9 Research Branch Technical Bulletin 1998-3E
Broad-scale Assessment of Agricultural Soil Quality in Canada Using Existing Land Resource Databases and GIS
K.B. MacDonald and F. Wang Soil Program at Guelph Greenhouse and Processing Crop Research Centre Agriculture and AgriFood Canada
W.R. Fraser and G.W. Lelyk Land Resource Unit Brandon Research Centre Agriculture and AgriFood Canada
Canada
Broad-scale Assessment of Agricultural Soil Quality in Canada Using Existing Land Resource Databases and GIS
_
K.B. MACDONALD AND F. WANG Soil Program at Guelph Greenhouse and Processing Crop Research Centre, Research Branch Agriculture and Agri-Food Canada 70 Fountain Street Guelph, ON NIH 3N6 W.R. FRASER AND G.W. LELYK Land Resource Unit Brandon Research Centre, Research Branch Agriculture and Agri-Food Canada Ellis Building, University of Manitoba Winnipeg, MB R3T 2N2
Technical Bulletin 1998-3E Research Branch Agriculture and Agri-Food Canada
Hardcopy of this publication could be obtained from: Soil Program at Guelph, Greenhouse and Processing Crop Research Centre Research Branch, Agriculture and Agri-Food Canada 70 Fountain Street E. Guelph, ON NIH 3N6 Tel: 519-8262086 Fax: 519-8262090 Email:
[email protected] The hyper-text version of this document is available at AFFC's intranet: http://l42.61 .197.4/olru/sq/
@Minister of Public Works and Government Services Canada 1998 Cat. No. A54-8/1998-3E ISBN 0-662-26774-5
TABLE OF CONTENTS Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii Sommaire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix 1.0 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.0 Conceptual Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Current Understanding on Soil Quality: An Overview . . . . . . . . . . . . . . . . . . . 2.2 A Hierarchical Framework of Soil Quality Assessment . . . . . . . . . . . . . . . . . . . 2.3 Concepts of Inherent Soil Quality (ISQ) and Soil Quality Susceptibility (SQS) 2.4 Basic Procedures of Soil Quality Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Approaches Specific to This Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.0 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 GIS System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Spatial Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 ISQ Rating Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 SQS Indicators and Spatial Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.0 Results and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Major Agricultural Regions of Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Current Status of Inherent Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 ISQ and Potential Land Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 .4 Areas Susceptible to Change in Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Implications of Changing Land Use and Management Practices on Soil Quality 5.0 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 .1 The Sensitivity of ISQ Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Future Applications of ISQ and SQS Procedures . . . . . . . . . . . . . . . . . . . . . . . 6.0 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
. . .. . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . .. .. .. .. ..
. . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . .
. .1 . .2 . .2 . .4 . .6 . 10 . 11 .13 .13 .13 . 15 . 16 . 26 . 29 . 29 . 30 . 47 . 49 . 56 .60 . 60 . 67 .69
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70 Appendix 1 Key Terms and Acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix 2 Descriptions of Soil and Landscape Data Attributes (items) from CanSIS/NSDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix 3 A Detail Description of ISQ92 Procedures . . . . . . . . . . . . . . . . . Appendix 4 A Detail Description of ISQ94 Procedures . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . 76 . . . . . . . . . . . . .77 . . . . . . . . . . . . . 82 . . . . . . . . . . . . . 85
List of Figures Figure 2-1 Multi-dimensional perspective on soil quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Figure 2-2 Soil health/quality as an indicator of environmental/ecosystem health . . . . . . . . . . . . 4 Figure 2-3 A hierarchical framework for soil quality assessment . . . . . . . . . . . . . . . . . . . . . . . . 5
Figure 2-4 Soil Quality change in relation to soil modifying processes and land use and management practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2-5 Kinds and direction of soil quality changes and research approaches . . . . . . Figure 2-6 Basic steps of soil quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3-1 Illustration of the spatial framework for soil quality assessment and reporting in Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3-2 The organization of ISQ procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3-3 Levels of ISQ generalization and mapping/reporting . . . . . . . . . . . . . . . . . . . Figure 3-4 GIS procedures to identify and map SQS indicators . . . . . . . . . . . . . . . . . . . Figure 4-1 The major agricultural regions of Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4-2 Regional differences of ISQ ratings in the Prairies provinces . . . . . . . . . . . . Figure 4-3 Inherent soil quality (ISQ) element map of the Prairies provinces (a,b,c,d,e) Figure 4-4 Regional differences of ISQ ratings in the Mixedwood Plains Ecozone . . . . . Figure 4-5 Inherent soil quality (ISQ) element map of the Mixedwood Plains Ecozone (a,b,c,d,e) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4-6 Soil quality susceptibility (SQS) map of the Prairies Provinces (a,b) . . . . . . . Figure 4-7 Soil quality susceptibility (SQS) map of the Mixedwood Plains Ecozone (a,b) Figure 4-8 Changes of selected SQS indicators of land use and management in the Prairies provinces from 1981 to 1991 . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4-9 Changes of selected SQS indicators of land use and management in the Mixedwood Plains Ecozone from 1981 to 1991 . . . . . . . . . . . . . . . . . Figure 5-1 Comparison of the results of ISQ92 and ISQ94 in the Prairies provinces . . . . Figure 5-2 Location of the scale sensitivity test area in Southern Manitoba . . . . . . . . . . Figure 5-3 Comparison of ISQ94 ratings at different scales in Southern Manitoba . . . . .
. . . . . .8 . .. . . .9 . . . . . 11 . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
15 17 24 28 29 33 34 41
. . . . . 42 . . . . . 52 . . . . 54 . . . . . 58 . . . .
. . . .
. . . .
59 61 63 63
Table 2-1 Inherent soil quality elements for crop production . . . . . . . . . . . . . . . . . . . . . . . . Table 2-2 Aspects of soil quality susceptibility to change . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3-1 Data sources for broad-scale assessment of soil quality in Canada . . . . . . . . . . . Table 3-2 . Rating scale of ISQ nutrient retention element . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3-3 Rating scale of ISQ physical rooting conditions element . . . . . . . . . . . . . . . . . . . Table 3-4 Rating scale of ISQ chemical rooting conditions element . . . . . . . . . . . . . . . . . . . Table 3-5 Matrix for determining rating points of ISQ overall chemical rooting conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3-6 Typical criteria for indicator selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3-7 Selected SQS indicators and the criteria and threshold values . . . . . . . . . . . . . . . Table 4-1 Proportion of area ISQ rated of total land area in the Prairies provinces . . . . . . . . Table 4-2 Summary of ISQ assessment in the Prairies provinces . . . . . . . . . . . . . . . . . . . . . Table 4-3 Proportion of area ISQ rated of total land area in the Mixedwood Plains Ecozone Table 4-4 Summary of ISQ assessment in the Mixedwood Plains Ecozone . . . . . . . . . . . . . Table 4-5 Potential land supply based on ISQ assessment in comparison to actual land use in the Prairies provinces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .
