulting road network layer for the European Region is reported in Figure 2. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.2.
Figure 2. Road network for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk
3.3 Hydrographic network The extension of the hydrographic network over the study area has also been extracted from the 1:1 000 000 scale Global Insight Plus dataset 2009. This dataset is of better quality than the road network coming from the same source which has been verified by overlaying the road network on top of satellite images. 13
In this dataset, the network is separated into two layers: polygons to represent lakes, wetlands and wider parts of rivers; while lines are reserved for the representation of the narrow parts of rivers and streams. Both line and polygon features are classified into the following categories: • perennial • non-perennial. These data are not in the public domain and hence not part of the atlas. Refer to the Europa Technologies website mentioned in section 3.2 for information regarding the terms of use as well as the cost of this dataset. Since the data are unprojected, the global hydrographic network layer needed only to be clipped to the borders of the region as described in section 3.2 for the roads. This was done separately for the polygons and the lines to give respectively two files, named st_drain_p.shp and st_drain_l.shp. The resulting hydrographic network layer for the European Region is reported in Figure 2. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.3.
Figure 3. Hydrographic network for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk
3.4 Surface geology The surface geology layer for the study area has been compiled from six separate digital maps: Surficial geology of Africa, Bedrock geology of the Arabian peninsula and selected adjacent areas, Surficial geology of Iran, Geologic map of south Asia, Generalized geology of the former Soviet Union and Generalized geology of Europe including Turkey, all of these 14
maps being freely downloadable from the US Geological Survey (USGS) world energy website: http://certmapper.cr.usgs.gov/rooms/we/index.jsp [Accessed December 15, 2010]. This dataset is freely redistributable and can also be found in the data section of the WHO eatlas of disaster risk (Volume 1, version 2.0). Please acknowledge the atlas as noted in the metadata if you want to use it. The four geological maps which contribute to the region’s geology layer were not meant to provide continuous global coverage. They were prepared by a variety of sources and authors at different scales and based on different nomenclatures. As an indication the classes observed over the region can be found in Annex 1. In addition to that, the indication of the lithology is not always reported on these maps. This results for example in important discontinuities along the boundaries between the Islamic Republic of Iran and its neighbouring countries. Work has been done in order to combine the different maps into one unique layer that can be used in the context of the atlas. The following process was first applied to download all into Esri shapefile format and clipped to the borders of the study area. 1. Download the digital maps in Esri shapefile format from the following addresses on the USGS web site [Accessed December 15, 2010]: a. Surficial geology of Africa: http://certmapper.cr.usgs.gov/data/we/ofr97470a/spatial/shape/geo7_2ag.zip b. Bedrock geology of the Arabian peninsula and selected adjacent areas: http://certmapper.cr.usgs.gov/data/we/ofr97470b/spatial/shape/geo2bg.zip c. Surficial geology of Iran: http://certmapper.cr.usgs.gov/data/we/ofr97470g/spatial/shape/geo2cg.zip d. Geologic map of south Asia: http://certmapper.cr.usgs.gov/data/we/ofr97470c/spatial/shape/geo8ag.zip e. Generalized geology of the former Soviet Union: http://certmapper.cr.usgs.gov/data/we/ofr97470e/spatial/shape/geo1ec.zip f. Generalized geology of Europe including Turkey: http://certmapper.cr.usgs.gov/data/we/ofr97470i/spatial/shape/geo4_2l.zip.
2. Unzip each downloaded file to its own folder, and rename each shapefile using names that correspond to the source map (i.e. Africa_geol.shp, Arabian_geol.shp, Iran_geol.shp, South_asia_geol.shp, Soviet_geol.shp, Euro_geol.shp). 3. Clip each shapefile to the borders of the study area following the process outlined in Annex 2.
15
4. Save each clipped shapefile. Name each clipped shapefile according to its source shapefile (i.e. Africa_geol_clip.shp, Arabian_geol_clip.shp, Iran_geol_clip.shp, South_asia_geol_clip.shp, Soviet_geol_clip.shp, Euro_geol_clip.shp). It was then necessary to work on reducing the discontinuities and differences that existed in term of nomenclature between the six maps. In particular, the following procedure was applied in order to group some of the classes used in the Islamic Republic of Iran to obtain a better match with the more simplified nomenclature used for the neighbouring countries. This was done using the following process. 1.
In ArcView, display the Arabian_geol_clip.shp, South_asia_geol_clip.shp and Iran_geol_clip.shp layers in the active view.
2.
Activate the Iran_geol_clip.shp layer.
3.
Visually detect discontinuities appearing at the border between the Islamic Republic of Iran and the neighbouring countries and see if it is possible to group some of the classes used in the Islamic Republic of Iran to obtain a correspondence with the classes used in the neighbours. This work was done by a geologist in order to ensure the quality of this operation.
4.
Modify the class of the polygon concerned in the Islamic Republic of Iran in order to make it correspond to the nomenclature observed in the neighbouring countries.
5.
Use the Xtools>Merge Themes function to merge the clipped shapefiles into a single geological layer.
6.
Save the output as st_ar_geol_merge.shp.
7.
Use the procedure outlined in Annex 3 to assign “no data” to small islands that are not covered by the original surface geology maps within the borders of the study area
8.
Save the result as st_ar_geol_nd.shp.
Finally, the following procedure was used to group into a single feature the multiple polygons presenting the same class. 1.
Open the attribute table of the st_ar_geol_nd.shp shapefile.
2.
Select the header of the column containing the classes to be used for the merge (“geology_name”).
3.
Click on the “summarize” icon ; add the “shape” field and specify st_geology.dbf as the name of the new dbf file. Click OK.
16
The resulting geology layer for the European Region is reported in Figure 4. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.4.
Figure 4. Surface geology for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
3.5 Tectonic The tectonic layer (land and sea) was extracted from the Digital Tectonic Activity Map (DTAM) created by the United States National Aeronautics and Space Administration at the Goddard Space Flight Center near Washington DC. The DTAM is used as a digital atlas of tectonism and volcanism of the past one million years and is drawn using the Robinson projection. It locates and classifies various types of geological fault zones, active spreading centres and volcanic centres active within that period. The DTAM was created using numerous remote sensing data sources and geospatial databases (seismicity, volcanism and plate motions) and is a unique tool for understanding the physiographic nature of the Earth. This map can be freely downloaded in jpg, gif, pdf and tif formats from: http://denali.gsfc.nasa.gov/dtam/data.html [Accessed December 15, 2010]. The following five-step procedure was applied to create an unprojected layer in GIS format to be used in the atlas.
17
1.
Download the 2002 Digital Tectonic Activity Map (DTAM) in tif format (DTAM.tif) from the above mentioned web site.
2.
In ArcView, use steps a to e below to project the global international boundaries layer (section 3.1) from the geographic projection into the Robinson one: a. make sure that the Grid and Theme Projector v.2 and ImageWarp extensions are active in ArcView b. add the International Boundaries Dataset (IBD) shapefile (Un_cntrypoly_01.shp) produced by the UN Cartographic Section (See section 3.1) and the DTAM.tif file as an image c. click either the button or use the Grid Projector>Grid>Theme Projector function and select Un_cntrypoly_01.shp from the list as the theme to project. d. in the Theme Projector window: - specify the parameters for the current projection as follows: Category = projection of the world Type = Geographic Current Projection Units = decimal degrees - specify the parameters for the new projection as follows: Category = projection of the world Type = Robinson New Projection Units = meters and click OK e. save the resulting shapefile as Global_int_bord_rob.shp.
3. Use the coastline of the reprojected version of the International Boundaries Dataset as a reference to warp the DTAM.tif image using the ImageWarp extension as follows. a. make sure that the ImageWarp extension is active in ArcView b. select the ImageWarp>ImageWarp function c. in the ImageWarp Session Setup window that appears, select DTAM.tif as the image to be rectified and the International Boundaries Dataset as the theme to serve as the reference for the warping; press OK to continue d. in the next dialogue box, which ask “do you want to set the projection for to view?”, answer “No” and move to the next window e. in the Ground Control Point table, click the New GCP Table radio button and select the name for the new table that will contain the list of the point of the reference for the work; click OK to continue. The reprojection process is started, and the image will appear in the FROM view, while the shapefile will appear in the TO and TO*** ROAM views f. in the TO view, zoom in to a place in the region and put a first reference point on an international border g. when this is done, zoom in to the same place in the FROM view h. select the GCP Pick tool in the icon window and put a first point in the TO view at a place that will be the reference and indicate if you want to keep the point just entered in the next window i. put a point at the corresponding place in the FROM view j. complete the same operation explained above for two other points 18
k.
l. m. n.
o.
p.
after having selected three GCP pairs, put a new point in the TO view and then click the Compute From button in the icon window. This will automatically place a point in the corresponding location in the FROM view click on the GCP Select tool and using the pointer select this newly created point in the FROM view select the Move GCP tool and click at the place where the fourth point should have normally been located. This will move the fourth point to its correct location repeat until you have a sufficient number of points on the map or when the results of the previous step does in fact locate a point in the FROM view exactly where it should have been. Once you are happy with your points of reference, click on the Calculate RMS icon and choose the third order of calculation. An information window will appear indicating the RMS (root mean square) error for the output map. The RMS error is an indication of the level of geometric distortion that has not been corrected measured by the discrepancy between the locations of the points in the reference layer and the location of the same points in the layer that has been warped. The lower the RMS error, the more accurate the transformation if you are happy with the result, click on the Go button to generate the new warped image. You also need to choose the resampling method (use nearest neighbour, bilinear interpolation or cubic convolution), the output image type (tif, jpg, bil, bsq or bip) and the name for the new image (e.g. DTAM_warp.tif) if the fit between the warped image and the reprojected International Boundaries Dataset is unsatisfactory, restart the process from step a. If you are satisfied you can go to step 4.
4. Use the following steps to create a line shapefile for each of the different types of fault that appear on the warped image (reverse, normal, major active and transform) for the study area only (this does not need to be done for the total spreading rate and active volcanic centres): a. add the DTAM_warp.tif file to the active view. b. create a new line themes using the View>New Theme function and by selecting “line” in the scroll down menu c. manually digitize all the faults of the first type d. save the editing work using a name that will be specific to the fault type e. repeat for the other types of fault until you have four vector shapefiles (reverse fault, normal fault, major active fault and transform fault). 5. Use the following process to merge the shapefiles containing the different fault types and create and clip them to the borders of the Region using the following process. a. make sure that the XTools extension is active in ArcView b. select the Xtools>Merge Themes function to create one unique shapefile from the four created previously and save the result as tectonic_merge.shp c. add the study area international boundaries shapefile (st_int_bord.shp) in the active view d. select the Xtools>Clip with Polygon(s) function
19
e.
f. g. h.
select tectonic_merge.shp as the theme that contains the features you wish to clip and st_int_bord.shp as the polygon theme that contains the polygons to use as the reference for the clipping name the shapefile to be created as st_ar_tectonic_rob.shp use the Grid and Theme Projector v.2 extension to unproject the st_ar_tectonic_rob.shp from the Robinson projection to the geographic one save the result as all_tectonic.shp.
The resulting tectonic layer for the European Region is reported in Figure 5. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.5.
Figure 5. Tectonic layer for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
3.6 Location of weather stations The locations of the weather stations in the region covered by this version of the e-atlas were extracted from the Global Surface Summary of the Day Dataset produced by the US National Climatic Data Center (NCDC). These data can be freely downloaded from ftp://ftp.ncdc.noaa.gov/pub/data/inventories/ [Accessed December 15, 2010]. The original data are distributed as an ASCII text file, listing all weather stations authorized to report to the World Meteorological Organization (WMO). Weather station attribute data include WMO station number, station name, country/state ID, latitude, longitude and elevation. 20
The following procedure was used to create a shapefile containing the location of the climatic stations located within the study area and those not further than 300 km away in order to ensure good interpolation within the region. 1.
Download the file containing the list of stations sorted by country (Ish-history.txt) from the ftp site mentioned above.
2.
Open the Ish-history.txt in a text editor (e.g. Microsoft Office Word) and delete the header (all lines before the column headings of the table) and save the changes.
3.
Open the modified version of Ish-history.txt in a spreadsheet program (e.g. Microsoft Office Excel) and save it as a DBF 4 (dBASE IV) (*.dbf) table with the name stnlistsorted.dbf to convert the text file into a format recognized by ArcView.
4.
Use the following process to convert stnlist-sorted.dbf into a shapefile: a. launch ArcView and add the stnlist-sorted.dbf file in the table window b. open the view in which the table will be mapped. Use the View>Add Event Theme function, choose the stnlist-sorted.dbf file in the windows that opens and specify LON as the X field and LAT as the Y field c. convert the Event Theme to a shapefile by using the Theme>Convert to Shapefile function d. save the result as stations.shp. The resulting layer should contain at least 30 166 stations (depending upon the publication date of the station list and the number of stations that have been added or removed).
5. Use the following process to create a 300 km buffer around the region. a. in View>Properties change the Distance Units to kilometres b. upload the study area international borders layer (st_int_bord.shp) in the view c. select the Theme>Create Buffer function and: - in the first dialogue window, choose st_int_bord.shp as the item to buffer - in the second dialogue window, select “At a Specified Distance” as the method used to create the buffer and enter “300” as number for the width of the buffer. Make sure that the distance units are set to kilometres - in the third dialogue window, create the buffer using the “only inside of the polygon parameter”. Specify that the buffer be saved as a new theme and assign a name to this layer (i.e. st_int_bord_buffer_300.shp). The new buffer theme will automatically be added to the current view.
21
6. Use the following process to select the stations in the buffered region covered by this version of the e-atlas: a. make sure that the XTools extension is active in ArcView b. open the Xtools>Clip with Polygon(s) function c. select stations.shp as the theme that contains features that you wish to clip d. select st_ar_int_bord_buffer_300.shp as the polygon theme that contains the polygons that will be used as the reference for the clipping e. name the new shapefile st_stations. The resulting weather station location layer for the European Region is reported in Figure 6. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.6.
Figure 6. Location of the weather stations for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
22
3.7 Number of previous flood events The distribution of the number of previous flood events was extracted from the flood polygons compiled by the Dartmouth Flood Observatory between 1985 and 2009 in its Global Active Archive of Large Flood Events. This archive is derived from a wide variety of news, government and direct and remote sensing sources and can be freely downloaded in MapInfo format from: http://floodobservatory.colorado.edu/Archives/index.html [Accessed December 15, 2010]. In order to generate the distribution of the number of previous flood events for the study area it was necessary to combine the polygons for each observation using the following process: 1.
Download the two global archive flood MapInfo files (FloodArchive.MIF and FloodArchive.MID for all floods between 1985 and 2009 from the Dartmouth Flood Observatory website mentioned above.
2.
Make sure that the MapInfo Conversion and Xtools extensions are active in ArcView.
3.
Click on the button containing the blue diamond conversion process.
4.
In the MapInfo Conversion dialogue box that opens, select the global flood event MapInfo file FloodArchive.MIF and click OK.
5.
In the next dialog box enter the name of the shapefile to be created as FloodArchive.shp.
6.
Upload the study area international boundary template st_int_bord.shp in the view.
7.
Use the XTool>Clip with Polygon(s) function to clip the FloodArchive layer to the extension of the region covered by this version of the e-atlas and save the output file as st_ar_flood.shp.
8.
Open the attribute table of the clipped layer st_ar_flood.shp and start editing it using Table>Start Editing.
9.
Add a field named “frequency”.
in the project window to start the
10. Set the default value of “frequency” to “1” for each record. This field indicates that there is one flood event for the individual polygon. It will be summed with other frequencies from other polygons to yield the flood frequency for a specific location. 11. Sum and convert the overlapping polygons to grid by using the script called Theme.OverlappingPolys_2_grid_revised.ave,, created by D . Jann and revised by J. Ardron, using the following steps: 23
a. open a New Script window in the Project window b. Load “Theme.OverlappingPolys_2_grid_revised.ave” script into this new script window by going to Script Menu>Load Text c. Activate the shapefile st_ar_flood.shp d. Compile and run the overlapping polygons to grid script by pressing respectively the Compile and Run buttons e. In the next window select “Same as st_int_bord.shp” as the “Output Grid extent” in order to obtain a grid presenting the same extent as the original shapefile f. for the Output Grid cell size choose 0.008333 degrees. This resolution corresponds to 1 km at the equator, the resolution selected for the e-atlas g. do not modify the default Number of Rows and Number of Columns values and click OK h. Choose “frequency” field as cell value i. The resulting grid named st_flood_fr contains the value representing the sum of the overlapping polygons. The resulting number of previous flood events distribution layer for the European Region is reported in Figure 7. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.7.