. . . . . .
.7 10 14 20 20 21
. . . . . . .
. . . . . . .
22 26 27 31 32 39 39
. . . .
. . . .
List of Tables
iv
. . 47
Table 4-6 Potential land supply based on ISQ assessment in comparison to actual land use in the Mixedwood Plains Ecozone . . . . . . . . . . . . . . . . . . . . . . . . . 48 . Table 4-7 Proportion of susceptible areas of soil quality change in the Prairies provinces . . . . . 50 Table 4-8 Proportion of susceptible areas of soil quality change in the Mixedwood Plains Ecozone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Table 4-9 Change in selected SQS indicators in the Prairies provinces from 1981 to 1991 . . . . 56 Table 4-10 Change in selected SQS indicators in the Mixedwood Plains Ecozone from 1981 to 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Table 4-11 Conservation tillage and no-till practices used in the Prairies provinces (1991) . . . . 57 Table 4-12 Conservation tillage and no-till practices used in the Mixedwood Plains Ecozone(1991) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57 Table 5-1 Comparison of ISQ94 and ISQ92 ratings in the Prairies provinces . . . . . . . . . . . . . . 62 Table 5-2 Comparison of ISQ94 ratings at different scales in Southern Manitoba . . . . . . . . . . . 64 Table 5-3 Comparison of detailed and broad scale ISQ ratings for selected SLC polygons in Southern Manitoba . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Table A2-1 SLC (version 1 .0) DOMinant and SUBdominant file attributes . . . . . . . . . . . . . . . 77 Table A2-2 SLC (version 1 .0) Component (CMP) file attributes . . . . . . . . . . . . . . . . . . . . . . 78 Table A2-3 SLC (version 1 .0) Carbon Layer (CLYR) file attributes . . . . . . . . . . . . . . . . . . . . 78 Table A2-4 DSM Soil Map Unit File (SMUF) attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Table A2-5 Soil Name File (SNF) attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Table A2-6 Soil Layer File (SLF) attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Table A3-1 Areas or soils excluded and the thresholds used in ISQ92 procedures . . . . . . . . . . 82 Table A3-2 Point system of ISQ92 rating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Table A3-3 The rating criteria of selected ISQ attribute used in ISQ92 procedures . . . . . . . . . 84 Table A4-1 Data attributes used in ISQ94 procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Table A4-2 Areas or soils excluded and the thresholds used in ISQ94 procedures . . . . . . . . . . 87 Table A4-3 Root restrictions and thresholds for layer exclusions . . . . . . . . . . . . . . . . . . . . . . . 87 Table A4-4 ISQ94 program variables for ISQ element rating . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Table A4-5 Matrix for determining rating points of ISQ aeration porosity . . . . . . . . . . . . . . . . 90 Table A4-6 Relationship between available water holding capacity and surface texture . . . . . . 90 Table A4-7 Matrix for determining rating points of ISQ available water holding capacity . . . . . 91 Table A4-8 Rating scale of ISQ nutrient retention element . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Table A4-9 Rating scale of ISQ physical rooting conditions element . . . . . . . . . . . . . . . . . . . . 92 Table A4-10 Rating scale of ISQ chemical rooting conditions element . . . . . . . . . . . . . . . . . . . 93 Table A4-11 Matrix for determining rating points of ISQ overall chemical rooting conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .94
ACKNOWLEDGEMENTS This research was mainly funded by the Soil Quality Evaluation Program (SQEP) through the National Soil Conservation Program (NSCP, http://res .agr.ca/lond/pmrc/nscp/ ) during 19911993. The State of the Environment Directorate, Environment Canada provided funds for Soil Quality Reporting in the Prairies in 1994 . The continuous development and improvement of the prototype system for assessing the inherent soil quality (ISQ) during 1995-96 was partially funded by the Canada-Ontario Agriculture Green Plan (http://res .agr .ca/lond/). The authors also wish to acknowledge people providing comments for development of the conceptual framework, particularly Drs. Don Acton and Wayne Pettapiece and early team members Mr. Andy Moore and Ian Jarvis .
EXECUTIVE SUMMARY Under the National Soil Quality Evaluation Program (SQEP), a series of research projects were conducted on soil and environmental quality in Canada. The main research findings have been summarized in a series of technical reports and a 1995 publication, The Health of Our Soils. The SQEP project to develop and demonstrate a system of soil quality assessment at national and regional scales is documented in this technical bulletin. The bulletin presents a conceptual framework which defines aspects of soil quality as either static assessments, termed inherent soil quality (ISQ), as a quasi-dynamic assessments termed soil quality susceptibility to change (SQS), or as actual soil quality change (SQC). It is not feasible to measure actual soil quality change at broad regional and national scales and current process-based models are not adapted to broad scale operation. Consequently, this study concentrates on the development of GIS procedures to assess ISQ and SQS regionally and nationally. The framework is used to outline the steps needed to conduct soil quality assessments (ISQ and SQS). The capability is demonstrated and results are presented in map and tabular form for the major agricultural regions of Canada. At a very general level the results are compared to other measures of agricultural land use and quality . The sensitivity of the assessments to different data sources is evaluated. Soil quality is a composite expression of properties and processes that interact to determine its ability to perform a number of basic functions, such as supporting crop production, buffering the environment from nutrients and other chemicals and partitioning water and gases. The inherent aspects of soil quality function are quite complex and have been simplified into components or, elements for assessment. Within the crop production function, four basic elements of soil quality were defined; (i) available porosity, (ii) nutrient retention, (iii) physical rooting conditions, and (iv) chemical rooting conditions . Attributes were selected from standard land resource data bases, and used in interpretive algorithms developed within the project to estimate each individual soil quality element. A series of indicators related to conditions of soil, landscape, land use and management practices were used to assess SQS. Indicators include attributes such as ; shallow topsoil, low organic carbon content of topsoil, steep slope, highly erodible surface texture, high intensity land use, and land management practices which expose the soil to degradation, . GIS procedures were used to assist in the spatial identification and mapping SQS indicators. ISQ and SQS assessment procedures were demonstrated in the major agricultural regions of Canada - the Prairies and the Boreal Plains ecozones in western Canada and the Mixed Wood Plains ecozone in eastern Canada. These three ecozones include about 91 % of the farmland and 95% of the cultivated land of Canada. Overall results from ISQ assessments indicate that approximately 37% of the land area in the 3 prairie provinces meet the climatic and soil requirements for spring seeded cereal crop production . In eastern Canada, approximately 83% of the land area within the Mixed Wood Plains ecozone in southern Ontario and Quebec meet the
climatic and soil criteria for cereal crop production. The ISQ assessment indicates that a relatively large proportion of agricultural soils in the three ecozone are in `Good' or `Good to Moderate' rating range. In the Prairies and the Boreal Plains ecozones, about 70-80% of the land area was rated `Good' or `Good to Moderate' for all four ISQ elements . Marginal ("Poor") ISQ ratings varied from 3 to 11 % for the different ISQ elements . At the overall ISQ rating level, about 46% received `Good' or `Good to Moderate' ratings, about 34% were rated as `Moderate to poor' and about 19% were rated as 'Poor' . The estimates of potential land supply by ISQ methods in the prairie provinces and the Mixed Wood Plains were close to the estimates of the first 4 agricultural capability classes of the Canada Land Inventory (CLI) . Comparing these figures to the actual land use as reported in the 1991 Census of Agriculture shows that nearly 85% of land suitable for agricultural use is currently being farmed . The SQS assessment indicated that for most provinces, the steep slope indicator identified a small but significant proportion of the potential agricultural land (9 - 15%) except for Quebec where the agricultural area was predominantly marine and fluvial sediments. Western Canadian soils were formed under grassland vegetation, with high organic carbon content in the surface horizon. eastern Canadian soils were formed under forest vegetation and have lower organic carbon content in the native state. They also have a longer history of intensive crop production. Low organic carbon content of the topsoil is a more serious problem in eastern Canada than in western Canada. Soil quality susceptibility to change due to land use and management factors (as recorded in the 1991 Census of Agriculture) varies widely between and within the different ecozones . A large proportion of the agricultural area in the Mixedwood Plains ecozone, especially in southwestern Ontario, is susceptible to soil quality change as it is predominantly in annual crops, with intensive row cropping . In the Prairies and Boreal Plains ecozones of western Canada, row cropping is insignificant, but approximately 3-7% of the farmland area has a high percentage (>30% of farmland) in summerfallow . Most of the summerfallow area occurs in the portion of the Prairies ecozone, where soil moisture is limiting for annual dryland crop production, and summerfallow has.traditionally been used as part of the crop rotation. At greatest potential risk are the areas identified as susceptible to soil quality degradation due to both biophysical conditions and land use and management practices ; these represent about 2-3% of the Prairies and Boreal Plains ecozones, and 7- 9% of the Mixedwood Plains ecozone. The soil quality assessment procedures developed in this project are defined generically so that they can be readily adapted to other quality functions or refined with more precise definitions to apply to specific crop types or more detailed map scales with more detailed land resource databases. ISQ procedures can be used in conjunction with process based models to evaluate sequential changes in soil quality as a result of model predictions of altered soil conditions and properties .
SOMMAIRE Dans le cadre du Programme national d'évaluation de la qualité des sols, on a mené une série de travaux de recherche sur la qualité des sols et de l'environnement au Canada. Les principales conclusions des recherches ont été résumées dans une série de rapports techniques et une publication, parue en 1995, intitulée La santé de nos sols. Le présent bulletin technique a trait au projet de démonstration d'un système d'évaluation de la qualité des sols à l'échelle régionale et nationale . Nous présentons un cadre conceptuel qui définit les différents aspects de la qualité des sols sous la forme d'évaluations statiques (qualité intrinsèque des sols), d'évaluations quasi dynamiques (sensibilité de la qualité des sols au changement) ou en fonction du changement réel de la qualité des sols . E n'est pas possible de mesurer le changement réel de la qualité des sols à l'échelle régionale et nationale, et les modèles actuels, basés sur des processus, ne sont pas adaptés à des opérations à grande échelle . En conséquence, l'étude a consisté principalement à élaborer des méthodes SIG pour l'évaluation de la qualité intrinsèque des sols et de la sensibilité de la qualité des sols au changement à l'échelle régionale et nationale. Le cadre conceptuel sert à indiquer dans les grandes lignes les étapes nécessaires d'une évaluation de la qualité des sols. Des cartes et des tableaux démontrent la capacité du système et présentent les résultats pour les grandes régions agricoles du Canada. Les résultats sont comparés, à un niveau très général, à d'autres mesures de l'utilisation et de la qualité des terres agricoles . II est également question de l'influence de différentes sources de données et échelles cartographiques sur les évaluations . La qualité des sols repose sur un ensemble de propriétés et de processus en interaction qui déterminent la capacité des sols à remplir un certain nombre de fonctions de base, comme la production végétale, l'effet tampon sur l'environnement des points de vue chimique et biologique et la séparation de l'eau et de gaz. Les paramètres de qualité des sols sont très variés et très complexes ; ils ont été simplifiés sous la forme de composantes ou d'éléments aux fins des évaluations . En ce qui touche la fonction culturale, quatre éléments de base de la qualité des sols ont été définis : i) porosité disponible, ii) rétention des substances nutritives, üi) conditions physiques d'enracinement et iv) conditions chimiques d'enracinement . Des attributs tirés de bases de données standard sur les ressources en terres ont été utilisés dans des algorithmes d'interprétation élaborés dans le cadre du projet pour évaluer chaque élément de la qualité des sols. On a utilisé une série d'indicateurs liés aux conditions des sols, aux paysages et aux pratiques d'aménagement et d'utilisation des terres pour évaluer la sensibilité de la qualité des sols au changement. Ces indicateurs sont les suivants : faible profondeur de la couche arable, faible teneur en carbone organique, forte inclinaison, textures superficielles très érodables, utilisation très intensive des terres et pratiques d'aménagement favorisant la dégradation des sols . L'identification