Figure 7. Distribution of the number of previous flood events for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
24
4. Raster layers 4.1 Digital Elevation Model (DEM) The Shuttle Radar Topography Mission (SRTM) global elevation data covers almost 80% of the globe, almost all terrestrial land surfaces. Its coverage extends between 60° north and 56° south latitudes. SRTM is a joint project between the United States National Aeronautics and Space Administration (NASA) and the Department of Defense’s National Geospatial Intelligence Agency (NGA) to produce near-global digital elevation data coverage at a relatively high spatial resolution. The data is handled and distributed by the United States Geological Survey and can be downloaded from their web site http://dds.cr.usgs.gov/srtm/version2_1/. The basic product has a 1 arc-second resolution, but it is publicly available for the continental US only by agreement with the NGA. This data was resampled to 3 arc–second resolution by averaging a 3 by 3 cell area. This 3 arc-second resolution data has now been released for the entire terrestrial surface, and a “Finished” product can be found at http://www2.jpl.nasa.gov/srtm/cbanddataproducts.html [Accessed December 15, 2010]. It is used for the WHO e-atlas studies because it provides a major advance in the accessibility of high quality elevation data. The resolution of this data at the equator is approximately 90 metres, but has been resampled to 1 km resolution to correspond to the characteristics of the project. The following steps were applied in order to extract the elevation distribution grid covering the e-atlas. 1. Download the resampled SRTM 90 data to 1 km resolution from the CGIAR-CSI web site http://srtm.csi.cgiar.org/ [Accessed December 15, 2010].or http://srtm.csi.cgiar.org/SELECTION/inputCoord.asp [Accessed December 15, 2010] 2. Unzip the downloaded file SRTM_1km_TIF.rar 3. Add the image SRTM_1km.tif to the active view 4. Make sure that the Spatial Analyst extension is active in ArcView 5. Convert the image SRTM_1km.tif to a grid file by using the Theme>Convert to Grid option. The outputs is saved as SRTM_1km 6. Activate the Analysis>Map Calculator and enter the following formula in the box: ([SRTM_1km] >= 32768).con ([SRTM_1km] - 65536, [SRTM_1km]). The resulting grid is named “Map Calculation “. This calculation corrects the misinterpretation of all data in the original .tif image as unsigned integer data (all elevations are above sea level) and subsequently allows for negative elevations (below sea level) 7. Set the ocean mask values to NODATA using the Analysis>Map Calculator option and entering the following formula in the box: ([Map Calculation] = -32768).setnull([Map Calculation]) 25
8. Save the result as srtm_clean 9. Clip the srtm_clean to the borders of the study area as follows: a. add the st_int_bord.shp layer to the active view. b. activate srtm_clean and select the Grid Analyst>Extract Grid Theme Using Polygon function c. click “yes” to continue. d. select st_int_bord.shp from the dropdown list to be used as the layer on which the grid needs to be clipped and click OK e. make the output grid the active theme and choose the Theme>Convert to grid function to create the final DEM f. save it as st_dem. The resulting DEM for the European Region is reported in Figure 8. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.8.
Figure 8. DEM for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
26
4.2 Slope The slope distribution layer as been derived from the Digital Elevation Model (DEM) presented in section 4.1 by calculating the maximum rate of change between each cell and the eight adjacent ones. As the DEM used is already unprojected and presents a good resolution, only the following steps were applied in order to create the slope distribution grid. 1.
Use steps a to d to project the st_dem (the elevation layer) into the Equal-Area Cylindrical projection to allow the horizontal and vertical dimensions to be expressed in the same units (metres). If the dimensions are measured in different units the derived slope measurements will be incorrect: a. make sure that Grid and Theme Projector v.2 and Spatial Analyst extensions are active in ArcView b. click either the button or use the Grid Projector>Grid and Theme Projector functions c. select st_dem from the list as the grid to project d.
In the Grid Projector window: - specify the parameters for the current projection as follows: - Category = projection of the world - Type = Geographic - Current Projection Units = decimal degrees - specify the parameters for the new projection as follow: - Category = projection of the world - Type = Equal-Area Cylindrical - New Projection Units = meters
e.
in the next window, enter the new cell size as “1000”.
f.
save the output grid as st_ar_dem_m.
2. Activate the Map Calculator from the Analysis menu and: a. create the slope grid in degrees by entering the following formula in the box: [st_ar_dem_m].slope(Nil,FALSE) b. save the output as st_ar_slope_m_d c. create the slope grid in percent by entering the following formula in the box: [st_ar_dem_m].slope(Nil,TRUE) d. save the output as st_ar_slope_m_pr. 3. Unproject the st_ar_slope_m_d and st_ar_slope_m_pr layers from the Equal-Area Cylindrical projection to the geographic one using the following steps: a. click either the button or use the Grid Projector>Grid and Theme Projector function
27
b.
select either st_ar_slope_m_d or st_ar_slope_m_pr from the list as the grid to project.
4. In the Grid Projector window: a. specify the parameters for the current projection as follows: - Category = projection of the world - Type = Equal-Area Cylindrical projection - Current Projection Units = meters b. specify the parameters for the new projection as follows: - Category = projection of the world - Type = Geographic - New Projection Units = decimal degrees 5. In the next window, specify the new cell size as 0.008333. This resolution corresponds to 1 km at the equator, the resolution selected for the e-atlas. 6. Save the output grids as st_slp_dd for the slope in degrees and st_slp_pr for the slope as a percentage. The resulting slope layer (both degrees and percentage) for the European Region is reported in Figure 9. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.9.
Figure 9. Slope layer for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk These two datasets are freely redistributable; they can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use them. 28
4.3 Aspect The aspect distribution layer was derived from the Digital Elevation Model (DEM) presented in section 4.1 by calculating the downslope direction of the maximum rate of change in the elevations between each cell and its eight neighbours. It can essentially be thought of as the direction of the maximum slope and is measured in positive integer degrees from 0 to 360, measured clockwise from north. Cells of zero slope (flat areas) are assigned an aspect value of –1. Like the slope layer, the creation of this dataset requires that both horizontal (x, y) and vertical (z) dimensions be expressed in the same units (metres). If the dimensions are measured in different units the derived aspect measurement will be incorrect. As the DEM used is already unprojected and presents a good resolution, only the following steps were applied in order to create the slope distribution grid: 1.
Activate the reprojected elevation layer created in section 4.2 (st_ar_dem_m).
2.
Use the Surface>Derive Aspect function to create the desired grid and save it as st_ar_aspect_m.
3.
Unproject the st_ar_aspect_m from Equal-Area Cylindrical projection to geographic projection by using the process described under step 3 in section 4.2.
4.
Name the final output grid st_asp.
The resulting geology layer for the European Region is reported in Figure 10. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.10. This layer is freely redistributable; it can be found in the data section of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
29
Figure 10. Aspect for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk
4.4 Flow accumulation The flow accumulation distribution layer could be derived from the Digital Elevation Model (DEM) by calculating the amount of upstream area that drains into each cell using the same process described in section 4.4 of the Methodology and implementation process for generating the dataset document that can be found in the first volume of the WHO e-atlas of disaster risk for the Eastern Mediterranean Region, version 1.0. In this version, however, the flow accumulation layer from HydroSHEDS (Hydrological data and maps based on SHuttle Elevation Derivatives at multiple Scales) has been used because it is derived from SRTM 90 meter resolution DEM and therefore offering a much precise suite of geo-referenced data sets (vector and raster), including stream networks, watershed boundaries and drainage directions, and ancillary data layers such as flow accumulation, distance and river topology information than the 1 km DM presented in section 4.1. It has been developed by the Conservation Science Program of World Wildlife Fund (WWF), in partnership with the US Geological Survey; the International Centre for Tropical Agriculture (Colombia); the Nature Conservancy; the government of Australia; McGill University, Montreal, Canada; and the Center for Environmental Systems Research (CESR) of the University of Kassel, Germany. The flow accumulation maps can be freely downloaded in Esri grid format from: http://hydrosheds.cr.usgs.gov [Accessed December 15, 2010].
30
The study area of the region covered by this version of the e-atlas is covered by using part of three map sheets: Africa, Europe and Asia. The following procedure was applied to derive the flow accumulation grid for the region covered by this version of the e-atlas. 1.
Download the Africa, Europe and Asia flow accumulation digital maps (af_acc_30s.zip, eu_acc_30s.zip, as_acc_30s.zip) in Esri grid format at 30 arc-seconds from the USGS web site http://hydrosheds.cr.usgs.gov [Accessed December 15, 2010].
2. Unzip each downloaded file to its own folder to extract af_acc_30s, as_acc_30s and eu_acc_30s. 3. In ArcView, upload these extracted grids in a view. 4. Use the Transform Grid>Mosaic function to merge af_acc_30s, as_acc_30s and eu_acc_30s and create a grid where each cell corresponds to the minimum distance from the drainage network. 5. Use the Theme>Save Data Set function to save the resulting grid as af_as_eu_fa. 6. Use the following steps to clip the af_as_eu_fa to the study area international boundaries: a. make sure that the Grid Analyst extension is uploaded in ArcView b. upload the international boundary level used for the e-atlas (st_int_bord.shp) c. make the af_as_eu_fa the active theme and use the Grid Analyst>Extract Grid Theme Using Polygon function. d. select the st_int_bord.shp from the drop list as the layer to be used as the reference for the clipping. e. make the new output grid the active theme and use the Theme>Convert to Grid function saving it as st_fa. The resulting flow accumulation distribution for the European Region is reported in Figure 11. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.11.
31
Figure 101. Flow accumulation for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
4.5 Land cover The land cover data used in this version is generated by European Space Agency’s GlobCover project. This GlobCover land cover map is based on ENVISAT’s Medium Resolution Imaging Spectrometer (MERIS) Level 1B data acquired in full resolution mode with a spatial resolution of 300 metres for the period December 2004–June 2006. Its 22 land cover classes are defined using the UN Land Cover Classification System (LCCS). This product is available though two access points [Accessed December 15, 2010]: - the ESA GlobCover web site (http://ionia1.esrin.esa.int). - the Pôle d’Observation des Surfaces continentales par Télédetection (POSTEL) web site (http://postel.mediasfrance.org). In order to display the GlobCover map for the study area, the following steps were applied. 1. Register to the GlobCover web site at: http://ionia1.esrin.esa.in 2. Once registered, login and download the zip file containing a raster version of the GlobCover global land cover map for the period December 2004–June 2006, from the ftp site: ftp://postel.mediasfrance.org, 3. Unzip the downloaded file to extract:
32
a. GLOBCOVER_200412_200606_V2.2_Global_CLA.tif: this is the full resolution data in GEOTIFF format, where the ID values stand for the land cover classes values b. GLOBCOVER_200412_200606_V2.2_Global_CLA.tif.vat.dbf: this dbf file stores the attribute table of the raster c. Glocover_Legend.xls: this Microsoft Excel file contains the legend of the global land cover map. The ID values of the GEOTIFF raster are linked with the corresponding land cover labels and RGB codes d. Globcover_Global_Legend.lyr/.avl/.dsr: these files contain the colour map of the GlobCover global land cover, in an ArcInfo format (.lyr), in an ArcView format (.avl) and in an Envi format (.dsr). 4. Add the image GLOBCOVER_200412_200606_V2.2_Global_CLA.tif to the active view. 5. Make sure that the Spatial Analyst and Grid Analyst extensions are active in ArcView. 6. Convert this image to a grid file by using Theme>Convert to grid option. The output is saved as glob_cove_300; This grid has a resolution of 300 m. 7. Resample this grid by using the Analysis>Resample option, and enter 0.008333 as output cell size in the next window. This resolution corresponds to 1 km at the equator, the resolution selected for the e-atlas. 8. Save the result as glob_cover 9. Clip the glob_cover to the borders of the study area as follows. a. add the st_int_bord.shp layer to the active view. b. activate glob_cover and select the Grid Analyst>Extract Grid Theme Using Polygon function c. click “yes” to continue d. select the st_int_bord.shp as the layer on which the grid needs to be clipped from the drop list, and click OK e. make the output grid the active theme and choose the Theme>Convert to Grid function to generate the final land cover grid for the study area f. save the output grid as st_lc. 10. Double-click on the st_lc grid theme in the table of content to open the legend editor. 33
11. Load the Globcover_Global_Legend.avl legend file and choose as field “value”. Each GlobCover class is now associated with its Globcover colour code. The resulting land cover distribution layer for the European Region is reported in Figure 12. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.12.
Figure 112. Land cover distribution for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
4.6 Soil type The soil type used is the 30 arc-second resolution (0.008333 degrees or 1 km at the equator) Harmonized World Soil Database (HWSD) version 1.1 produced in 2009 by the International Institute for Applied System Analysis (IIASA) in partnership with ISRIC–World Soil Information, the Food and Agriculture Organization of the United Nations (FAO), the European Soil Bureau Network and the Institute of Soil Science, Chinese Academy of Sciences. Four source databases were used to compile this version of the HWSD: the European Soil Database (ESDB), the 1:1 000 000 soil map of China, various regional SOil and TERrain database (SOTER) and the FAO–UNESCO Soil Map of the World. This last was used in version 1.0 of the WHO e-atlas. The HWSD is composed of a GIS image file that can be linked to an attribute database in Microsoft Access format. These two components are separate data files that can be linked through ArcView.
34
The HWSD attribute database provides information on the soil unit composition for each soil mapping unit. The database shows the composition of each soil mapping unit and standardized soil parameters for top- and subsoil. A soil mapping unit can have up to nine soil unit/topsoil texture combination records in the database1. The core fields for identifying a soil mapping unit are: • • • •
MU_GLOBAL (Global Mapping Unit Identifier): the harmonized soil mapping unit identifier of HWSD providing the link between the GIS layer and the attribute database MU_SOURCE1 and MU_SOURCE2: the mapping unit identifiers in the source database SEQ: the sequence of the soil unit in the soil mapping unit composition Soil unit symbol using the FAO-74 classification system (SU_SYM74) or the FAO90 classification system (SU_SYM90) or FAO-85 interim system (SU_SYM85), and the (SU_SYMBOL) symbol stands for the dominant major HWSD soil group. It is used here to spatialize the main HWSD. The major soil groupings used for the HWSD map are given in Annex 4
•
T_USDA_TEX_CLASS: topsoil texture (0–30 cm) classified into 13 classes according to USDA and FAO soil texture classifications
•
S_USDA_TEX_CLASS: subsoil texture (30–100 cm) classified into 13 classes according to USDA and FAO soil texture classifications
•
T_TEXTURE: topsoil texture containing the three classes (coarse, medium and fine) for 0–30cm.
The full contents of the HWSD database in Microsoft Access (general information on the soil mapping unit composition, information related to phases, physical and chemical characteristics of topsoil (0–30 cm) and subsoil (30–100 cm), and the HWSD map (in .bil format) can be downloaded from http://www.iiasa.ac.at/Research/LUC/External-World-soildatabase/HTML/HWSD_Data.html?sb=4 [Accessed December 15, 2010]. In order to obtain the distribution of the soil type in Esri grid format, the following procedure was followed. 1. Download the zip file HWSD_RASTER.zip containing a raster soil map in .bil file format, the HWSD.mdb file containing the soil attribute database in Microsoft Access 2003 format and HWSD_META.mdb containing the Soil Attribute Database metadata from the International Institute for Applied Systems Analysis (IIASA) web site: http://www.iiasa.ac.at/Research/LUC/External-World-soildatabase/HTML/HWSD_Data.html?sb=4 [Accessed December 15, 2010]. 2. Unzip the downloaded file HWSD_RASTER.zip to extract the 1 km resolution HWSD image raster file format in .bil format hwsd.bil.