spatiale et la cartographie des indicateurs de sensibilité de la qualité des sols au changement s'appuyaient sur des méthodes SIG. On a fait la démonstration des méthodes d'évaluation de la qualité intrinsèque des sols et de la sensibilité de la qualité des sols au changement dans les grandes régions agricoles du Canada, soit dans les écozones des Prairies et des Plaines boréales dans l'Ouest et l'écozone des Plaines à forêts mixtes dans l'Est. Ces trois écozones renferment environ 91 % des terres agricoles et 95 % des terres cultivées du Canada. Selon les résultats généraux des évaluations de la qualité intrinsèque des sols, environ 37 % des terres des trois provinces des Prairies présentent les conditions climatiques et pédologiques nécessaires à la production de céréales semées au printemps . Dans l'est du pays, quelque 83 % des terres de l'écozone des Plaines à forêts mixtes, dans le sud de l'Ontario et du Québec, répondent aux conditions climatiques et pédologiques nécessaires à la production de céréales. L'évaluation de la qualité intrinsèque des sols révèle qu'une proportion relativement grande des sols agricoles des trois écozones entrent dans les catégories de qualité « Bonne» ou « Bonne à moyenne ». Dans les écozones des Prairies et des Plaines boréales, environ 70 à 80 % des terres sont classées « Bonne » ou « Bonne à moyenne » pour tous les éléments évalués. De 3 à 11 % des terres entrent dans la catégorie de qualité marginale « Mauvaise » pour ces mêmes éléments. Dans l'ensemble du Canada, environ 46 % des terres appartiennent à la catégorie « Bonne » ou « Bonne à moyenne », environ 34 % à la catégorie « Moyenne à mauvaise » et environ 19 % à la catégorie « Mauvaise ». Les estimations du bassin de terres à potentiel agricole par les méthodes d'évaluation de la qualité intrinsèque des sols dans les provinces des Prairies et de l'écozone des Plaines à forêts mixtes étaient proches des estimations propres aux quatre premières catégories possibiltés agricoles des sols de l'Inventaire des terres du Canada. Si l'on compare ces données aux données sur l'utilisation des terres du Recensement de l'agriculture de 1991, on constate que près de 85 % des terres à potentiel agricole sont cultivées actuellement . Selon l'évaluation de la sensibilité de la qualité des sols au changement, une petite mais néanmoins importante partie des terres à potentiel agricole (de 9 à 15 %) dans la plupart des provinces présentaient une forte inclinaison, alors que cette caractéristique était un facteur moins limitant dans les régions dominées par les sédiments marins et fluviaux de l'écozone des Plaines à forêts mixtes au Québec. Les terres de l'Ouest canadien étaient constituées de prairies aux sols à forte teneur en carbone organique dans les horizons de surface . Les terres de l'Est étaient couvertes de forêts et les sols avaient une teneur initiale en carbone organique moindre dans les horizons de surface . La sensibilité de la qualité des sols au changement attribuable à l'utilisation et à l'aménagement des terres (selon le Recensement de l'agriculture de 1991) varie grandement dans et entre les différentes écozones . Dans une grande proportion des zones agricoles de l'écozone des Plaines à forêts mixtes, surtout dans le sud-ouest de l'Ontario, la qualité des sols est sensible au changement vu que ceux-ci produisent en majeure partie des cultures annuelles et font l'objet d'une. culture sarclée intensive. Dans les écozones des Prairies et des Plaines boréales, ce type de
culture est quasi inexistant; néanmoins, un grand nombre d'agriculteurs (plus de 30 % des terres agricoles) pratiquent la culture sur jachères dans environ 3 à 7 % des zones agricoles . La plupart des terres mises en jachère se trouvent dans la partie de l'écozone des Prairies où la teneur en eau du sol limite les cultures sèches annuelles ; dans cette région, la rotation des cultures intègre la mise en jachère depuis longtemps . Les zones dont les sols sont susceptibles de se dégrader à cause des conditions biophysiques ainsi que de (utilisation et de l'aménagement des terres sont exposées à un plus grand risque; elles représentent de 2 à 3 % des terres des écozones des Prairies et des Plaines boréales et de 7 à 9 % de celles de l'écozone des Plaines à forêts mixtes . Comme les méthodes d'évaluation de la qualité des sols mises au point dans le cadre du projet sont générales, elles peuvent être adaptées à d'autres fonctions de qualité ou raffinées par l'ajout de définitions plus précises pour être appliquées à des types de culture particuliers ou à des échelles cartographiques plus fines avec des bases de données plus détaillées sur les ressources en terres. Les méthodes d'évaluation de la qualité intrinsèque des sols et de la sensibilité de la qualité des sols au changement peuvent être employées de concert avec des modèles basés sur des processus pour l'évaluation de changements successifs de la qualité des sols quand ces modèles prévoient l'altération des conditions et des propriétés de sols.
1.0 Introduction
In the past, experts who carried out broad regional and national assessments of soil/land quality (for example, the Canada Land Inventory) relied on their personal knowledge to compensate for gaps in data and inconsistencies between geographical areas. However, current assessments should be carried out in a documented, reproducible way with a minimum of subjectivity. The development of digital soil and land resource databases, such as the Canadian Soil Information System and National Soil DataBase (CanSIS/NSDB, http://res .agr.ca/ecorc/program3/cansis/ ), and the wide use of geographical information system (GIS) technology make it possible to develop procedures for such automated and reproducible assessments . The characterization of soil quality and degradation is a basic agri-environmental issue in Canada as it is in many other nations . It has been the focus of much recent attention and research efforts (Conte et al ., 1982; Nowland, 1987; Nowland and Halstead, 1986 ; Science Council of Canada, 1986; Mathur and Wang, 1991 ; Acton 1991a) . During 1989-1993, the National Soil Quality Evaluation Program (SQEP) of Canada supported and coordinated a series of research activities on soil and associated environmental quality. One of these research studies, the development of a GIS-based system to facilitate improved regional and national assessment of agricultural soil quality in the major crop production regions of Canada, is the subject of this report. The main objectives of the study are: to translate the current understanding of soil quality into a conceptual framework for broad-scale assessment, to develop procedures within GIS environment and to demonstrate the capability using digital land resource databases and other related data, and to evaluate the sensitivity of the procedures to different data sets and to the various scales of mapping . Components ofthe rationale and major results of this study have been documented and reported in previous publications (MacDonald et al., 1991 ; 1992; 1993 ; 1994 ; and 1995). These reports also reflect the course of development and improvement of both methodology and databases. There was a need, however, for a comprehensive and detailed document to summarize the current `state of the art' with respect to the major research methods and findings . This Technical Bulletin describes the soil quality concepts and the GIS-based procedures for soil quality assessment at regional and national scales using existing land resource databases .