1
FAO/IIASA/ISRIC/ISS-CAS/JRC, 2009. Harmonized World Soil Database (version 1.1). FAO, Rome, Italy and IIASA, Laxenburg, Austria. (http://www.iiasa.ac.at/Research/LUC/External-World-soildatabase/HTML/HWSD_Data.html?sb=4)
35
3. In ArcView add the image hwsd.bil to the active view. 4. Make sure that the Spatial Analyst and Grid Analyst extensions are active in ArcView. 5. Convert this image to a grid file by using Theme>Convert to Grid option. Save the output as glob_hwsd. 6. Clip glob_hwsd to the borders of the study area as follows. a. add the st_int_bord.shp layer to the active view b. activate glob_hwsd and select the Grid Analyst>Extract Grid Theme Using Polygon function c. click “yes” to continue d. select the st_int_bord.shp as the layer on which the grid needs to be clipped from the drop list, and click OK e. make the output grid the active theme and choose the Theme>Convert to grid function to generate the final land cover grid for the study area f. save the output grid as st_ar_ hwsd. 7. In Microsoft Access, open the soil attribute database HWSD.mdb. 8. Using the Objects>Tables option, open the following tables: a. HWSD_SMU with fields: MU_GLOBAl (harmonized soil mapping unit identifier of HWSD (global)), SU_Symbol (soil mapping unit symbol), and SU_ code (soil mapping unit code) b. D_symbol with fields: Code (soil mapping unit code), Symbol (soil mapping unit symbol), and Value (name of the harmonized soil mapping unit). 9. Using File>Export, export these Access tables into .dbf format and save as HWSD_database.dbf and HWSD_name.dbf. 10. Link the grid HWSD Emro soil mapping st_ar_ hwsd to harmonized attribute database HWSD_database.dbf as follows: a.
make sure that the Compiled Table Tools extension is uploaded in ArcView
b.
add the st_ar_ hwsd grid in the view and open its attribute table
c.
in the ArcView project, select Table and add HWSD_database.dbf 36
d.
find the column in common between the st_ar_ hwsd attribute table (Value) and the HWSD_database.dbf one (MU_GLOBAl)
e.
highlight first the header of the MU_GLOBAL column in the HWSD_database.dbf and then the header of the Value column in st_ar_ hwsd by clicking on them
f.
keeping the attribute table of the st_ar_ hwsd file activated, join the two tables by clicking the join button on the tool bar
g.
apply the C-Tables Tools>Make joins permanent function. This grid contains the symbol and code information for each soil mapping unit.
11. In order to associate to this grid the name of the harmonized soil mapping unit, use the following procedure: a.
add the st_ar_ hwsd grid in the view and open its attribute table
b.
in the ArcView project, select Table and add HWSD_name.dbf
c.
find the column in common between the st_ar_ hwsd attribute table (SU_ code) and the HWSD_name.dbf one (Value)
d.
highlight first the header of the SU_ code column in the HWSD_name.dbf and then the header of the Value column in st_ar_ hwsd by clicking on them
e.
keeping the attribute table of the st_ar_ hwsd file activated, join the two tables by clicking the join button on the tool bar
f.
apply the C-Tables Tools>Make Joins Permanent function. This grid contains in the symbol, code and name information for each soil mapping unit
g.
convert the HWSD name to a new grid by using the Analysis>Map Calculator option and entering the following formula in the box: ([st_ar_ hwsd.value])
h.
Save the result as st_soil_type. This layer is organized in 33 groups.
The resulting soil type layer layer for the European Region is reported in Figure 13. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.13.
37
Figure 13. Soil types for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk
4.7 Soil texture Soil texture is used to describe the relative proportion of different grain sizes of mineral particles in a soil. Particles are grouped according to their size into what are called soil separates (clay, silt and sand). The soil texture classification corresponds to a particular range of separate fractions, and is diagrammatically represented by United States Department of Agriculture (USDA) soil texture triangle (http://en.wikipedia.org/wiki/File:SoilTextureTriangle.jpg ). In the e-atlas, the topsoil (0–30 cm) and the subsoil (30–100 cm) texture layers are also derived from the Harmonized World Soil Database version 1. These layers are classified into 13 classes according to USDA and FAO soil texture classifications, namely: • • • • • • • • • • • • •
clay (heavy) silty clay clay silty clay loam clay loam silt silt loam sandy clay loam sandy clay loam sandy loam loamy sand sand. 38
The topsoil texture can be classified into the three simplified textural classes used in the FAO/UNESCO Soil Map of the World: •
coarse textured: sands, loamy sands and sandy loams with less than 18 percent clay and more than 65 percent sand
•
medium textured: sandy loams, loams, sandy clay loams, silt loams, silt, silty clay loams and clay loams with less than 35 % clay and less than 65 % sand; the sand fraction may be as high as 82 percent if a minimum of 18 percent of clay is present
•
fine textured: clays, silty clays, sandy clays, clay loams and silty clay loams with more than 35 percent clay.
These textural classes reflect the relative proportions of clay in the soil (fraction less than 0.002 mm), silt (0.002–0.05 mm) and sand (0.05–2 mm). When there is no soil on the surface the following nomenclature appears in the dataset: DS = sand dunes, ST = salt flats, RK = rock debris; GG = glaciers; WR = water bodies; UR = urban, mining, etc. The areas concerned are classified as “unsuitable” in the final dataset. The following procedure was applied to the st_ar_hwsd grid produced in section 4.6 to obtain the three soil texture layers for the e-atlas (topsoil texture, subsoil texture and simplified topsoil texture. 1. In ArcView, add the st_ar_hwsd grid to the view. Its attribute table contains all the information appearing in the HWSD_database.dbf, with T_USDA_TEX_CLASS, S_USDA_TEX_CLASS and T_texture. 2. Activate this grid, and apply Analysis>Map Calculator option to enter the following formula in the box: ( [st_ar_hwsd.T_USDA_TEX_CLASS]) to create the topsoil texture grid with 13 classes; The result is saved as st_t_ text. 3. Associate to the st_t_ text the name of the USDA texture class by using the following procedure: a. in Microsoft Access, open the soil attribute database HWSD.mdb b. using the Objects>Tables option, open the following tables: • •
D_USDA_TEX_CLASS with fields: Code (top- and subsoil texture code) and Value (name of the USDA top- and subsoil texture). This table contains 13 classes D_Texture with fields: Code (simplified topsoil texture code) and Value (name of the simplified top texture). This table contains three classes
39
c.
Using File>Export, export these Access tables into .dbf format and save as USDA_texture.dbf (containing the code and name of the USDA top- and subsoil texture and simple_texture.dbf (containing the code and name of the simplified top texture)
d. link the st_t_ text to USDA_texture.dbf as follows: • make sure that the Compiled Table Tools extension is uploaded in ArcView • add the st_t_ text grid to the view and open its attribute table • in the ArcView project, select Table and add USDA_texture.dbf • find the column in common between the st_t_ text attribute table (Value) and the USDA_texture.dbf one (Code) • highlight first the header of the Code column in the USDA_texture.dbf and then the header of the Value column in st_t_ text by clicking on them • keeping the attribute table of the st_t_text file activated, join the two tables by clicking the join button on the toolbar • apply the C-Tables Tools>Make Joins Permanent function. This grid contains the 13 top soil texture classes. 4. Repeat steps 2 and 3 for the other textural fields (S_USDA_TEX_CLASS and T_texture) to generate the st_s_text subsoil texture grid with 13 classes and the st_spl_t_txt simplified topsoil texture grid with three classes. The resulting soil texture layers (topsoil, subsoil and simplified) for the European Region are reported in Figure 14. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.14.
40
a)
b)
c) Figure 14. Soil texture ((a) topsoil; (b) subsoil; (c) simplified) for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk 41
4.8 Annual maximum total precipitations over 3 consecutive days The process used to generate the annual maximum total precipitations over 3 consecutive days distribution maps for a five years return periods is very similar to the one used for modelling the distribution of the heat wave as described in the Heat wave hazard modelling that can be found in the Methodology document for the WHO e-atlas of disaster risk, Volume 1. Exposure to natural hazards Version 2.0 This methodology goes through the following steps: 1. Extract the daily precipitations for the 3304 stations located in and around (300 km buffer) the region covered by this version of the e-atlas from the global surface summary of day dataset produced by the US National Climatic Data Center (NCDC) from 1997 to 2008 (http://www7.ncdc.noaa.gov/CDO/cdo, [Accessed December 15, 2010]). 2. Calculate the total daily precipitations for a given period of 3 consecutive days and year of observation (1997-2008) using the EatlasClimMod 1.0 application 3. Apply the Gumbel frequency analysis on the measures from point 2 to obtain the annual maximum precipitations for a five years return period 4. Identify the relevant parameters for each climatic zone and selection of the regression model to spatialize the annual maximum precipitations using a stepwise regression analysis. 5. Perform the spatial interpolation of the annual maximum precipitations for each climatic zone using the selected regression models. 6. Merge the maps for each climatic zone to create the annual maximum total precipitations over 3 consecutive days map for the entire region covered by this version of the e-atlas The only difference, compare to heat wave, is that the annual maximum total precipitations over 3 consecutive days was directly obtained for each climatic station without any need to apply a particular formula. The following sections describe in details the above mentioned steps.
4.8.1 Extraction, preparation and pre-processing of the daily precipitation data The source of the precipitations data is the Global Summary of the Day (GSOD) dataset produced by the National Climatic Data Center (NCDC). Accessible from the internet (http://www7.ncdc.noaa.gov/CDO/cdoselect.cmd?datasetabbv=GSOD&countryabbv=&geore gionabbv= [Accessed December 15, 2010]), this dataset gives access to the following 18 surface meteorological elements for over 9000 stations: 42
• • • • • • • • • • • •
Mean temperature (.1 Fahrenheit) Mean dew point (.1 Fahrenheit) Mean sea level pressure (.1 mb) Mean station pressure (.1 mb) Mean visibility (.1 miles) Mean wind speed (.1 knots) Maximum sustained wind speed (.1 knots) Maximum wind gust (.1 knots) Minimum and Maximum temperature (.1 Fahrenheit) Precipitation amount (.01 inches) Snow depth (.1 inches) Indicator for occurrence of: fog, rain or drizzle, snow or ice pellets, hail, thunder and tornado/Funnel Cloud
Historical data are generally available for 1929 to the present, but the period 1973-present is the most complete. The following sections describe how the precipitations have been extracted from this dataset to cover the 1997 - 2008 period. 4.8.1.1 Extraction and preparation of the daily precipitation data The process used to extract the daily precipitation amount is outlined by the following steps: 1.
Download and save the global surface summary of day data in ASCII format from: http://www7.ncdc.noaa.gov/CDO/cdo web site, after choosing the geographic region (Africa, Asia, Europe, Middle East) and the date range from 01/01/1997 to 31/12/2008.
2. Open each meteorological regional file in Microsoft Office Access and save them as a dBase IV (*.dbf) table using a specific naming convention (e.g. africa.dbf). These files will not open in Microsoft Office Excel because they contain more than 65 356 records. The fields’ names and their description can be found in Annex 6. 3. Only keep the columns that are needed for analysis as follows: a. b.
open africa.dbf, and choose C-Tables Tools>Delete Multi-Fields and select all the fields to be deleted (all except: Field 1: STN, Field 3: YEARMODA and Field 22: PRCP) repeat this step for the other regions.
4. Convert the total precipitations from inches to millimeters using the following steps: a. b. c. d.
open africa.dbf table and add a new field called PRCP_mm by choosing the number type select the header of the PRCP_mm column and click on the Calculate button type the following formula in the Calculator window: “([PRCP].AsNumber)*25.4 save the africa.dbf table 43
e.
repeat steps a to d for the other regions
5. Create a files for each year of observation (1997…2008) from africa.dbf as follows (these files are needed for the pre-processing presented in section 4.8.1.2): a. open africa.dbf and add a field called “date_string” in the attribute table by choosing the string type, in order to convert the date into string type b. select the header of the “date_string” column and click on the Calculate button c. type the following formula in the Calculator window: [YEARMODA].AsString d. add another new column on the right and call it “year” e. select the header of the “year” column and click on the Calculate button f. type the following formula in the calculator window: [date_str].Left(4). This field contains only the year of observation for each station g. select the year = 1997 and save the table under year.txt (year = 1997, 1998, … , 2008) by selecting File>Export and Export Format = Delimited Text h. using WordPad, delete the first record containing the fields names in order to have files with only numerical contents. This operation facilitates the processing in Matlab i. repeat steps a to h for the other years of observation and the other regions j. Put all the files created year.txt (year = 1997, 1998, … , 2008) into a specific folder for example named “Africa/data”, “Europe/data”, etc. 4.8.1.2 Pre-processing of the daily precipitation data The pre-processing of the daily precipitation data takes place using the EatlasClimMod 1.0 application under Matlab 6.0 (or higher). The following steps have to be followed in order to start the EatlasClimMod 1.0 application: 1. Run the MATLAB software. The window presented in Figure 15 will appear.
44
Figure 15. Matlab interface 2. Specify the path to the EatlasClimMod 1.0 application as the current directory using the browsing button
on the upper right side of the window.
3. Launch the EatlasClimMod 1.0 application, by going to the File>Open, select the GUI_HWI_Gumbel.m file as shown in Figure 16 and click on Open.
Figure 16. Open file window in MATLAB to run the EatlasClimMod 1.0 application The EatlasClimMod main program will then be opened in Matlab as shown in Figure 17.
45
Figure 17. Window appearing in Matlab once the EatlasClimMod file has been opened 4. From there, press F5 or click on Run button to run the EatlasClimMod 1.0 application. The start up screen of this application will then appear as shown in Figure 18.
Figure 18. EatlasClimMod 1.0 application startup screen This start up screen gives access to two menus: Operation and About. The About menu give the user has access to the Help file or to the summary screen window (Figure 19) 46
Figure 19. EatlasClimMod 1.0 © summary screen The Operation menu gives access to five options: ¾ Preprocessing: used for data pre-processing; ¾ Heat index: used for the calculation of the daily heat index; ¾ Wave modelling: used to calculate the annual maximum wave for any variable (total precipitations over 3 consecutive days here) over a given number of consecutive days ¾ Unique stations files: used to save the data of each weather station in a separated file; ¾ Gumbel analysis: used to predict the precipitations for different return periods; this options contains two sub-options: • •
All stations: used to apply the Gumbel method on all weather stations One station: used to apply the Gumbel method on a single weather station
¾ Exit: used to close the application. From there, the pre-processing step consists in removing the records with no data for precipitations (lines with a value of 9999.9), then sorting the records by station and by date. This is done using the following steps in the EatlasClimMod 1.0 application: 1. Click on Operation>Preprocessing. This will display the next screen of the wizard as shown in Figure 20.
47
Figure 20. Window used to specify the parameters for the pre-processing of the climatic data The user can then pre-process the data created in section 4.8.1.1 for: a. a single year by checking the One year option, b. a period of consecutive years by checking the A period option. It is important to remember that this stage that the data files are named by date year.txt (1997.txt, 1998.txt, etc.) and stored in the “data” folder (see section 4.8.1.1 in this document). This window is also used to specify the path to the folder in which the generated under section 4.8.1.1 are located as well as the folder in which the preprocessed data should be saved. This is done by clicking on the browse button respective path.
next to the
2. Clik Ok once the information entered in the window reported on Figure 25 has been completed. This step can take time depending on the processor characteristics, the size of the data files and the complexity of the sorting operation. A status screen is therefore displayed and indicates which year is being currently processed (Figure 21).
Figure 21. EatlasClimMod 1.0 © status screen for the pre-processing operation
48
At the end, the program will save the new files under the name sort_year.txt (sort_1997.txt, sort_1998.txt, etc.) in the folder selected by the user under step 1. An example of the content of such files is presented on Figure 22.
Figure 22. Example of file resulting from the pre-processing of the precipitations, with: column 1: weather station number, column 2: Date of the measure, column 3: daily total precipitations (mm)
4.8.2 Calculation of the total precipitations for a given period of consecutive days and annual maximum total precipitation over 3 consecutive days for each weather station and year of observation EatlasClimMod 1.0 has been programmed in such a way that it can directly calculate the total precipitations for a given period of consecutive days and the annual maximum precipitations for each period of observation and weather station. Furthermore, instead of using the sum operation as is the case for precipitations it enables us to use the mean operation for heat wave. This particular component of the EatlasClimMod 1.0 application is based on the following three functions: 1. fct_MainProgram() for extracting the block of data records of each station and calculating the data frequency. This function uses the two following functions . 2. WaveModelling () for the calculation of the total precipitations by translation of a window of size equal to the given period of consecutive days. This calculation is done only for data in successive dates. This function has been also used to select the maximum precipitations for each weather station.
49
3. check_succession() for checking the condition of succession of the dates in the window. To use these functions: 1. In EatlasClimMod 1.0 click on Operation>Wave modelling. This will display the next wizard window as shown in Figure 23.