2 .0 Conceptual Framework The conceptual framework for assessing and monitoring soil quality in Canada was extensively discussed during the implementation of the Soil Quality Evaluation Program and has been documneted by Acton and Padbury (1994). In this report, a brief overview of the historical development of the term soil quality is provided (Section 2.1). This is followed by a summary of the soil quality terms specifically used and adapted by this study, including; a hierarchical framework for soil quality (Section 2.2), working definitions of several key concepts, such as inherent soil quality (ISQ, soil quality susceptibility (SQS) and soil quality change (SQC) (Section 2.3), and the basic steps of soil quality assessments (Section 2.4). Finally, the approaches taken by this study are outlined in Section 2 .5 . 2.1 Current Understanding on Soil Quality : An Overview 2.1.1 Shift and expansion of the context of soil quality Soil quality is a term that has undergone a significant shift and expansion in recent years. Historically, the concept of soil quality was defined mainly in terms of suitability for crop production or other agricultural land uses (Bouma 1989a; Bouma 1989b ; van Diepen et al., 1991). Recent discussions and reviews (Acton, 1991 a, 1991b, 1992 ; Larson and Pierce, 1991 ; Rodale Institute, 1991 ; Acton et al., 1992 ;1994 and 1995; Doran, et al., 1994; Acton and Gregorich, 1995; Warkentin, 1995; Doran, et al., 1996 ; Bouma, 1997 ; Karlen et al., 1997; Sims et al., 1997 ; Lal, 1998 ) have indicated that there is a significant shift and expansion in the context of interpretations to be included as components of soil quality . The major change in the scope of soil quality is the inclusion of functions which are not directly related to productivity . For example, Karlen et al. ( 1996) has defined soil quality as "the capacity of a specific kind of soil to function, within natural or managed ecosystem boundaries, to sustain plant, animal productivity, maintain or enhance water and air quality, and support human health and habitation." Lal (1998) suggests an even more expansive definition which includes the following four principal soil functions ; Sustain biomass production and biodiversity including preservation and enhancement of gene pool, Regulate water and air quality by filtering, buffering, detoxification, and regulating geochemical cycles, Preserve archeological, geological and astronomical records and Support socioeconomic structure, cultural and aesthetic values and provide engineering foundation . The expanded definition of soil quality extends its range of influence from a direct impact on the quantity of food and fibre produced to considerations of environmental quality and food quality .
In addition, the environmental filtering function is directly concerned with the cycling of toxic elements, biological entities and heavy metals in the environment and, indirectly, with the effects on human and animal health (Rodale Institute, 1991 ; Hortensius and Nortcliff, 1991) . The water partitioning function is directly related to the quantity of surface and groundwater availably. In its expanded form soil quality becomes an important component of a holistic assessment of environmental quality and an important environmental indicator to monitor the effects of land and water management . Soil quality changes occur as a result of environmental processes which vary greatly over space and within and between ecosystem boundaries (Larson and Pierce, 1991) . Crop production has been emphasized as the primary soil quality function, especially in the context of agricultural land uses (Arnold, et al, 1990 ; MacDonald and Moore, 1991 ; Pettapiece, 1995). Acton and Gregorich (1995) recently defined soil quality/health in this manner, namely soil quality for agriculture is the soil'sfitness to support crop growth without resulting in soil degradation or otherwise harming the environment . Consideration of the various soil functions 'on an equal basis' has been suggested by the Rodale Institute (1991) . In reality, the major functions are closely interrelated, and occur concurrently in any farming system at any management level, whether low (natural or undisturbed) or Viability highly manipulated (or managed) . For e' ~i il,i't , l ° ;R`.~ example, land area which is used for intensive agricultural production may Sod Modification Processes also be required to serve as an environmental buffer to retain nutrients from manure in the rooting zone for crop uptake and also to partition precipitation into soil storage rather than allowing surface runoff and contamination of adjacent water with sediment . The combination of nutrient, crops and water creates conditions in the soil atmosphere which can be controlled to promote or inhibit Temporal changes 10 processes such as denitrification -10 (MacDonald et al., 1994). Figure 2-1. A multi-dimensional perspective on soil quality Figure 2-1 illustrates the broad context of a multi-dimensional perspective on soil quality (Figure 2-1) . 2.1.2 Adoption of `soil health' as a synonymous term of soil quality In the past decade, the term `health' has been used extensively in environmental applications as a metaphor drawn from human health, even though there is not always a parallel between medical
health and environmental issues (Ayala, 1987 ; Schaeffer et al., 1988; Rapport 1992, Haskell et al., 1992) . Terms such as `ecosystem health' and `environmental health' are often cited in the literature. Recently the term `health' is also adopted by some soil scientists (Haberm 1991, Acton et al 1995) as an equivalent term of `quality' . In general, human health can be defined in a negative manner as the absence of disease or in positive manner as the resilient and robust characteristics of a healthy human body . The term `health', as used in soil quality assessment, can also be viewed in negative or positive manner. Soil health is a composite picture of the state of the soil's many physical, chemical, and biological properties and of the processes that interact to determine this quality or health (Acton and Gregorich 1995) . In the broad, holistic context of ecosystems, soil health is an indicator of environmental or ecosystem health (Figure 2-2).
Environmental Sustainability
Figure 2-2. Soil health/quality as an indicator of environmental/ecosystem health
We agree with Acton and Gregorich (1995) that soil quality and soil health can be used interchangeably, however ; for consistency, soil quality is the term used in this report . 2.2 A Hierarchical Framework of Soil Quality Assessment Soil quality approaches have to be developed to operate at a variety of scales and associated levels of data availability (Halvorson et al., 1997 ; Karlen et al ., 1997). At the farm level, Romig et al. (1995;1997) have demonstrated a descriptive qualitative assessment using a soil health scorecard. For assessments at localized scales (e.g. field or farm) it is frequently possible to collect site-specific information . However, as the area to be assessed increases, constraints of
time and money usually restrict the information available to data which can be extracted from readily available standard databases . At broad regional and national scales, generalized soil and land resource databases have been used to characterize the wide range of spatial variation of soil/land quality from one area to another (Thomasson and Jones, 19898; van Diepen et al., 1991) .
SOIL QUALITY FUYCTIINS
I"
I
rRIDUCTt "11
: w
t
Databases atm
Soil
London 8 Management
u"CiMr
Figure 2-3. A hierarchical framework for soil quality assessment
The relationship between standard land resource databases and soil quality assessments has been structured hierarchically for this project as shown in Figure 2-3. At the highest level, soil quality is shown as a composite of several soil functions . Most soil functions are too complex to be estimated directly but can be defined by a series of somewhat arbitrary elements which contribute to the function. The framework consists of increasingly specific levels ; functions, elements, attributes and databases. The combination of attributes selected depends, to a large extent, on the availability of data. In general terms, it is possible to define the elements which contribute to a soil quality function. The same elements are estimated regardless of the level of assessment but the kinds and quality of data used for the estimates will vary with the application and the data which are available. For effective data integration in a GIS environment, attributes from a variety of data sources can be extracted and combined into elements of the soil function of interest . The integration can be ; 1) additive, multiplicative or some other arithmetic combination; 2) flexible enough for input additional data and adjustable weighting and 3) implemented by using specific algorithms and GIS operations. The more precisely the soil quality function is defined, the more specific the elements and attribute limits can also be defined to make a more detailed assessment. For example, soil quality for crop production can be defined for a certain crop and management, such as spring seeded wheat under dryland agriculture. This permits the soil quality elements and attribute limits to be specifically tailored to the needs of the particular crop. 2 .3 Concepts of Inherent Soil Quality (ISQ) and Soil Quality Susceptibility (SQS) Soil Quality is actually a series of related concepts; some deal with current conditions, and others with soil quality change over time. For the purposes of this study, it was necessary to define these terms more explicitly . 2.3.1 Inherent soil quality (ISQ) and ISQ elements Some soil properties are mainly determined by naturally-controlled factors and processes, such as parent materials and chemical and physical weathering, They are relatively stable in short and medium time scales. In order to distinguish these properties from the more dynamic aspects of soil quality which are considered within the context of susceptibility and change, the concept of inherent soil quality was developed. Inherent soil quality (ISQ) is defined as those properties ofthe soil which contribute to the capacity ofthe soil to support a specific criticalfunction (such as crop production) and which are relatively unchanging through time. Inherent soil quality is defined by in situ soil and landscape properties such as; slope, soil horizon
thickness, texture, and soil organic matter. The inherent soil quality of any particular site can change over longer time periods, due to soil formation, or due to soil degradation by wind, water, compaction, or other factors . Inherent soil quality could be further characterized by elements, which are key components contributing to a specific soil function (Figure 2-3). For the crop production function, four key elements were defined (Table 2-1). A severe limitation in any one element would be detrimental to the overall soil quality function, and would result in a lower ISQ rating. Table 2-1. Inherent soil quality elements for crop production ISQ Element
Description
Available Porosity
The capacity ofthe soil to retain and supply moisture to the crop, and also its ability to provide sufficient aeration for healthy root development.
Nutrient Retention
The capacity ofthe soil to retain plant available nutrients and release them as required by the plants
Physical Rooting Conditions
The quantity or volume of available soil material that is physically suitable for root development.
Chemical Rooting Conditions
The quantity or volume of available soil that is chemically suitable for root development. This is based on the absence ofexcessive or noxious chemical conditions such as salts, pH extremes, heavy metals, pesticides which inhibit the growth of crops or degrade the safety of the produce
Normally, data about soil or landscape properties and related environmental features are collected and recorded as attributes (Figure 2-3) and are stored in standard land resource data bases. The major challenge faced with this project was to define inherent soil quality, in terms of ISQ elements, using the available soil attributes and databases. Sections 3 and 4 of this report describe attributes selected, and how they were integrated into the rating algorithms, classes, and threshold values for each ISQ element. The ISQ elements for other functions of soil, such as the environmental buffer and water partitioning may be defined in a similar fashion but would have a different set of ISQ elements . For example, the water partitioning function would involve ISQ elements for the capacity of surface water recharge for certain soil-landscape types and the water transmission capacity below the rooting zone of the soil . 2.3.2 Soil quality change (SQC) and soil quality susceptibility (SQS) Ideally, the procedure for assessment of soil quality should precise enough to determine soil quality change (SQC). At localized scales it is possible to achieve this objective, however, at
broad regional and national scales, the level of resolution (both spatial and temporal) in standard databases is not adequate for change assessment . Soil Quality Change is a dynamic concept . In a broad sense, soil quality change may include both change over space, i .e . between different map areas, and changes through time, i.e . temporal change. In this report, soil quality change is defined in terms of temporal changes, as the alteration of overall soil quality, or certain soil quality properties, between two points in time (t2 - t,) . Temporal changes in soil quality may be regarded as either random, regular-periodical, or trend changes (Varallyay et al., 1990) . It is difficult, but necessary, to distinguish the natural random or cyclical variability from the trend changes in soil quality, and to define the threshold or limiting range of change beyond which the results will be regarded as significant or Nahnal Agents Htnir&i Agents cause for concern (Lal et al. 1989). The (Physical, chemical (Land use and processes responsible for soil modification mid biological) nuuragenmit piacUces) and soil quality change include basic physical, chemical and biological processes (Lal et al, 1989 ; Varallyay et al, 1990). The rates of these processes are determined by the land use and management practices (Figure 2-4). Changes in soil quality may be Compractio~a >g positive (agradation), negative " Fa~obioii (w , wuid and age) (degradation), or neutral (fluctuation) . Land use and management practices can affect " Acidilication both the direction and rates of change of soil quality (Figure 2-4 and 2-5). " ToAficadon (he%tvy iilet:als) '
Matter Content
ofOrgank
Figure 2-4. Soil Quality change in relation to soil modifying processes and land use and management practices
Several different approaches can be taken to assess soil quality change . It can be measured directly by research at specific monitoring sites over long time periods . Since most soil quality attribute changes are subtle and long term, very precise and repetitive monitoring of specific plots is required. While vital for scientific understanding of soil quality change, it is costly, and only feasible for limited numbers of soil and landscape conditions .
Another option to measure soil quality change is to use data from historical sampling sites, such as ground truth information collected during detailed soil surveys. This option was tested in Manitoba, using a network of soil inventory sites that were re-sampled to estimate soil quality change . No definitive trends in soil quality change were noted. This was attributed to the lack in precision in the original recording of soil horizon attributes and site location, and the relatively
short (10 year) time interval . Soil inventory data are only available in limited areas . In general, the location and horizon attribute information have not been recorded with sufficient precision for comparative re-sampling . For a more general approach, process-based models can be used in conjunction with field measurements to separate trends in soil quality from random and cyclical changes . These models
" Monitoring " Modelling " Inference/indication
Site speck measurements Site specific and general measurements General measurements
Detection of change Quantification of trends and prediction of change Indication of soil quality susceptibility to change
Figure 2-5. Kinds and direction of soil quality changes and research approaches normally require large quantities of site specific information and site calibration. Models can also be used to predict future changes in soil quality based on scenarios of climate, land use and management. In future, generalized models may be able to use soil map databases, and to predict changes in soil attributes in each landscape unit over time and to estimate how these change affect soil quality. Such models are not yet available for Canadian conditions, although a number of models are under development (Environmental Indicator Working Group, 1994). As Hamblin (1991) points out, all research methodologies are limited either by constraints of space, in that they are too specific to be extrapolated to whole regions, or by constraints of time, being measured over too short a period to be predictive of any long-term trend. Therefore, it is very difficult to incorporate and extrapolate the limited existing data of measured soil quality change into regional or national assessments. In this study, an alternative approach was developed to address soil quality change at broad scales . It consists of identifying and assessing the relative `susceptibility' of soil quality to change
using existing databases and GIS tools. Soil Quality Susceptibility (SQS) to change is defined as an estimate ofthe ease or likelihood that a soil modifying process will change some basic land resource attributes and result in a net change in soil quality. There are two major aspects of SQS: (i) biophysical (intrinsic) characterized by soil and landscape conditions and (ii) land use and management (extrinsic) characterized by land use and management practices, past, current and proposed (Table 2-2). Table 2-2. Aspects of soil quality susceptibility to change. Aspects
Description
Biophysical Susceptibility
Soil and landscape conditions which make the soil more or less susceptable to processes which modify the quality of the soil for the critical function of interest. For the crop production function, this can include such factors as slope, silt content, and soil structural stability.