Figure 23. Window used to specify the parameters for calculating the annual maximum total precipitations over 3 consecutive days In this window, the user: - can decide to calculate the annual maximum total precipitations over 3 consecutive days for a single year by checking the One year radio button, or over a period of consecutive years by checking the A period one. - specifies the path to the files resulting from the pre-processing operation (see section 4.8.1.2) and the path to the folder where the resulting files will be saved (naming this folder “WaveModelling” is recommended, this is -
-
(Figure 23). done by clicking on the browse button specify the number of consecutive days to be considered for measuring the total precipitations. This information is to be entered in the Wave window size field. In the context of the e-atlas, a period of 3 consecutive days has been used. Check the Sum option to calculate the total precipitations for each period of 3 consecutive days and then to calculate the annual maximum total precipitations over 3 consecutive days
2. After completing all the fields in the window, click on the OK button to run the wave modelling process. This step takes few minutes. A status screen is therefore displayed and indicates which year is being currently processed (Figure 24).
50
Figure 24. EatlasClimMod© status screen for the wave modelling process At the end of the treatment, EatlasClimMod 1.0 will have produced new files “final_year.txt” (example final_1997.txt) containing for each station the annual maximum precipitations wave and the annual frequency computing for each year of observation using the following formula: =n/365, where n is the total number of days of observations per year. Figure 25 present one example of such file.
Figure 25. Example of file resulting from the wave modelling operation in EatlasClimMod 1.0 (final_1998.txt) (with column 1: Station, column 2: annual maximum total precipitations over 3 consecutive days, column 3: annual frequency)
51
4.8.3 Calculation of the annual maximum total precipitations over 3 days for a two, five, height and ten years return period 4.8.3.1 Creation of weather station specific files Applying the Gumbel frequency method requires the creation of weather stations specific files containing the total annual maximum precipitations obtained through the process presented in section 4.8.2. This operation is carried out in the EatlasClimMod 1.0 application using the following steps 1. Click on Operation>Unique stations files. This leads to the window presented on Figure 26.
Figure 26. Window used to specify the parameters for creating the weather station specific files In this window, the user must specify: a. the period of consecutive years for which he wants the data to be extracted by weather station, b. the path to the folder containing the files resulting from the calculation of the annual maximum total precipitations over 3 consecutive days (see section 4.8.2) c. the path to the folder in which he wants the results to be saved For b. and c. the path can be changed by clicking on the browse button 2. Once the parameters entered in this window, click on Ok to start the process. This will automatically run the two functions behind this process, namely: save_list_STNs() and save_one_STN().
52
The resulting files will be named as follow: - final_STN_NO_REDON.txt for the file containing all the source data sorted by weather station and year and with all the redundancies deleted - station_number.txt (for example, 170220.txt) for the weather station specific files (see Figure 27 for an example of such file).
Figure 27. Example of weather station specific file with column 1: entire studied year , column 2: total annual maximum precipitations over 3 consecutive days, column 3: annual frequency) 4.8.3.2 Application of the Gumbel frequency analysis The Gumbel frequency analysis technique has been programmed and included in the EatlasClimMod 1.0 application in order to calculate the annual maximum total precipitations over 3 consecutive days for any given climatic station and return period. The use of this application requires the introduction of two thresholds which are used as filters to remove any weather station from the calculation in case these are not respected. While the user can specify these thresholds manually in EatlasClimMod 1.0, they have been fixed as follow in the context of the WHO e-atlas. Namely, a weather station would not be taken into account if: - the dataset for that given station does not contain a daily observation for at least 70% of the days in the year (255 days), - the number of year of observation for that station, after applying the first filter, is lower than 8 years. The threshold at eight years will give a good prediction of the annual maximum total precipitations over 3 consecutive days for return periods that do not exceed eight years. In this work, we have nevertheless also calculated the annual maximum precipitations for a 10 year return period even if this result should be taken with precaution. 53
In EatlasClimMod 1.0, the Gumbel frequency analysis is applied in two steps: -
Application of the Gumbel frequency analysis on all the stations Correction and/or adjustment of the original dataset for unusual observations (typing mistakes, outliers,…).
The steps to be followed for these two steps are described in the coming sections. 4.8.3.2.1 Application of the Gumbel frequency analysis on all the weather stations
Here are the steps to be followed in order to apply the Gumbel frequency analysis on all the weather stations at the same time: 1. In EatlasClimMod 1.0, click on Operation>Gumbel analysis; 2. Choose the All stations option. This will open the window presented in Figure 28.
Figure 28. Interface window of the application of the Gumbel method to all stations In this dialogue box, the user must specify: a. the path to the folder containing the weather station specific files (see section 4.8.3.1), b. the path to the folder where the resulting files will be saved 54
c. the two thresholds described in section 4.8.3.2 (annual frequency and minimum number of years of observations). For a. and b. the path to the selected folder can be changed by clicking on the browse button 3. Once all the parameters entered, click on the button to start the process, This will automatically run the Apply_Gumbel_allSTN() function which itself calls the following four functions: • fix_threshold(): used to remove the lines in the dataset for which the annual frequency is lower than the fixed threshold; • Gumbel(): produces the tables containing the values of the parameters involved in the Gumbel frequency method (Gumbel reduced variable, empirical frequency, mean and standard deviation) • fct_rank(): used to rank to the annual maximum total precipitations over 3 consecutive days (the rank will be the same in the case of similar values for different years) • index_return(): computes the annual maximum total precipitations over 3 consecutive days for two, five, eight and ten year return periods. In the window, the graph for each station is appearing one after the other once the analysis completed. The passage from a station graph to another one is done automatically. Figure 29 show one example of such window.
Figure 29. Example of window appearing as the gumbel analysis is completed on each weather station
55
At the end of the treatment, the EatlasClimMod 1.0 application will have produced a graph, plotting the annual maximum total precipitations over 3 consecutive days versus the Gumbel reduced variate, for each of the station and stored this graph as an image file, named numSTN.jpg (with numSTN = the station number, for example 85940.jpg, 600600.jpg, 601410.jpg), in the folder selected previously. Figure 30 present one example of such graph.
Figure 30. The annual maximum total precipitations over 3 consecutive days versus the Gumbel reduced variate distribution for weather station 172370 with the annual frequency threshold 70% and eight year return period In addition to these graphs, the application also generates a summary file WaveModelledVariable_allSTN_return_2-5-8-10_Fq-0.7_NbrY-8.txt for each study zone and places it in the same folder. This file contains, for each station, the annual maximum total precipitations over 3 consecutive days for two, five, eight and ten year return periods as well as the correlation value between the annual maximum precipitations and a gumbel reduced variable (see figure 30). Figure 31 presents an example of such a file.
56
Figure 31. Example of summary file resulting from the application of Gumbel analysis on all the stations in a given region On the basis of the graph and summary file, the user is then identifying any potential errors in the original datasets, potential outliers (points far from the regression line on the graph) and/or correlation value lower than 0.80 for example. In these cases, the user writes down the number of the concerned stations and follows the steps reported in the following section. 4.8.3.2.2 Correction and/or adjustment of the original dataset for unusual observations
The application of the Gumbel frequency analysis on all the weather stations might reveal some data entry mistakes and/or outliers (see section 4.8.3.2.A). When identified, such cases needs to be corrected in the original dataset in order to improve the correlation in the analysis and therefore reduce the error on the final values for the annual maximum total precipitations over 3 consecutive days for these stations. The following case illustrates how to modify an error in the original dataset and run again the Gumbel analysis on that particular weather station. The graph generated by the analysis for the weather station n°385650 (Figure 32) shows an isolated point (red circle on the graph). Let’s for example consider that this particular measure correspond to a year (1997) where the measurement instruments at the weather 57
station has been changed and that this resulted in wrong measurement of some of the climatic variable
Figure 32. Graph resulting from the Gumbel frequency analysis for weather station 385650 with the isolated point indicated by the red circle It would therefore be appropriate to remove this point from the analysis. To do so: 1. Make a copy of the file for that station (385650.txt) located in the “Stations files” folder 2. Using WordPad, open the version of the file (385650.txt) located in the “Stations files” folder (Figure 33)
Figure 33. Weather station n° 385650 specific file in which the year 1997 record is highlighted in blue 58
3. delete the record (the all line) corresponding to year 1997 (249.94 mm) 4. Save the file in the same location (do not move it); it is always preferable to make a copy before modifying a file. Once the record deleted, it is possible to run again the Gumbel frequency analysis but on that weather station only. For that: 1. In EatlasClimMod 1.0, go to Operation>Gumbel analysis>One station (Figure 34).
Figure 34. Interface window of the application of the Gumbel method to one station In this window, the user must specify: a. the path to the folder containing the weather station specific files (see section 4.8.3.1), b. the number of the station to be corrected (385650) in this case c. the two thresholds described in section 4.8.3.2 (annual frequency and minimum number of years of observations). . 2. After completing all the boxes, click on the button 3. The new graph is then displayed. if the result is satisfactory, save the graph and the associated text file by selecting the path to the resulting files with the button , then by validating the change by clicking on the button
59
(Figure 35),
Figure 35. Interface result of Gumbel application after correction This operation will have for result to modify the previous version of the summary file (Figure 31), file which is necessary for the rest of the process.
4.8.4 Identification of the independent variables and selection of the regression model Spatial interpolation is widely used for translating irregular scattered meteorological data (data collected at discrete locations [i.e. at points]) into continuous data surfaces (rasters). The choice of interpolation method is especially important in the WHO Regions where meteorological data are sparse, and there are large value changes over short spatial distances. Additionally, the spatial density, distribution and spatial variability of sampling stations influence the choice of interpolation technique (MacEachren and Davidson, 1987). Given a set of meteorological data, researchers are confronted with a variety of stochastic and deterministic spatial interpolation methods to estimate meteorological data values at unsampled locations: •
deterministic estimation methods including inverse distance weighting (Legates and Willmont, 1990; Eischeid et al., 1995; Lennon and Turner, 1995; Willmott and Matsuura, 1995; Collins and Bolstad, 1996; Ashraf et al., 1997; Dodson and Marks, 60
•
1997) and spline methods (Eckstein, 1989; Hutchinson and Gessler, 1994; Hulme et al., 1995; Lennon and Turner, 1995; Collins and Bolstad, 1996) stochastic techniques including the kriging and cokriging techniques (Matheron, 1963; Hudson and Wackernagel, 1994; Collins and Bolstad, 1996; Hammond and Yarie, 1996; Holdaway, 1996; Ashraf et al., 1997, El Morjani, 2003) and polynomial regression (Myers, 1990; Collins and Bolstad, 1996; Benzi et al., 1997; Chessa and Delitala, 1997; Hargy, 1997; Vogt et al., 1997; Agnew and Palutikof, 2000; El Morjani, 2003; Li et al., 2006).
For a summary description of these methods, refer to Collins and Bolstad (1996) and El Morjani (2003). The characteristics of the data found for the region covered in this version of the e-atlas (low spatial data density, a high spatial variability and the absence of meteorological data for many countries) resulted in implausible outputs when applying the inverse distance weighted and kriging interpolation methods, more specifically as follows. The application of the inverse distance weighting method over a test area (Islamic Republic of Iran and Pakistan) generated specking or “birds eye” effects around the station locations, which was not plausible as the spatial variation for climatic variables was not following a regular trend. The application of the kriging technique over the same test area produced results that were inconsistent with the original data. Whatever the model used (spherical, exponential or Gaussian) the statistical cross-validation was not able to fit the theoretical spatial semivariogram. This might be because the density of weather stations is too low and the study area too large to support the use of the kriging interpolation method. It has therefore been necessary to find another model that produces results of good quality. A set of variables that are significantly correlated to the precipitations have been taken in consideration, namely: • • • • • •
Elevation (Z), Mean elevation within a 3×3 pixel window of each cell (Z9), Aspect (Asp) as a measure of the local climate effect (microclimate) that can be generated by the orientation of the slope, Slope (Slp). Distance from the relative longitude (d_X) and latitude (d_Y), Distance to the nearest coastline (d_Coast) to account for maritime influences on the precipitations.
With the variables identified, to which their squares were also added, stepwise (back and forth steps) linear regression was used to identify their statistical significance, if any, and their relative contribution to the determination of the dependent variable (annual maximum total precipitations over 3 consecutive days), thereby eliminating any insignificant variables. Because of their respective climatologic characteristics, this process has been applied separately on the following five zones: 61
• • • • •
Zone 1: African continent Zone 2: Middle East countries (Bahrain, Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, United Arab Emirates, West Bank and Gaza Strip, Yemen) Zone 3: Afghanistan, Islamic Republic of Iran, Pakistan Zone 4: Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Mongolia, Tajikistan, Turkmenistan and Uzbekistan Zone 5: Albania, Austria, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Greece, Hungary, Latvia, Lithuania, Moldova, Republic of, Montenegro, Poland, Romania, Serbia, Slovakia, Slovenia, former Yugoslav Republic of Macedonia, Turkey and Ukraine.
The final map was created by aggregating, in a seamless way, the results obtained for each zone. A stepwise linear regression analysis was performed separately for each of these zones using S-Plus software. The validation of each regression was carried out using R2 variance analysis as well as a detailed probability and residual analysis in order to identify the significant variables and therefore select the best regression model possible. This process has been implemented in three steps, which are explained in the following sections: • • •
preparation of the GIS layers containing the spatial distribution of the causal factors and dependant variable (annual maximum total precipitations over 3 consecutive days for a five year return periods), preparation of the stepwise regression analysis, application of the stepwise regression analysis.
Due to the climatological characteristics of the study area, the stepwise regression analysis has been applied separately to the five zones described here above.
4.8.4.1 Preparation of the GIS layers containing the spatial distribution of the causal factors and dependant variable As reported in above the causal factors retained to explain the spatial variability of the annual maximum total precipitations over 3 consecutive days are: 1. 2. 3. 4. 5.
Elevation (Z) Mean elevation within a 3×3 pixel window around each cell (Z9) Aspect (Asp). Distance from the relative longitude (d_X) and latitude (d_Y) Distance to the nearest coastline (d_Coast)
The spatial distribution of the elevation can be directly and easily derived from the dataset generated in the section 4.1. When it comes to the last 4 causal factors, specific process had to be applied in order to get the appropriate GIS layer for the analysis. 62
Before applying the steps reported in the next sections, each of the layers used in this process first had to be projected into a metric projection system. This has been done using the process presented in Annex 7. In addition to that, and as mentioned in section 4.8.4, the all region covered in this version of the e-atlas had to be cut into 5 zones during the analysis to take their respective climatologic characteristics into account. Each of the layers used in the regression analysis therefore had to be cut according to the extent of the respective zone to which a 300 km buffer has been added in order to ensure a good interpolation at the edge of each of the zone. This process is described in Annex 8. 4.8.4.1.1. Preparation of the distance to the nearest coastline layers
First, the coastlines have been extracted from the projected version of the international boundaries layer (st_ar_int_bord_km.shp) using the following steps: 1. Make sure that the XTools Extension is uploaded in ArcView. 2. Display in the view both the international boundaries (st_ar_int_bord_km.shp) and the layer containing the global coastline border coming from the SALB project Un_coast_01.shp reprojected into metric projection system (Annex 7) (Un_coast_01_km.shp). 3. Use the XTools> Clip with polygon(s) function. 4. Select the Un_coast_01_km.shp as the theme that contains features that you wish to clip. 5. Select st_ar_int_bord_km.shp as the polygon theme that contains the polygons that will be used as the reference for the clipping. 6. Specify the name for the new shapefile to be created as st_ar_coast_km. The resulting projected coastline layer has then cut according to the 5 climatic zones (see Annex 8) to give the following layers: zone1_coast_buffer_300.shp, zone2_coast_buffer_300.shp, zone3_coast_buffer_300.shp, zone4_coast_buffer_300.shp, zone5_coast_buffer_300.shp. From there, the distance from the coastline was computed as follows starting with the first climatic zone: 1. In ArcView, display the zone1_coast_buffer_300.shp 2. Navigate to Analysis>Find Distance to create a grid of distances from the coasts in kilometres: 3. in the next menu, select Output Grid Specification and specify as follow: 63
a. b. c. d. e.
set Output Grid Extent = Same As zone1_int_bord_buffer_300.shp set Output Grid Cell Size = As Specified Below set Cell Size = 1 km use the default number of rows and columns save the result as zone1_dist_coast.
4. Repeat steps 1 to 3 on the buffered coastline shapefiles for the other zones changing the name of the resulting files accordingly. 4.8.4.1.2. Preparation of the distance from the relative latitude/longitude layer
The following process was followed for generating the climatic zone specific relative latitude layers. 1. Create a line shapefile that will pass by the point located at the extreme South of the of the frist climatic zone as follows: a. b. c. d.
In ArcView, add the zone1_int_bord_buffer_300.shp file in the view (See annex 8) create a new line theme using View>New Theme, manually digitize a straight horizontal line passing by the point located at the extreme South of the of this particular zone. This is going to be the zero degrees relative latitude for this zone, save the editing work as zone1_Y.shp.
2. Navigate to Analysis>Find Distance to create a grid of distances from the line generated in the zone1_Y.shp file. 3. in the next menu, select Output Grid Specification and specify as follow: a. set Output Grid Extent = Same As zone1_int_bord_buffer_300.shp b. set Output Grid Cell Size = As Specified Below c. set Cell Size = 1 km d. use the default number of rows and columns e. save the result as zone1_dist_Y. 4. Repeat steps 1 to 3 on the other zones. The process used to create a layer containing the relative longitude for study area is identical to the process described for the relative latitude drawing this time a vertical line passing by the point located at the extreme West of each climatic zone. This would represent the zero degrees relative longitude. The resulting output is then saved as zonen_dist_X (n corresponds to the number associated with each climatic zone). 4.8.4.1.3. Preparation of the mean elevation distribution layer
The mean elevation distribution grid was derived from the DEM using the following steps: 1. In ArcView, make sure that the Grid Analyst extension is. 64
2. Add the st_ar_dem* grid in the view and activate it. 3. Use the Analysis>Neighborhood Statistics function specifying the following in the windows that appears: a.
statistic = Mean
b.
under neighbourhood, type of neighbourhood for analysis = Rectangle
c.
select the “cell” dial; and set “Width” and “Height” to three cells
d.
save the resulting grid as st_ar_Z9.
The resulting grid has then been projected using the steps reported in Annex 7 before being cut according to the extent of each climatic zones to which a 300 km has been added and this following the process reported in Annex 8. The grids resulting from these operations are named: Zone1_ Z9, Zone2_ Z9, Zone3_ Z9, Zone4_Z9 and Zone5_Z9. 4.8.4.1.4. Preparation of the aspect layers
The aspect, or slope direction, layer has been included in the regression analysis under the form of a dummy variable. This is being done as we don’t want to consider broad directions (N, NE, E, SE,…) and not any small variations in the direction of the slope. In order the use the aspect distribution layer as a dummy variable in the regression analysis, eight grids named Aspect_X (where X is N, NE, E, SE, S, SW, W, and NW) have been derived according to the following classification: • • • • • • • •
0°–22.5° and 337.5°–360°: North 22.5°–67.5°: North East 67.5°–112.5°: East 112.5°–157.5°: South East 157.5°–202.5°: South 202.5°–247.5°: South West 247.5°–292.5°: West 292.5°–337.5°: North West
For example, in Aspect_N cells which slope is directed towards the North are given a value of 1 while any other cells are given the value 0. The procedure used to create these grids is outlined in the following steps. 1. In ArcView, upload the aspect distribution grid st_ar_aspect* into the view. 2. Select the Analysis>Map Calculator function and enter the following formulas in the Calculator window: [st_ar_aspect]>=67.5 and [st_ar_aspect]Appends Tables Together function to create one unique table from the five created previously and save the result as st_ar_PRC_5.dbf. 6. Merge the st_ar__PRC_2_5_8_10.dbf with the st_ar_stations.shp* shapefile as follows: a. b. c.
in ArcView add the st_ar_PRC_5.dbf table in the table window add the st_ar_stations.shp* in the view and open its attribute table Select the header of the STN column in the st_ar_PRC_5.dbf table and the header of the Number column in the attribute table of the st_ar_stations.shp* shapefile e. keeping the attribute table of the st_ar_stations.shp* file active, join the two tables by clicking the join button on the tool bar f. use the C-Tables Tools>Make Joins Permanent function to fix all the columns added in the attribute table of the st_ar_stations.shp* shapefile g. save the resulting shapefile as st_ar_ PRC.shp. 66
The resulting shape file has then been projected using the steps reported in Annex 7 before being cut according to the extent of each climatic zones to which a 300 km has been added and this following the process reported in Annex 8. The files resulting from these operations are named as follow: zone_1_PRC.shp, zone_2_PRC.shp, zone_3_PRC.shp, zone_4_ PRC.shp, zone_5_PRC.shp.
4.8.4.3 Preparation of the stepwise regression analysis Before performing the stepwise regression analysis on each of the five zones it was necessary to prepare a table which contained, for each weather station, the annual maximum total precipitations over 3 consecutive days for five year return period as well as the variables extracted from each grid prepared in section 4.8.4.1. The procedure used to create this table is outlined in the following steps: 1. In ArcView, make sure that the Grid Analyst extension is uploaded, 2. In a view, add the seven causal factor distribution grids for the first climatic zone zone1_dem, zone1_Z9, zone1_slope, zone1_asp_Y, zone1_dist_coast,, zone1_dist_X, zone1_dist_Y and the shapefile containing the distribution of the weather stations to the annual maximum total precipitations over 3 consecutive days for the five year return period have been associated for the first zone (zone1_PRC.shp). 3. Make the zone1_PRC.shp shapefile the active theme and use the Grid Analyst>Extract X, Y and Z Values for Point Theme from Grid functions 4. Select the first grids listed in step 2 from the drop list. The function will add and then populates three new fields in the attribute table of zone1_PRC.shp (Xval, Yval and Zval), the last one storing the value extracted from the grid layer. 5. Open the attribute table of the zone1_PRC.shp shapefile and click on the header of the “Zval” column. 6. Rename this field to correspond to the name of the raster layer (i.e. Z for elevation, Z9 for mean elevation, SLP for slope, ASP_Y for aspect (where Y is N, NE, E, SE, S, SW, W and NW), d_Coast for the distance from the coastline, d_X for the distance from the relative longitude, d_Y for the distance from the relative latitude) using the C-Tables Tools>Rename/Resize/Copy Field(s) function. 7. Repeat steps 3 to 5 on the other causal factor distribution grids until the Zval for each of them is integrated into the attribute table of the zone_1_PRC.shp shapefile. 8. Save the resulting table as zone1_PRC_regression.dbf.
67
9. Repeat steps 2 to 8 on the other zones, changing the names of the resulting files accordingly. 4.8.4. 4 Application of the stepwise regression analysis Once the stepwise regression table ready (see section 4.8.4.3), it is possible to perform the stepwise regression analysis on each zone. This procedure is done using S-Plus software as follows. 1. Launch the S-Plus software. 2. Choose File>Import Data>From File function to import the stepwise regression table created above for the first climatic zone (zone1_PRC_regression.dbf). 3. Choose Statistics>Regression>Stepwise function. 4. In the “Stepwise Linear Regression” dialogue box that appears: a. b.
under Data Set scroll down the list and click on zone1_PRC_regression click on the Create Formula box for the Upper Model and use PRC_5 as the response and add all explanatory variables (Z, Z9, SLP, ASP_X, d_Coast, d_X, d_Y) as Main Effects and Quadratic with: PRC_5 = annual maximum total precipitations over 3 consecutive days for a five year return period Z = elevation Z9 = mean elevation SLP = slope ASP_X = aspect (where X is N, NE, E, SE, S, SW, W and NW) d_Coast = distance from coastline d_X = distance from the relative longitude d_Y = distance from the relative latitude
c.
click OK to run the procedure.
A report is created that shows the result of this selection procedure with the coefficient of the variables selected and their significance, residual standard error, multiple R2 and probability (F statistic). Table 9 shows the report obtained for a five year return period in zone 1.
68
Table 9. Results of the regression analysis for the annual maximum total precipitations over 3 consecutive days for zone 1 and for a five year return period Variable (Intercept) Z9 d_coast d_Y2 d_coast2 d_Y3 d_X3 Residual standard error Degrees of freedom Multiple R2 F statistic Probability (F statistic)
Regression coefficient 164.1456380056 -0.0063207498 -0.0601654632 -0.0000263097 0.0000358803 0.0000000053 0.0000000003
Standard error 6.52463 0.00425 0.02310 0.00000 0.00002 0.00000 0.00000
t value 25.15781 -1.48520 -2.60360 -9.41860 1.66126 7.18709 5.16708
Probability Pr(>|t|) 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
3.59449 446 0.78 51.92338249 0.0000
The regression equation explaining the annual maximum total precipitations over 3 consecutive days for this particular return period (five years) and zone (zone 1) can be read as follows: PRC_5 = -0.0063207498* Z9 – 0.0601654632*d_coast – 0.0000263097*d_Y2 + 0.0000358803* d_coast 2 + 0.0000000053 *d_Y3 + 0.0000000003 *d_X3+164.1456380056 5. Repeat steps 2 to 4 for the other zones changing the names of the files accordingly. Annex 9 present the regressions obtained for all climatic zones.
4.8.5 Spatialization of the predicted annual maximum total precipitations over 3 consecutive days The annual maximum total precipitations over 3 consecutive days distribution map for a five year return period and zone is created by applying the regressions found in section 4.8.4.4 on the corresponding grids as follows (example for the first climatic zone): 1. Make sure that all the seven causal factor distribution layers for the first zone are uploaded in the view. 2. Select the Analysis>Map Calculator function and enter the following formula in the calculator window: ([zone_1_ Z9]* (-0.0063207498))–([zone_1_dist_coast]* 0.0601654632)– ([zone_1_dist_Y]*[zone_1_dist_Y]* 0.0000263097)+( [zone_1_dist_coast]* [zone_1_dist_coast]* 0.0000358803)+( [zone_1_dist_Y]*[zone_1_dist_Y]* [zone_1_dist_Y]* 0.0000000053) +([zone_1_dist_X]*[zone_1_dist_X]* [zone_1_dist_X]* 0.0000000003) +164.1456380056 3. Save the output grid as zone1_PRC_5. This corresponds to the spatial distribution of the annual maximum precipitations over the first zone for a five year return period. 69
4. Unproject the zone1_PRC_5 layer from the Equal-Area Cylindrical projection to the Geographic one using the following steps: a. click either the function
button or use the Grid Projector>Grid and Theme Projector
b. select zone1_PRC_5 from the list as the grid to project c. In the Grid Projector window: •
Specify the parameters for the current projection as follows: - Category = projection of the world - Type = Equal-Area Cylindrical projection - Current Projection Units = kilometres
• specify the parameters for the new projection as follows: - Category = projection of the world - Type = Geographic - New Projection Units = decimal degrees d. In the next window, specify the new cell size = 0.008333. e. Save the output grids as zone1_PRC_5_d. 5.
Repeat steps 1 to 4 on the other zones using the corresponding regressions.
The annual maximum total precipitations over 3 consecutive days distribution maps for each climatic zone were then merged to generate three grids, covering the study area using the following steps: 1. In ArcView, make sure that the Grid Transformation Tool extension is uploaded. 2. Select the Transform Grid>Mosaic function to create one unique grid from zone1_PRC_5_d, zone2_PRC_5_d, zone3_PRC_5_d, zone4_PRC_2_d and zone5_PRC_2_d. 3. Save the result as st_ar_PRC_5_d. Finally the following process was applied to clip the annual maximum total precipitations over 3 consecutive days distribution mosaic map to the borders of the region covered in this version of the e-atlas. 1. In ArcViewmake sure that the Grid Analyst extension is uploaded, 2. upload the AFRO international boundary level used for the e-atlas (afro_int_bnd.shp)
70
3. Activate st_ar_PRC_5_d and select the Grid Analyst>Extract Grid Theme Using Polygon function •
Click “yes” to continue.
•
Select the afro_int_bnd.shp from the dropdown list to be used as the layer on which the grid needs to be clipped and click OK
•
Make the output grid the active theme and choose the Theme>Convert to Grid function to create the AFRO annual maximum precipitations distribution map for a five year return period
•
Save it as afro_PRC_5.
4. Repeat steps 2 and 3 for the others WHO Regions (EURO and EMRO) and save the outputs as emro_PRC_5, and euro_PRC_5. The resulting annual maximum total precipitations over a 3 days period distribution layer for the European Region are reported in Figure 36. Please refer to the e-atlas DVD itself for the maps covering the other two WHO Regions. The associated metadata for the complete dataset can be found in Annex 10.15.
Figure 36. Annual maximum total precipitations over a 3 days period distribution for the countries of the European Region covered in this version of the WHO e-atlas of disaster risk This layer is freely redistributable; it can be found in the data section of the first volume of the e-atlas. Please acknowledge the e-atlas as noted in the metadata if you want to use it.
71
References and further reading Agnew MD, Palutikof JP, 2000. GIS-based construction of baseline climatologies for the Mediterranean using terrain variables. Climate research, 14:115–27. Ashraf M, Loftis JC and Hubbard KG, 1997. Application of geostatistics to evaluate partial weather station networks. Agricultural and forest meteorology, 84(3–4):255–71. Benzi R, Deidda R, Marrocu M, 1997. Characterization of temperature and precipitation fields over Sardinia with principal component analysis and singular spectrum analysis. International journal of climatology, 17(11):1231–62. Chessa PA, Delitala AM, 1997. Objective analysis of daily extreme temperatures of Sardinia (Italy) using distance from the sea as independent variable. International journal of climatology, 17(13):1467–85. Collins FC Jr, Bolstad PV, 1996. A comparison of spatial interpolation techniques in temperature estimation, Proceedings of the Third International Conference/Workshop on Integrating GIS and Environmental Modeling. January 21–25, 1996, Santa Fe, New Mexico, USA. Dodson R, Marks D, 1997. Daily air temperature interpolated at high spatial resolution over a large mountainous region. Climate research, 8(1):1–20. Eckstein BA, 1989. Evaluation of spline and weighted average interpolation algorithms. Computers & geosciences, 15(1):79–94. Eischeid JK, Baker FB, Karl TR, Diaz HF, 1995. The quality control of long-term climatological data using objective data analysis. Journal of applied meteorology, 34(12):2787–95. El Morjani Z, 2003. Conception d’un système d’information à référence spatiale pour la gestion environnementale; application à la sélection de sites potentiels de stockage de déchets ménagers et industriels en région semi-aride (Souss, Maroc). Doctoral thesis, University of Geneva. Terre et environnement, 42. Hammond T, Yarie J, 1996. Spatial prediction of climatic state factor regions in Alaska. Ecoscience, 3(4):490–501. Hargy VT, 1997. Objectively mapping accumulated temperature for Ireland. International journal of climatology, 17(9):909–27. Holdaway MR, 1996. Spatial modelling and interpolation of monthly temperature using kriging. Climate research, 6:215–25. Hudson G, Wackernagel H, 1994. Mapping temperature using kriging with external drift: theory and an example from Scotland. International journal of climatology, 14:77–91.
72
Hulme M, Conway D, Jones PD, Jiang T, Barrow EM, Turney C, 1995. Construction of a 1961–1990 European climatology for climate change modelling and impact applications. International journal of climatology, 15:1333–63. Hutchinson MF, Gessler PE 1994. Splines—more than just a smooth interpolator, Geoderma, 62:45–67. Legates DR, Willmott CJ, 1990. Mean seasonal and spatial variability in global surface air temperature. Theoretical and applied climatology, 41:11–21. Lennon JJ, Turner JRG, 1995. Predicting the spatial distribution of climate: temperature in Great Britain. Journal of animal ecology, 64:370–92. Li J, Huang JF, Wang XZ, 2006. A GIS-based approach for estimating spatial distribution of seasonal temperature in Zhejiang province, China. Journal of Zhejiang University science A, 7(4):647–56. MacEachren AM, Davidson JV, 1987. Sampling and isometric mapping of continuous geographic surfaces. American cartographer, 14(4):299–320. Matheron G, 1963. Principles of geostatistics. Economic geology, 58:1246–66. Myers RH, 1990. Classical and modern regression with applications. Boston, PWS-Kent Publishing. Vogt JV, Viau AA, Paquet F, 1997. Mapping regional air temperature fields using satellitederived surface skin temperatures. International journal of climatology, 17(14):1559– 79. Willmott CJ, Matsuura K, 1995. Smart interpolation of annually averaged air temperature in the United States. Journal of applied meteorology, 34(12):2577–86.