Land Use and Management Susceptibility
Past, current and proposed land use and management practices which make the soil more or less susceptible to processes which modify of the soil for the critical function of interest . For the crop production function, this can include such factors as crop type, rotation, and tillage .
Although SQS does not provide estimates of actual soil quality change, soil-landscapes with different SQS classes can be mapped to show areas with varying degrees of susceptibility to soil quality change . More detailed assessment of SQS will identify where programs to change land use and management would be most beneficial. SQS is normally considered under current climatic and land use and management conditions, however, it can also be estimated under a variety of alternative climatic and land use and management conditions, over particular time periods. This can be used to study the implications of such land use and management alternatives on long term soil quality change and agricultural sustainability. 2.4 Basic Procedures of Soil Quality Assessment Soil Quality assessment can be consdered as a sequence of steps, as outlined in Figure 2-6. These can be applied to areas as small as a field site, or very broad areas as large as a country . Historically, soil quality assessments have been done intuitively by local farmers or agronomists . Soils were considered as suitable or unsuitable for crop production, based on long term yields, or performance of similar soils elsewhere. Risks of declining soil quality from water erosion, wind erosion, and other factors were also considered, and influenced certain land use and crop management options . Past experience and perceptions of soil quality change, or the lack of it, was often a major influence in such decision making. Long term sustainability of lands for crop production was largely dependent on the insight and commitment to long term stewardship by 10
individual land owners .
MOO
Step 1 : Estimate the inherent soil quality for one or tnore specific soil functions, using soil and land resource information. Assess the biophysical conditions causing soil quality to be susceptible to change, using soil and landscape information Assess the hcumn-imposed conditions causing soil quality to be susceptible to change using land use and management information. Step 4: Combine the procedures of ISQ and SQS assessment over time to predict change in soil quality : a) by estimation; b) through monitoring and modeling Step 5: Re-assess soil quality at some time in future using land-resource data. Figure 2-6. Basic steps of soil quality assessment
2.5 Approaches Specific to This Study As stated in the introduction, the overall objective of this study is to develop an operational GISbased system and a set of procedures which allow an improved assessment of agricultural soil quality for the major crop production regions of Canada . To achieve the objective, we have stratified the nature and approaches of this study as follows : It is mainly a macro- or broad-scale assessment at regional and national level using (approx . 1 :1 million to 1 :5 million scale) . More detailed assessments (approx. 1 :50,000 map scale) are included to indicate how methodologies can be extended to more detailed scales, and for validation and sensitivity analysis of the broad scale results . It is a spatially oriented assessment and analysis. The spatial variation and patterns of inherent soil quality (ISQ) and Soil Quality Susceptability to potential change (SQS) across or within a region are analysed and mapped for intra- or inter-regional comparisons .
The assessment and analysis are implemented with a set of GIS procedures using currently available digital data bases . The use of an automated GIS-based approach allows for fully reproducible results, to facilitate comparison of the assessments at different time intervals and with updated land resource information . For the purposes of this study, soil quality was assessed for the crop production function . The logic and algorithms for assessing inherent soil quality (ISQ) were developed for the generic requirements of annual cereal crop production . Cereals are a major crop type grown in all major agricultural areas of Canada, and are appropriate for broad level comparisons . Actual soil quality change (SQQ cannot be estimated and mapped directly for large land areas, as measured data about soil modification and change is only available for a few, selected monitoring sites. Mapping of predicted soil quality change, under various climatic and land management scenarios is possible in the future . This will require integration of calibrated, process based soil degradation models with GIS databases, and soil quality assessment techniques developed in this study. The most appropriate substitute, at the present time, are maps indicating soil quality susceptibility to change (SQS) for different regions, under specified land use and management practices .
3.0 Methods 3.1 GIS Systems The system and GIS procedures were developed using Arc/Info' GIS software from the Environmental Systems Research Institute, Redlands, California, USA and PAMAP GIS software from PCI Pacific GeoSolutions Inc., Victoria, Canada. The main procedures have been tested on various computer platforms, such as VAX mainframe, VAX and DEC Alpha UNIX workstations, and PC Windows environment . Some ISQ rating algorithms were developed and implemented in dBASE IV and linked with the PAMAP GIS . 3.2 Data Sources Based on the soil quality conceptual framework documented in Chapter 2, the minimum data sets required for assessing inherent soil quality and soil quality susceptibility were identified. They are: - Soil and land resource data - Topographic data - Climatic data - Land use and management data. In general, soil and land resource data are regarded as prerequisite or "first order' data sets, as they describe the fundamental soil properties required to characterize inherent soil quality for crop production . The additional data sets are required to estimate the susceptibility to change of soil quality, or to make actual predictions of soil quality change . In Canada, reasonable amount of soil and related data are available at regional and national levels in digital format. Table 3 summarizes the main features of some of these data sets and indicates how they can be used for soil quality assessment. Soil, land resource data and some climate and landscape (topographic) data were obtained from the Canadian Soil Information System and National Soil Data Base (CanSIS/NSDB, http://res .agr.ca/ecorc/Program3/cansis/) of Agriculture and Agri-Food Canada; Census of Agriculture data were purchased from Statistics Canada, and the AVHRR land cover data were purchased from the Manitoba Centre for Remote Sensing and vectorized by GIS Division, Natural Resources Canada. They were all either available or converted to standard GIS formats . The reliability assessment in Table 3-1 is subjective based on the documentation attached to each data set and our data checking and sensitivity analysis. The quality of the attribute information was quite variable, particularly for the land resource layer which was compiled from a variety of data sources ranging from expert estimates to summaries from detailed soil surveys. 1 The mention of a trademark, proprietary product or vendor does not imply endorsement by Agriculture and Agri-Food Canada to the exclusion ofother products or vendors . 13
Table 3-1 . Data sources for broad-scale assessment of soil quality in Canada (used in this study) Category
Data
Soil and land resource
Soil Landscapes of Canada (SLC)* map with 40 attributes for dominant (DOM) and subdominant (SUB) soils in each SLC polygon (DOM/SUB files are also called SLC extended legend)
CanSIS/NSDB, Agriculture and Agri- Food Canada (AFFC)
Soil Carbon Data Base linked to SLC polygons with component (CMP) and layer (LYR) files containing more than 20 attributes
CanSIS/NSDB, AFFC.
Soil Map Unit File (SMUF) in Detail Soil Map (DSM) database
CanSIS/NSDB, AFFC.