73
Annex 1. Classes observed for the geology of the region covered by this version of the e-atlas acid
Acid igneous unknown age and type
ae
Acid volcanics unknown age
ai
Acid intrusives unknown age
basic
KJ
Cretaceous–Jurassic
KJs
Undifferentiated Jurassic and Cretaceous sedimentary rocks
Basic igneous unknown age and type
Kl
Lower Cretaceous
be
Basic volcanics unknown age
Ks
Cretaceous sedimentary rocks
bi
Basic intrusives unknown age
KTrs
C
Carboniferous
Middle Triassic–Lower Cretaceous sedimentary rocks
CD
Carboniferous–Devonian
Kv
Cretaceous volcanics
Cm
Cambrian
Mi
Mesozoic igneous
MiPi
Mz–Pz igneous
Mz
Mesozoic
MzCzi
Mesozoic–Cenozoic intrusives
MzCzv
Mesozoic–Cenozoic volcanics
Mzi
Mesozoic intrusives
Mzim
Mesozoic intrusive and metamorphic rocks
Mzo
Mesozoic ophiolites
MzPz
Mesozoic–Paleozoic
MzPzi
Mesozoic–Paleozoic intrusives
Mzv
Mesozoic volcanics
N
Neogene
O
Ordovician
OCm
Ordovician–Cambrian
Osm
Ordovician metamorphic and sedimentary rocks
pC
Precambrian undifferentiated
PC
Permian–Carboniferous
pCm
Precambrian
pCmi
Precambrian intrusives
pCmv
Precambrian volcanics
Pg
Paleogene sedimentary rocks
Pi
Paleozoic igneous
Pr
Permian rocks
CmpCm Cambrian–Precambrian Cmsm
Cambrian sedimentary and metamorphic rocks
Cs
Carboniferious sedimentary rocks
Czi
Cenozoic intrusives
CzMzi
Cenozoic–Mesozoic intrusives
CzMzv
Cenozoic–Mesozoic volcanics
Czv
Cenozoic volcanics
D
Devonian
DS
Devonian–Silurian
DSO
Devonian–Silurian–Ordovician
Du
Upper/Middle Devonian
H2O
Open water
ie
Intermediate volcanics unknown age
ii
Intermediate instrusives unknown age
inter
Intermediate igneous unknown age and type
Io
Ophiolites (undifferentiated)
J
Jurassic
Jl
Lower Jurassic
Jms
Jurassic metamorphic and sedimentary rocks
JTr
Jurassic–Triassic
K
Cretaceous
KC
Cretaceous–Carboniferous
74
Pz
Paleozoic
SOc
Ordovician and Silurian carbonates
Pzl
Lower Paleozoic rocks
T
Tertiary
PzpC
Paleozoic–Precambrian
Ti
Tertiary intrusives
Pzu
Upper Paleozoic metamorphic and intrusive rocks
TK
Tertiary–Cretaceous
Pzv
Paleozoic Volcanics
TKi
Tertiary–Cretaceous intrusives
Q
Quaternary
TKim
Qe
Quaternary (eolian)
Cretaceous and Tertiary igneous and metamorphic rocks
Qf
Quaternary (fluvial)
TKs
Cretaceous and Tertiary sedimentary rocks
QN
Quaternary–Neogene
TKv
Tertiary–Cretaceous volcanics
Qp
Pleistocene
To
Tertiary
Qs
Quaternary sand and dunes
Tr
Triassic
Qsk
Quaternary (sabkha)
Trms
Triassic metamorphic and sedimentary rocks
QT
Quaternary–Tertiary
TrP
Triassic–Permian
QTv
Quaternary–Tertiary volcanics
TrPr
Permian and Triassic rocks
Qv
Quaternary volcanics
Ts
Tertiary sedimentary rocks
S
Silurian
Tv
Tertiary volcanics
Salt
Salt undifferentiated
vs
Volcanics and sedimentary deposits undifferentiated
SO
Silurian–Ordovician
75
Annex 2. Procedure to clip a shapefile dataset to the borders of the region covered by this version of the e-atlas using the international boundary template (extracted from the SALB editing protocol) The following steps were used to clip shapefile datasets to the borders of the region covered by this version of the e-atlas. 1.
Set the editing parameters as follows: a. click on View>Properties>Distance Unit and set the distance units to metres b. select the Theme>Properties>Editing function c. select the box called “General” in the snapping window and in the view, click and hold the right button of the mouse and select “enable general snapping”. A new icon will then appear in the menu d. click on this new icon e. draw a circle representing the snapping distance in the view. The snapping tool allows vertices to snap to each other at the defined tolerance distance, thus enabling a perfect connection between segments.
2. Prepare the shapefile for clipping: a. click on Theme>Start Editing to enter the editing mode b. make sure that all the vertices forming the extent of the shapefile to be clipped are located outside of the study area international borders. If this is not the case, move them, paying particular attention to vertices that form intersections between two polygons in the shapefile to be clipped c. if there are a large number of external vertices use the steps below to eliminate some of them before moving the vertices - zoom in on the polygon to be edited and select it with the transparent arrow
d.
- scroll down ArcView’s drawing tools (represented by the icon ) and select the Draw Line to Split Polygon tool , and do the following using the mouse pointer in the view: i. left-click a first time outside the polygon (in the white area). ii. make several other left-clicks inside the polygon in order to create a kind of polygon iii. for the last point of this polygon, double left click in the white area. This will complete your new polygon. iv. erase it by clicking on Delete on your keyboard select the newly cut polygon with the transparent arrow tool and move the few remaining vertices outside the international boundary.
3. Clip the shapefile to the international boundary layer. a. make sure that the XTools extension is active in ArcView b. display both the shapefile to be clipped and the international boundaries of the Region in the active view 76
c. d. e. f.
select the Xtools>Clip with Polygon(s) function. select the shapefile to be clipped as the theme that contains the features that you wish to clip select the international boundary shapefile as the theme on which the theme mentioned in the previous point will be clipped specify the name for the new shapefile to be created and click on OK.
77
Annex 3. Procedure to assign the “no data” value to islands without data coverage within the region covered by this version of the e-atlas The following procedure was used to assign “no data” to islands located within the study area that are not covered by the geology, soil and other datasets. 1.
Make sure that the XTools extension is active in ArcView.
2.
Select the Xtools>Convert Multipart Shapes to Single Part function to group the polygons of the study area international boundary shapefile into a single feature.
3.
Save the result as st_ar_int_border_single.shp.
4.
In the attribute table of st_ar_int_border_single.shp, create a string field called ND and assign the “No data” value to all the polygons.
5.
Select the XTools>Identify function and: a. in the first dialog box, select st_ar_int_border_single.shp as the theme containing the features that are needed b. in the next dialog box, select ND as the field from the Input theme that should be kept in the Output theme. c. in the third dialog box, select the shapefile containing the features for which it is needed to identify the gaps (e.g. soil type layer); and in the next window the field from this layer that containing the attributes that will appear in the Output file. d. in the fifth dialog box, type the name of the output file and click OK.
6. In the attribute table of the output file, select all the polygons presenting an empty cell in the column containing the result of the identification process and assign “no data”. This process does not allow identification of potential gaps existing within individual polygons. It only allows identification of polygons that are completely uncovered by the data layer.
78
Annex 4. List of soil grouping types present in the region covered by this version of the e-atlas Acrisols (AC):
Soils with subsurface accumulation of low activity clays and low base saturation
Alisols (AL):
Soils with sub-surface accumulation of high activity clays, rich in exchangeable aluminium
Andosols (AN):
Young soils formed from volcanic deposits
Anthrosols (AT):
Soils in which human activities have resulted in profound modification of their properties
Arenosols (AR):
Sandy soils featuring very weak or no soil development
Calcisols (CL):
Soils with accumulation of secondary calcium carbonates
Cambisols (CM):
Weakly to moderately developed soils
Chernozems CH):
Soils with a thick, dark topsoil, rich in organic matter with a calcareous subsoil
Ferralsols (FR):
Deep, strongly weathered soils with a chemically poor, but physically stable subsoil
Fluvisols (FL):
Young soils in alluvial deposits
Glaciers (GG):
Glaciers
Gleysols (GL):
Soils with permanent or temporary wetness near the surface
Greyzems (GR):
Acid soils with a thick, dark topsoil rich in organic matter
Gypsisols (GY):
Soils with accumulation of secondary gypsum
Histosols (HS):
Soils which are composed of organic materials
Kastanozems (KS):
Soils with a thick, dark brown topsoil, rich in organic matter and a calcareous or gypsum-rich subsoil
Leptosols (LP):
Very shallow soils over hard rock or in unconsolidated very gravelly material
Lixisols (LX):
Soils with subsurface accumulation of low activity clays and high base saturation
Luvisols (LV):
Soils with subsurface accumulation of high activity clays and high base saturation
Nitisols (NT):
Deep, dark red, brown or yellow clayey soils having a pronounced shiny, nut-shaped structure
Phaeozems (PH):
Soils with a thick, dark topsoil rich in organic matter and evidence of removal of carbonates
Planosols PL):
Soils with a bleached, temporarily water-saturated topsoil on a slowly permeable subsoil
Plinthosols (PT):
Wet soils with an irreversibly hardening mixture of iron, clay and quartz in the subsoil
Podzols (PZ):
Acid soils with a subsurface accumulation of iron-aluminium-organic compounds
Podzoluvisols (PD):
Acid soils with a bleached horizon penetrating into a clay-rich subsurface horizon
Regosols (RG):
Soils with very limited soil development
Rock outcrop (RK):
Rock
Sand dunes (DS):
Sand dunes
Salt flats (ST):
Salt flats
Solonchaks (SC):
Strongly saline soils
Solonetz (SN):
Soils with subsurface clay accumulation, rich in sodium
Urban, mining, etc. (UR): Urban, mining, etc. Vertisols (VR):
Dark-coloured cracking and swelling clays
Water bodies (WR):
Water bodies
79
Annex 5: Procedure for converting a shapefile into a grid This procedure is used for converting a shapefile into a grid presenting a 1 km resolution. 1.
Make sure that the Spatial Analyst extension is active in ArcView.
2.
In the view activate the shapefile that you want to convert.
3.
Select the Theme>Convert to Grid function.
4. In the next window assign a directory and name for the grid that will be generated; click OK. 5. In the next window: a. select “Same as …..shp” as the “Output Grid extent same as” in order to obtain a grid presenting the same extent than the original shapefile b. for the Output Grid cell size choose 0.008333 degrees. This resolution corresponds to 1 km at the equator, the resolution selected for the e-atlas c. do not modify the default number of Rows and Number of columns values and click OK 6. In the next window select the header of column containing the value that will be converted to a grid and click OK. 7. Click “yes” to join the feature attributes to the grid, and also click “yes” to add the theme to the view. 8. Save the output grid as st_raster with raster as the specific raster (e.g. emro_geology).
80
Annex 6 - Description of the NCDC daily meteorological elements dataset This annex provides the indication of the type of the data (int[eger], real or char[acter]) as well as a description of each of the fields of the daily meteorological data coming from the global surface summary of the day data produced by the National Climatic Data Center (NCDC). Field
Type
Description
STN
Int
Station number
WBAN
Int
This is the historical “Weather Bureau Air Force Navy” number where applicable
YEARMODA Int
Year, month and day
TEMP
Real
Mean temperature for the day in degrees fahrenheit to tenths. Missing = 9999.9
Count
Int
Number of observations used in calculating mean temperature
DEWP
Real
Mean dew point for the day in degrees fahrenheit to tenths. Missing = 9999.9
Count
Int
Number of observations used in calculating mean dew point
SLP
Real
Real mean sea level pressure for the day in millibars to tenths. Missing = 9999.9
Count
Int
Number of observations used in calculating mean sea level pressure
STP
Real
Mean station pressure for the day in millibars to tenths. Missing = 9999.9
Count
Int
Number of observations used in calculating mean station pressure
VISIB
Real
Mean visibility for the day in miles to tenths. Missing = 999.9
Count
Int
Number of observations used in calculating mean visibility
WDSP
Real
Mean wind speed for the day in knots to tenths. Missing = 999.9
Count
Int
Number of observations used in calculating mean wind speed
81
Field
Type
Description
MXSPD
Real
Maximum sustained wind speed reported for the day in knots to tenths. Missing = 999.9
GUST
Real
Maximum wind gust reported for the day in knots to tenths. Missing = 999.9
MAX
Real
Maximum temperature reported during the day in degrees fahrenheit to tenths. Missing = 9999.9
Flag
Char
Blank indicates maximum temperature was taken from the explicit maximum temperature report and not from the hourly data. * indicates maximum temperature was derived from the hourly data (i.e. highest hourly or synoptic-reported temperature)
MIN
Real
Minimum temperature reported during the day in degrees fahrenheit to tenths—time of minimum temperature report varies by country and region, so this will sometimes not be the minimum for the calendar day. Missing = 9999.9
Flag
Char
Blank indicates minimum temperature was taken from the explicit minimum temperature report and not from the hourly data. * indicates minimum temperature was derived from the hourly data (i.e. lowest hourly or synoptic-reported temperature)
PRCP
Real
Total precipitation (rain and/or melted snow) reported during the day in inches and hundredths; will usually not end with the midnight observation—i.e. may include latter part of previous day. 0.00 indicates no measurable precipitation (includes a trace). Missing = 99.99
82
Field
Type
Description
Flag
Char
A = one report of 6-hour precipitation amount B = summation of two reports of 6-hour precipitation amount C = summation of three reports of 6-hour precipitation amount D = summation of four reports of 6-hour precipitation amount E = one report of 12-hour precipitation amount F = summation of 2 reports of 12-hour precipitation amount G = one report of 24-hour precipitation amount H = station reported 0 as the amount for the day (e.g. from 6-hour reports), but also reported at least one occurrence of precipitation in hourly observations—this could indicate a trace occurred, but should be considered as incomplete data for the day I = station did not report any precipitation data for the day and did not report any occurrences of precipitation in its hourly observations—it is still possible that precipitation occurred but was not reported
SNDP
Real
Snow depth in inches to tenths—last report for the day if reported more than once. Missing = 999.9 Note: most stations do not report 0 on days with no snow on the ground; therefore, 999.9 will often appear on these days
FRSHTT
Int
Indicators (1 = yes, 0 = no/not reported) for the occurrence during the day of: fog (‘F’—1st digit); rain or drizzle (‘R’—2nd digit). Snow or ice pellets (‘S’—3rd digit). Hail (‘H’—4th digit). Thunder (‘T’—5th digit). Tornado or funnel cloud (‘T’—6th digit).
83
Annex 7. Projection of a GIS layers into the metric projection system This operation was used to switch the map units of the different layers used in the analysis from decimal degrees to kilometres in order to be able to measure distances. The operation was performed separately on each the vector and raster layers. Taking the international boundaries as an example (st_ar_int_bord.shp*), all the vector layers have been projected using the following steps: 1.
In ArcView, make sure that the Grid and Theme Projector v.2 extension are uploaded. button or use the Grid Projector>Grid and Theme Projector function 2. Click either the from the menu. 3. In the window that open, select the st_ar_int_bord.shp* layer from the list as the theme to project. 4. In the next window that opens: a.
Specify the following parameters for the current projection Category = projection of the world Type = Geographic Current Projection Units = decimal degrees
b.
Specify the following parameters for the new projection Category = projection of the world Type = Equal-Area Cylindrical New Projection Units = kilometers.
5. Save the output theme as st_ar_int_bord_km.shp. The raster layers have also been projected. Here is for example the process followed for the Digital Elevation Model (DEM): 1. Click either the button or select Grid Projector>Grid and Theme Projector. 2. Select the st_ar_dem grid from the list in the next window. 3. In the next window that opens: a.
b.
c. d. e.
specify the parameters for the current projection Category = projection of the world Type = Geographic Current Projection Units = decimal degrees specify the parameters for the new projection Category = projection of the world Type = Equal-Area Cylindrical New Projection Units = kilometres in the next window, specify the new cell size = 1 choose Interpolation Method = Bilinear Interpolation and Transformation Order = 4 save the output grid as st_ar_dem_km 84
Annex 8. Creation of a 300 km buffer around each climatic zone and clipping of the different layers for the regression analysis In order to insure good interpolation at the edge of each of the 5 climatic zones considered in this version of the e-atlas to conduct the regression analysis, a 300 km buffer was added to each of these zones using the following process: 1. Extract the five zones (see section 4.8.4) from the projected version of the international boundaries layer (See Annex 7) using the following steps in ArcView: a. b. c. d. e. f.
In ArcView, make sure that the XTools Extension is uploaded add the projected version of the international boundaries layer st_ar_int_bord_km.shp, and open its attribute table, select the countries part of the first zone (see section 4.8.4) In the view, make the st_ar_int_bord_km.shp shapefile the active theme use the Theme>Convert To Shapefile function to create a shapefile containing only the countries part of the first zone and save it as zone1_int_bord_km.shp repeat steps c) to e) for the other four zones.
2. Create the 300 km buffer around each climatic zone as follows: a. b. c. d. e.
f.
from the View>Properties change the Map Units and Distance Units to kilometres select Theme>Create Buffer; the Create Buffers wizard appears in the first wizard dialog box, choose zone1_int_bord_km.shp as the items to buffer in the second dialogue box, select At a Specified Distance as the method used to create the buffer and Width of the Buffer = 300. Make sure that the distance units are set to kilometres in the third dialogue box, create the buffer using Only Inside of the Polygon Parameter. Specify that the buffer be saved as a new theme and save as zone_1_int_bord_buffer_300.shp. The new buffer theme will be added to the current view Repeat steps c) to e) on the other four zones.
Taking the projected version of the weather station location layer (st_ar_PRC_km.shp) as an example, the following steps have then been applied to clip each vector layer to the 5 buffered climatic zones. 1. In ArcView, make sure that the XTools Extension is uploaded, 2. In the view, display both the buffered international boundaries of the first zone zone1_int_bord_buffer_300.shp and the projected layer containing the distribution of the weather stations with the associated annual maximum precipitation st_ar_PRC_km.shp. 3. Use the XTools> Clip with Polygon(s) function. 85
4. Select st_ar_PRC_km.shp as the theme that contains features that you wish to clip. 5. Select zone1_int_bord_buffer_300.shp as the polygon theme that contains the polygons that will be used as the reference for the clipping. 6. Specify the name for the new shapefile to be created as zone_1_PRC.shp. 7. Repeat steps 2 to 6 for the others zones For the raster layers, taking the DEM as an example, the following steps are applied for the clipping. 1. In ArcView, make sure that the Grid Analyst extension is uploaded, 2. In the view, add the st_ar_dem_km grid and zone1_int_bord_buffer_300.shp shape file. 3. Make the first grid st_ar_dem_km active and use the Grid Analyst>Extract Grid Theme Using Polygon function. 4. Select first the zone1_int_bord_buffer_300.shp from the drop list to use in the clip. 5. Make the resulting grid the active theme and select the Theme>Convert to Grid function to save the output grid under zone1_dem. 6. Repeat steps 2 to 5 for the four other zones.
86
Annex 9. Final regression for a five year return period by climatic zone
Climatic Zone 2 Variable (Intercept) d_Y d_X d_coast Z9 d_X2 Z92 Residual standard error Degrees of freedom Multiple R2 F statistic Probability (F statistic)
Regression coefficient -482.3276281059 0.0329408748 0.1780254227 -0.0987618331 -0.0258939062 -0.0000164114 0.0000154675
Standard error 18.14147 0.00623 0.06603 0.02621 0.01751 0.00000 0.00000
t value -2.86858 5.28542 2.69602 -3.76698 -1.47840 -2.53983 1.63430
Probability Pr(>|t|) 0.00050 0.00000 0.00010 0.00080 0.00010 0.00000 0.00000
5.71174 113 0.79 6.88379 0.00000
Climatic Zone 3 Variable (Intercept) d_coast d_Y d_X2 d_Y2 Z9 Z92 d_Y3 Residual standard error Degrees of freedom Multiple R2 F statistic Probability (F statistic)
Regression coefficient -1207.820660836 -0.018302060 0.949934571 0.000000148 -0.000222914 -0.025113354 0.000004012 0.000000016 4.71174 110 0.74 14.6523 0.00000
87
Standard error 22.73147 0.00617 0.23718 0.00000 0.00005 0.00614 0.00000 0.00000
t value -3.74249 -2.96171 4.00501 1.60518 -3.91458 -4.08767 2.09815 3.71971
Probability Pr(>|t|) 0.00021 0.00325 0.00007 0.00931 0.00010 0.00005 0.00657 0.00023
Climatic Zone 4 Variable (Intercept) d_coast d_X d_Y d_X2 Z9 Z92 Residual standard error Degrees of freedom Multiple R2 F statistic Probability (F statistic)
Regression coefficient
Standard error
-408.1495262763 -0.0399317927 0.1525505904 0.0299113702 -0.0000139276 -0.0516409529 0.0000199815
104.29460 0.01311 0.03881 0.00551 0.00000 0.01269 0.00000
t value -3.91342 -3.04433 3.93014 5.42775 -3.80115 -4.06900 2.88038
Probability Pr(>|t|) 0.00013 0.00272 0.00012 0.00000 0.00020 0.00000 0.00451
5.27682 331 0.71 11.98388 0.00000
Climatic Zone 5 Variable (Intercept) d_coast Z9 d_X d_coast2 d_Y2 d_X2 Residual standard error Degrees of freedom Multiple R2 F statistic Probability (F statistic)
Regression coefficient 81.4554730534 -0.1342869543 0.0048612269 0.0263277069 0.0001833458 -0.0000113164 -0.0000064917 3.60292 659 0.77 30.36604 0.00000
88
Standard error 4.54884 0.02228 0.00297 0.00464 0.00004 0.00000 0.00000
t value 17.90684 -6.02537 1.63403 5.66395 4.22091 -7.43871 -5.64095
Probability Pr(>|t|) 0.00000 0.00000 0.00269 0.00000 0.00002 0.00000 0.00000
Annex 10. Metadata for the e-atlas region datasets Annex 10.1. Metadata for the international boundaries layer Dataset title
International boundaries of the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, coastline and international boundaries, political boundaries
Dataset topic category
International boundaries
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2010
Data exchange format
ArcView shapefile
Filename
st_int_bord.shp
Abstract
The international boundaries used for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) have been extracted from the global dataset produced by the United Nations International and Administrative Boundaries Task Group of the UN Geographic Information Working Group (UNGIWG)
Lineage
The global dataset has been extracted from the original dataset to cover the 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Data quality comments
This dataset is mapped at 1:100 000 scale and covers the 100 countries comprising the WHO Regions
Source map format
Digital map
Source map name
UN International Boundaries Database
Distributor
UN Cartographic Section
Spatial representation type
Vector
Geometric object type
Polygons
Map projection
Unprojected (Geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287°
89
Y min: –46.978931°, Y max: 63.459827° Access and use constraints
Free for UN institutions
Acknowledgements
UN 2011
Disclaimer
The boundaries and names shown and the designations used on this map do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the status of any country, territory, city, or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries All reasonable precautions have been taken by WHO to produce this layer. However this layer is being distributed without warranty of any kind, either express or implied regarding its content. The responsibility for its interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use
Online linkage
http://boundaries.ungiwg.org/
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
90
Annex 10.2. Metadata for the road network distribution layer Dataset title
Spatial distribution of road network distribution for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, roads, transport, infrastructure
Dataset topic category
Transportation
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2009
Data exchange format
ArcView shapefile
Filename
st_roads.shp
Abstract
This layer contains the spatial distribution of roads network for WHO Regions (Africa, Eastern Mediterranean and part of Europe). This dataset contains four types of road: 1) motorways/highways, 2) major trunk roads, 3) primary roads , 4) tracks, trails or footpaths
Lineage
The process used to extract the roads network layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.2 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The maximum horizontal error is reported to be lower than 2040 metres rounded to nearest 5 metres at 90% circular error, World Geodetic System (WGS-84). Nevertheless some GPS tests done worldwide have shown that this error could in fact be lower than 500 metres. Global Insight [Plus] is not a replacement for high-integrity datasets, such as airport databases, which are approved for critical use. If there is any doubt as to the suitability of Global Insight [Plus], consult Europa Technologies before using the product
Source map format
Digital map
Source map name
Global Insight Plus
91
Distributor
Europa Technologies
Spatial representation type
Vector
Geometric object type
Line
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
All data from Europa Technologies are licensed, not sold, and are subject to a standard end-user licence agreement (EULA) The standard Europa Technologies single user licence agreement (EULA) allows the data to be installed and used at a single computer workstation. Printed output and digital images of the map data can be distributed within the licensees’ organizations. Distributions outside the organization in printed or electronic form, including publishing on the internet, is covered by a separate license agreement. Please refer to the following web site for more information: http://www.europa-tech.com/
Acknowledgement
© 2009 Europa Technologies Ltd.
Online linkage
http://www.europa-tech.com/
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115 92
Annex 10.3. Metadata for the hydrographic network distribution layer Dataset title
Spatial distribution of drain network for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe , water resources, surface water, rivers, inland water, lake
Dataset topic category
Inland waters
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2009
Data exchange format
ArcView shapefile
Filename
st_drain_l.shp, st_drain_p.shp
Abstract
This layer contains the drain network lines (st_drain_l.shp) and polygons (st_drain_p.shp)) for WHO Regions (Africa, Eastern Mediterranean and part of Europe). This dataset contains two types of drain: perennial and non-perennial water
Lineage
The process used to extract the hydrographic network layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.3 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The maximum horizontal error is reported to be lower than 2040 metres rounded to nearest 5 metres at 90% circular error, World Geodetic System(WGS-84) Nevertheless some GPS tests done worldwide have shown that this error could in fact be lower than 500 metres Global Insight [Plus] is not a replacement for high-integrity datasets, such as airport databases, which are approved for critical use. If there is any doubt as to the suitability of Global Insight [Plus], consult Europa Technologies before using the product
Source map format
Digital map
Source map name
Global Insight Plus 93
Distributor
Europa Technologies
Spatial representation type
Vector
Geometric object type
Line and polygon
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
All data from Europa Technologies are licensed, not sold, and are subject to a standard end-user licence agreement (EULA) The standard Europa Technologies single user licence agreement (EULA) allows the data to be installed and used at a single computer workstation. Printed output and digital images of the map data can be distributed within the licensee organizations. Distributions outside the organization in printed or electronic form, including publishing on the Internet, is covered by a separate licence agreement Please refer to the following web site for more information: http://www.europa-tech.com/
Acknowledgements
© 2009 Europa Technologies Ltd.
Online linkage
http://www.europa-tech.com/
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC) El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30
Metadata contact
email:
[email protected] Metadata date Metadata language Metadata character set Metadata standard
20110301 English ASCII ISO 19115 94
Annex 10.4. Metadata for the surface geology distribution layer Dataset title
Spatial distribution of geology for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, geologic age, surface, lithology
Dataset topic category
Geology
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
1998 (Arabian Peninsula, South Asia, former Soviet Union), 1999 (Islamic Republic of Iran), 2002 (Africa), 2003 (Europe including Turkey)
Data exchange format
ArcView shapefile
Filename
st_geology.shp
Abstract
This layer contains the general geological age and bedrock type The description of the six digital maps: Surficial geology of Africa, Bedrock geology of the Arabian Peninsula and selected adjacent areas, Surficial geology of Iran, Geological map of South Asia, Generalized geology of the former Soviet Union and Generalized geology of Europe including Turkey have been used to create this dataset. This compilation is part of a map series of the world produced by the US Geological Survey World Energy Project
Lineage
The process used to produce the geology layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.4 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The attribute data identifies the geology of each polygon by either age of the rock formations as they occur on the surface, or rock type (i.e. sedimentary, igneous, volcanic, metamorphic, etc.) or facies (i.e. eolian, fluvial, salt, etc.). Overall accuracy of this attribute is based on the accuracy of the four source digital maps Even if the dataset covers the 100 countries comprising the WHO Regions, some islands were not mapped; they have 95
been assigned to “no data” Source map format
Digital map
Source map names
Surficial geology of Africa Bedrock geology of the Arabian Peninsula and selected adjacent areas Surficial geology of Iran Geologic map of South Asia Generalized geology of the former Soviet Union Generalized geology of Europe including Turkey
Distributor
US Geological Survey Earth Science Information Center (ESIC)
Spatial representation type
Vector
Geometric object type
Surface
Map projection
Unprojected (eographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
Public domain data from the US Government are freely redistributable. Please recognize the US Geological Survey (USGS) as the source of this information There is no legal requirement for users to acknowledge or credit USGS as the source for public domain information, but they may wish to do so as a courtesy. If you wish to acknowledge or credit USGS as an information source of data or products, use the following line of text: “Source of the geology data: US Geological Survey”
Acknowledgements
US Geological Survey
Online linkage
http://energy.usgs.gov/
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 96
80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected] Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
97
Annex 10.5. Metadata for the tectonic layer Dataset title
Spatial distribution of tectonic activity for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, tectonic, DTAM, faults, fault zones, structural geology, fracturation, tectonic maps, plate tectonics, earthquakes, lineaments, seismic, tectonophysics
Dataset topic category
Tectonic
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2002
Data exchange format
ArcView shapefile
Filename
all_tectonic.shp
Abstract
This dataset contains the location of several types of faults and active spreading centres for the WHO Regions (Africa, Eastern Mediterranean and part of Europe), extracted from the Digital Tectonic Activity Map (DTAM). The DTAM was created using numerous remotely sensed and GIS databases (seismicity, volcanism, and plate motions) and could be considered as a unique tool for understanding the physiographic nature of the Earth
Lineage
The process used to produce the tectonic layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.5 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) The DTAM is distributed in image formats (.jpg, .gif and .tif) in the Robinson projection
98
Data quality comments
The use of both a GIS historical database (with scale 1:24 000) and satellite imagery (with 30-metre resolution) provides us with a good degree of confidence regarding the quality of the Digital Tectonic Activity Map
Source map format
Image
Source map name
Digital Tectonic Activity Map (DTAM)
Distributor
NASA Goddard Space Flight Center (GSFC), Greenbelt, Maryland, US
Spatial representation type
Vector
Geometric object type
Line
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
The original Digital Tectonic Activity Map is free and cleared for general use
Acknowledgements
NASA modified by WHO (2010)
Disclaimer
All reasonable precautions have been taken by WHO to produce this layer. However this layer is being distributed without warranty of any kind, either express or implied regarding its content. The responsibility for its interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use
Online linkage
http://denali.gsfc.nasa.gov/dtam/data.html
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115 99
Annex 10.6. Metadata for the location of the weather stations Dataset title
Spatial distribution of location of the weather stations for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, precipitation, temperature, wind speed, dew point, sea level pressure, station pressure, humidity relative, weather stations
Dataset topic category
Weather stations
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Representativity
until 2009
Data exchange format
ArcView shapefile
Filename
st_stations.shp
Abstract
This layer contains the location of the weather stations coming from global surface summary of day data produced by the US National Climatic Data Center (NCDC). It also contains the indication of the station number, station name, country/state ID and elevation
Lineage
The process used to produce the location of the weather stations layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.6 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Source map format
Text file
Source map name
Global Surface Summary of Day
Distributor
US National Climatic Data Center (NCDC)
Spatial representation type
Vector
Geometric object type
Point
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
100
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
Data summaries and products which are available through the Global Surface Summary of the Day dataset are intended for free and unrestricted use in research, education, and other non-commercial activities. However, for non-US locations’ data, the data or any derived product shall not be provided to other users or be used for the re-export of commercial services
Acknowledgements
US National Climatic Data Center (NCDC) modified by WHO 2010
Online linkage
ftp://ftp.ncdc.noaa.gov/pub/data/inventories/
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
101
Annex 10.7. Metadata for the number of previous flood events distribution layer Dataset title
Spatial distribution of number of previous flood events for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, flood, inundation, flooded lands, wetlands
Dataset topic category
Environment
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Representativity
1985–2009
Data exchange format
ArcView grid
Filename
st_flood_fr
Abstract
This dataset contains the number and extension of the flood observed between 1985 and 2009 for WHO Regions (Africa, Eastern Mediterranean and part of Europe). It is based on the flood polygons coming from the Global Active Archive of Large Flood Events at Dartmouth Flood Observatory. These data are derived from a wide variety of news, governmental, instrumental and remote sensing source
Lineage
The process used to produce the flood frequency layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 3.7 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
102
Data quality comments
The statistics presented in the Dartmouth Flood Observatory Global Archive of Large Flood Events are derived from a wide variety of news and governmental sources. The quality of information available about a particular flood is not always in proportion to its actual magnitude, and the intensity of news coverage varies from nation to nation. In general, news from floods in less developed countries tends to arrive later and be less detailed than information from developed countries Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level
Source map format
Digital map
Source map name
Global Active Archive of Large Flood Events
Distributor
Dartmouth Flood Observatory
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc seconds (0.008333 degrees)
Access and use constraints
Dartmouth Flood Observatory (DFO) data are available to users free and unrestricted under the conditions below In making a request, the data user agrees that the Observatory may inform its funding sponsors about the use to which the data have been put and may transfer the names and addresses of the data users to these sponsors. Sponsors have included or do include the National Aeronautics and Space Agency (US), NATO (Science for Peace Programme) and the World Meteorological Organization (UN)
103
DFO can make available subsets of its database on request, as stated above. Requests for the entire database or substantial parts of it, or for updating thereof as new floods are recorded, cannot be entertained without entering into a cooperative agreement Please give credit to original source and to WHO when using these data as reported in the acknowledgements Acknowledgements
Dartmouth Flood Observatory modified by WHO (2010)
Disclaimer
All reasonable precautions have been taken by WHO to produce this layer. However this layer is being distributed without warranty of any kind, either express or implied regarding its content. The responsibility for its interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use.
Online linkage
http://floodobservatory.colorado.edu/Archives/index.html
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
104
Annex 10.8. Metadata for the Digital Elevation Model layer Dataset title
Spatial distribution of Digital Elevation Model (DEM) for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, Digital Elevation Model, DEM, Elevation, Altitude
Dataset topic category
Elevation
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2000
Data exchange format
ArcView grid
Filename
emro_dem
Abstract
This dataset contains the spatial distribution Digital Elevation Model (DEM) for WHO Regions (Africa, Eastern Mediterranean and part of Europe) based on the SRTM30 data sensed by the Shuttle Radar Topography Mission, flown in February 2000
Lineage
The process used to produce the Digital Elevation Model layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.1 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe
Data quality comments
The SRTM 90 m resampled to 1 km resolution , but has a better quality than GTOPO30 Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level
Source map format
Digital image map
Source map name
SRTM 3 arc-seconds
Distributor
US Geological Survey’s EROS Data Center
Spatial representation type
Grid
Map projection
Unprojected (geographic) 105
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc-seconds (0.008333 degrees)
Access and use constraints
Public domain data from the US Government are freely redistributable. Please recognize the US Geological Survey (USGS) as the source of this information There is no legal requirement for users to acknowledge or credit USGS as the source for public domain information, but they may wish to do so as a courtesy. If you wish to acknowledge or credit USGS as an information source of data or products, use the following line of text: “Source of the SRTM30 data: US Geological Survey”
Acknowledgements
US Geological Survey
Online linkage
http://seamless.usgs.gov/products/srtm3arc.php
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
106
Annex 10.9. Metadata for the slope distribution layer Dataset title
Spatial distribution of slope for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, slopes, Digital Elevation Model, DEM, topography
Dataset topic category
Slope
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2010
Data exchange format
ArcView grid
Filename
st_slp_dd, st_slp_pr
Abstract
This dataset contains the terrain slope for the WHO Regions (Africa, Eastern Mediterranean and part of Europe). It was developed by calculating the maximum rate of change between each cell and its eight neighbours from the SRTM3 elevation dataset resampled to 1 km. Each pixel is attributed with a value that represents the slope, in degrees (st_slp_dd) and as a percentage (st_slp_pr). These layers are distributed as integer data. However, in order to carry as much information as possible, the floating point data were multiplied by 100 and then converted to integer. For example, a slope value of 32.75 is represented in the slope data layer as 3275.
Lineage
The process used to produce the slope layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.2 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The accuracy of this dataset is directly related to the accuracy of the DEM Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level
Source map format
Digital map
Source map name
SRTM 3 arc-seconds 107
Distributor
US National Aeronautics and Space Administration (NASA)
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc-seconds (0.008333 degrees)
Access and use constraints
The DEM from which the slopes have been derived is in the public domain. Please recognize the US Geological Survey (USGS) as the source of the Digital Elevation Model used to derive this information Please give credit to original source and WHO when using this data as reported in the acknowledgements
Acknowledgements
US Geological Survey modified by WHO (2010)
Disclaimer
All reasonable precautions have been taken by WHO to produce this layer. However this layer is being distributed without warranty of any kind, either express or implied regarding its content. The responsibility for its interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use
Online linkage
Under construction
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
108
Annex 10.10. Metadata for the aspect distribution layer Dataset title
Spatial distribution of aspect for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe , aspect, Digital Elevation Model, DEM, topography, direction, orientation
Dataset topic category
Aspect
Geographic location
Afghanistan, Bahrain, Djibouti, Egypt, Iraq, Islamic Republic of Iran, Jordan, Kuwait, Lebanon, Libyan Arab Jamahiriya, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Somalia, Sudan, Syrian Arab Republic, Tunisia, United Arab Emirates, West Bank and Gaza Strip, Yemen
Publication date
2010
Data exchange format
ArcView grid
Filename
st_asp
Abstract
This layer contains the spatial distribution of slope direction (aspect) for WHO Regions (Africa, Eastern Mediterranean and part of Europe), derived from the SRTM3 elevation dataset resampled to 1 km dataset. This layer describes the maximum rate of change in the elevations between each cell and its eight neighbours. These layers are distributed as integer data. However, in order to carry as much information as possible, the floating point data were multiplied by 100 and then converted to integer. For example, an aspect value of 150.48 is represented in the aspect data layer as 15048.
Lineage
The process used to produce the aspect layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.3 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The accuracy of this dataset is directly related to the accuracy of the DEM Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level 109
Source map format
Digital map
Source map name
SRTM 3 arc-seconds
Distributor
US National Aeronautics and Space Administration (NASA)
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc seconds (0.008333 degrees)
Access and use constraints
The DEM from which the aspect has been derived is the public domain. Please recognize the US Geological Survey (USGS) as the source of the Digital Elevation Model used to derive this information Please give credit to original source and WHO when using this data as reported in the acknowledgements
Acknowledgements
US Geological Survey modified by WHO (2010)
Disclaimer
All reasonable precautions have been taken by WHO to produce this layer. However this layer is being distributed without warranty of any kind, either express or implied regarding its content. The responsibility for its interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use
Online linkage
Under construction
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115 110
Annex 10.11. Metadata for the flow accumulation distribution layer Dataset title
Spatial distribution of flow accumulation for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, flow accumulation, drainage basin boundaries, stream lines
Dataset topic category
Flow accumulation
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2010
Data exchange format
ArcView grid
Filename
st_fa
Abstract
This layer contains the spatial distribution of flow accumulation for WHO Regions (Africa, Eastern Mediterranean and part of Europe). Derived from the HydroSHEDS dataset, this layer defines the amount of upstream area draining into each cell and is useful for generating watershed boundaries and stream network
Lineage
The process used to produce the flow accumulation layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.4 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The accuracy of this dataset is better than HYDRO1k, a global hydrographic data set at 1 km resolution (USGS 2000), due to HydroSHEDS being based on a superior digital elevation model SRTM3 Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level
Source map format
Digital map
Source map name
HydroSHEDS
Distributor
US Geological Survey
Spatial representation type
Grid 111
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc seconds (0.008333 degrees)
Access and use Constraints
Users may use HydroSHEDS for non-commercial purposes. Any modification of the original data by users must be noted. The authors of HydroSHEDS may request reprints of publications and copies of derived materials. The user shall not reproduce, convert, (re)publish, (re)distribute, (re)broadcast, (re)transmit, sell, rent, lease, sublicense, lend, assign, time-share, or transfer, in whole or in part, or provide unlicensed third parties access to the data and related materials without explicit written permission from the authors
Acknowledgements
The Conservation Science Program of World Wildlife Fund (WWF), U.S. Geological Survey (USGS); the International Centre for Tropical Agriculture (CIAT); the Nature Conservancy (TNC); the Government of Australia; McGill University, Montreal, Canada; and the Center for Environmental Systems Research (CESR) of the University of Kassel, Germany.
Online linkage
http://hydrosheds.cr.usgs.gov
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
112
Annex 10.12. Metadata for the land cover distribution layer Dataset title
Spatial distribution of land cover for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, land cover, environment, natural resources, agriculture, forest
Dataset topic category
Land cover
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2004–2006
Data exchange format
ArcView grid
Filename
st_lc
Abstract
This layer contains the spatial distribution of land cover coming from the ESA 2004–2006 Globcover database. These data are based on ENVISAT’s Medium Resolution Imaging Spectrometer (MERIS) Level 1B data acquired in full resolution mode with a spatial resolution of 300 metres
Lineage
The process used to produce the land cover layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.5 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
The GlobCover land cover validation based on 3167 points shows that the global accuracy is 73.14% . Thus, the GlobCover land cover dataset can be considered to be of reasonable quality
Because of the methods and resolution used (1 kilometre) special care should be taken when using this dataset for application below the national level Source map format
Digital map
Source map name
GlobCover land cover map
Distributor
Pôle d’Observation des Surfaces continentales par Télédétection (POSTEL)
Spatial representation type
Grid 113
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc seconds (0.008333 degrees)
Access and use constraints
Public domain data from the POSTEL Service Centre with the agreement of ESA. Please recognize POSTEL as the source of this information
Acknowledgements
POSTEL
Online linkage
http://postel.mediasfrance.org
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
114
Annex 10.13. Metadata for the soil type distribution layer Dataset title
Spatial distribution of soil type distribution for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, soil type
Dataset topic category
Soil
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2009
Data exchange format
ArcView grid
Filename
st_soil_type
Abstract
This layer describes the soil types observed for WHO Regions (Africa, Eastern Mediterranean and part of Europe). It has been derived from the 2009 FAO/IIASA/ISRIC/ISSCAS/JRC Harmonized World Soil Database. In this layer the soil types are organized into 34 classes. Sand dunes, salt flats, rock debris, glaciers and water are also reported
Lineage
The process used to produce the soil type layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.6 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Dataset edition
Version 1.1
115
Data quality comments
Reliability of the information contained in the database is variable: the parts of the database that still make use of the Soil Map of the World such as west Africa and south Asia are considered less reliable, while most of the areas covered by SOTER databases are considered to have the highest reliability (central and southern Africa, central and eastern Europe). Even if the dataset covers the 100 countries comprising the WHO Regions, some islands were not mapped; they have been assigned to “no data”
Source map format
Digital map
Source map name
Harmonized World Soil Database
Distributor
International Institute for Applied Systems Analysis (IIASA)
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
2008-2009 COPYRIGHT FAO, IIASA, ISRIC, ISSCAS, JRC All rights reserved. No part of this Harmonized World Soil Database may be reproduced, stored in a retrieval system or transmitted by any means for resale or other commercial purposes without written permission of the copyright holders. Reproduction and dissemination of material in this information product for educational or other non commercial purposes are authorized without any prior written permission from the copyright holders provided the source is fully acknowledged. Full acknowledgement and referencing of all sources must be included in any documentation using any of the material contained in the Harmonized World Soil Database, as follows: Citation FAO/IIASA/ISRIC/ISSCAS/JRC, 2009. Harmonized World Soil Database (version 1.1). FAO, Rome, Italy and IIASA, Laxenburg, Austria
Acknowledgements
FAO/IIASA/ISRIC/ISS-CAS/JRC
Online linkage
http://www.iiasa.ac.at/Research/LUC/External-World-soildatabase/HTML/HWSD_Data.html?sb=4
Dataset language
English 116
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
117
Annex 10.14. Metadata for the soil texture distribution layer Dataset title
Spatial distribution of soil texture distribution for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe , texture, soil
Dataset topic category
Soil
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
2009
Data exchange format
ArcView grid
Filename
st_t_text, st_s_text, st_spl_t_txt
Abstract
The present dataset is the digital soil texture of the WHO Regions (Africa, Eastern Mediterranean and part of Europe), derived from 2009 FAO/IIASA/ISRIC/ISSCAS/JRC Harmonized World Soil Database. These dataset includes three soil texture layers for the e-atlas (topsoil texture with 13 classes (st_t_text), subsoil texture with 13 classes (st_s_text ) and the simplified topsoil texture with 3 classes (st_spl_t_txt).
Lineage
The process used to produce the soil texture layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.7 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Data quality comments
Reliability of the information contained in the database is variable: the parts of the database that still make use of the Soil Map of the World such as west Africa and south Asia are considered less reliable, while most of the areas covered by SOTER databases are considered to have the highest reliability (central and southern Africa, central and eastern Europe). Even if the dataset covers the 100 countries comprising the WHO Regions, some islands were not mapped; they have been assigned to “no data”
Source map format
Digital map 118
Source map name
Harmonized World Soil Database
Distributor
International Institute for Applied Systems Analysis (IIASA)
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Access and use constraints
2008-2009 COPYRIGHT FAO, IIASA, ISRIC, ISSCAS, JRC All rights reserved. No part of this Harmonized World Soil Database may be reproduced, stored in a retrieval system or transmitted by any means for resale or other commercial purposes without written permission of the copyright holders. Reproduction and dissemination of material in this information product for educational or other non commercial purposes are authorized without any prior written permission from the copyright holders provided the source is fully acknowledged. Full acknowledgement and referencing of all sources must be included in any documentation using any of the material contained in the Harmonized World Soil Database, as follows: Citation FAO/IIASA/ISRIC/ISSCAS/JRC, 2009 and WHO, 2010. Harmonized World Soil Database (version 1.1). FAO, Rome, Italy and IIASA, Laxenburg, Austria
Acknowledgements
FAO/IIASA/ISRIC/ISS-CAS/JRC modified by WHO (2010)
Online linkage
http://www.iiasa.ac.at/Research/LUC/External-World-soildatabase/HTML/HWSD_Data.html?sb=4
Dataset language
English
Dataset character set
ASCII
Metadata provider
World Health Organization Regional Office for the Eastern Mediterranean
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110301
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115 119
Annex 10.15. Metadata for the annual maximum total precipitations over 3 consecutive days distribution layer (five year return period) Dataset title
Spatial distribution of annual maximum total precipitation over 3 consecutive days for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
Theme keywords
WHO, Africa, Eastern Mediterranean, Europe, Geographic Information System (GIS), precipitation, rainfall,
Dataset topic category
Precipitations
Geographic location
The layer covers a total of 100 countries (22 for the Eastern Mediterranean, 46 for Africa and 32 for Europe)
Publication date
20110601
Representativity
1997–2008
Data exchange format
ArcView grid
Filename
Afro_PRC_5, emro_PRC_5, euro_PRC_5
Abstract
This dataset contains the distribution of the annual maximum total precipitations over 3 consecutive days for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) and a five year return period.
Lineage
The process used to produce the annual maximum total precipitations over 3 consecutive days layer for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) is described in section 4.8 of the Methodology and implementation process for generating the dataset document that can be found in the first volume (2nd version) of the WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe)
120
Data quality comments
Because of the resolution used (1 km), special care should be taken when using this dataset for application below the national level Even though the dataset covers all of the WHO Regions, the lack of historical data for many countries reduces the quality of the dataset for those areas
Distributor
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Spatial representation type
Grid
Map projection
Unprojected (geographic)
Reference system
WGS-84 datum
Geographic box
X min: –25.358747°, X max: 91.8287° Y min: –46.978931°, Y max: 63.459827°
Resolution
30 arc seconds (0.008333 degrees)
Redistributions constraints
The maximum precipitations distribution layers are copyrighted. The owner of the data agrees to the use, reproduction, distribution, display, publication and dissemination at no cost to third parties of the maximum precipitations distribution layer, in any manner and in any form whatsoever, subject to the copyright and acknowledgement mentioned in this metadata
Access and use constraints
These layers may not be reproduced, changed, adapted, translated, stored in a retrieval system or transmitted in any form or by any means without prior permission of the copyright holder, except to make a security backup. Requests for permissions, with a statement of purpose and extent, should be address to the VRAM programme at the WHO Mediterranean Centre for Health Risk Reduction (
[email protected])
Acknowledgement
WHO e-atlas of disaster risk for the WHO Regions (Africa, Eastern Mediterranean and part of Europe) 2nd version. Copyright © WHO 2011. All rights reserved
Disclaimer
All reasonable precautions have been taken by WHO to produce these layers. However these layers are being distributed without warranty of any kind, either express or implied regarding their content. The responsibility for their interpretation and use lies with the user. In no event shall the World Health Organization be liable for damages arising from its use 121
Online linkage
Under construction
Dataset language
English
Dataset character set
ASCII
Metadata provider
WHO Mediterranean Centre for Health Risk Reduction (WMC)
Metadata contact
El Morjani Zine El Abidine BP 3566 Poste Talborjt 80000 Agadir Morocco Telephone: +212 528 28 55 30 email:
[email protected]
Metadata date
20110601
Metadata language
English
Metadata character set
ASCII
Metadata standard
ISO 19115
122