Soil Name File (SNF) and Soil Layer File (SLF) with more detail soil attributes. Data from these files can be related to map polygons. Topographic
Source
Scale & Availability
Utility in This Study
low to medium (variable from area to area)
ISQ rating( ISQ92) procedures
medium
SQS indicators and ISQ94 procedures linking SLC map components to soil name and layer attributes (SNF and SLF).
Approx. 1 :20,000 50,000 Limited areas
medium to high
ISQ94 procedures linking detail soil map units to soil name and layer attributes
CanSIS/NSDB, AFFC.
Pedon-scale variable from province to province
medium to high
ISQ94 rating procedures
Landscape shape and slope attributes from SLC DOM, SUB and CMP files
CanSIS/NSDB, AFFC.
1:1 million All of Canada
low to medium
SQS indicator
Climate
EGDD and P-PE (1951-80 climatic normals), re-compiled to SLC polygons
Agronomic Interpretation Working Group, AFFC
1 :1 million All ofCanada
low to medium
Defining area suitable for annual crop production and adjusting ISQ porosity rating
Land use and management
Census of Agriculture (CoA) 1981 and 1991 recompiled to SLC polygons and containing common land use and management attributes.
Statistics Canada and AFFC (joint recompilation)
1 :1 million
medium
Agricultural areas
SQS indicator and estimation of past changes in land use and management
Vectorized AVHRR 1989 composite containing broad land cover classification
Natural Resources Canada
1 :1 million (approx .) All area of Canada
low to medium
Defining the extent ofagricultural areas and estimating area of agri-crop land
* For a complete list of acronyms used in this report, please see appendix 1
14
1 :1 million
Reliability
All land areas of Southern Canada 1 :1 million All of Canada
3.3 Spatial Framework Along with decisions regarding the attribute data to be used in this study, it was necessary to choose appropriate spatial units for the assessment, analysis, and map displays. The spatial framework acts as a set of `spatial folders' of data or information . The hierarchy of broad scale Canadian land resource data used in this report and its organization is briefly summarized in this section . Figure 3-1 illustrates the hierarchical nestings between SLC and the National Ecological Framework of Canada.
DETAIL SOIL MAP (DSM) DATA ( 1 :20,000 - 125,000 scale) (Map units NOT nested with SLC polygons)
i-,
J
SOIL-L ANDSCAPE (SLCI UNITS ( 1 :1 million scale) 22 SLC polygons in the Ecodistrict #565 ECODISTRICTS Approx . 1 :5 million scale)
Ecodistrict #565 as one of 12 Ecodistricts in the Lake Erie Lowland Ecoregion
LEVEL FOR: " Testing ISQ algorithms " Validation of results of SLC level ISQ ratings and SQS indication Primary LEVEL OF : " Data collection and integration " ISQ rating and mapping " Identification and mapping of ISQ indicators LEVEL OF : " Generalized ISQ analysis & mapping " Analysis and reporting of soil related environmental and policy issues.
Lake Erie Lowland as one of 4 Ecoregions in,the Mixedwood Plain Ecozone
ECOZONES
Mixedwood Plain as one of 15 Canadian Terrestrial Ecozones Figure 3-1. Illustration of the spatial framework for soil quality assessment and reporting in Canada
The Soil Landscapes of Canada (SLC) digital maps were selected as the appropriate spatial framework for land-related assessments and environmental reporting at regional and national scales . SLC maps are part of the Canadian Soil Information System and National Soil Data Base 15
(CanSIS/NSDB, http://res .agr .ca/ecorc/program3/cansis/) of Agriculture and Agri-Food Canada. SLC polygons are compiled at a scale of 1 :1 million by regional land resource experts, from a variety of more detailed historical data sources. SLC polygons also constitute the basic spatial units in the National Ecological Framework for Canada (Ecological Stratification Working Group, 1996). Ecological units, such as ecodistricts, ecoregions, and ecozones, are all "nested" to the Soil Landscapes of Canada polygonal coverage . This ecological framework permits further aggregation and generalization of soil quality assessments at even broader scales. Ecodistricts, typically consisting of several SLC polygons, were selected as the appropriate units for ISQ reporting at national and broad regional scales (approx . 1 :5 million). Detailed digital soil maps (DSM) are also available for selected areas of Canada from the National Soil Database of AAFC. These maps are linked to a standard set of soil attribute files (MacDonald and Valentine, 1992), permitting the development and testing of soil quality algorithms at the scale of detailed soil maps (approximately 1:20,000 to 1 :125, 000 ). Selected detail map data sets were used in this project, for development and testing of inherent soil quality rating algorithms for application at both detailed and broad scales . They were also used for validation and sensitivity analysis of broad level soil quality assessments in selected areas, as described in Section 5.1 .2. 3.4 ISQ Rating Procedures 3.4.1 Evolution of ISQ Procedures During the course of this study, two separate sets of inherent soil quality (ISQ) rating procedures were developed for application at regional and national levels . Both of them were based on the framework illustrated in Figure 2-3, Chapter 2. The same four elements of soil quality (Table 2-1) were defined under the crop production function of soil quality, as well as an overall soil quality rating for each soil polygon component. A generic cereal crop was selected in both cases, as this represents a major crop type grown in all crop production areas of Canada. The differences between the two ISQ procedures which resulted from different available soil databases were in the attributes, and algorithms used to define the soil quality elements. The first set of procedures, developed in 1992 (referred to as "ISQ92"), uses attributes contained in the Soil Landscapes of Canada extended legend as described in Table 3-1 and Table A2-1 of Appendix 2. The extended legend database has generalized soil attributes for dominant and subdominant soil landscapes in each SLC polygon (SLC DOM and SUBDOM attribute files) . The SLC extended legend properties were defined by local soil experts in each region, in terms of a few broad classes for each attribute. ISQ92 can be run for all agricultural areas of Canada, as the SLC extended legend databases are available for all of southern Canada (SLC v1 .0) . The SLC extended legend database has not be updated with further versions of the SLC map (currently v2.2). The ISQ92 rating program (Appendix 3) is relatively simple but, nevertheless, should be appropriate for regional or national level assessments .
16
The second, more detailed set of ISQ procedures was developed in 1994 ("ISQ94"), to take advantage of improved SLC file structures . The general structure of data describing SLC polygons had undergone a major revision such that each polygon was described by a series of components which occupied a specified proportion of the polygon area. Each soil component could be linked to two additional files, the Soil Names File (SNF) and Soil Layer File (SLF), which contain modal values of physical and chemical properties for each soil . The SLC Component file data structure was quite similar in concept to the Soil Map Unit File (SMUF) structure used to delineate soil map components and extents for detailed Canadian soil map polygons (at scales of 1 :20,000 to 1 :125,000). ISQ94 procedures can be applied to both SLC and detailed scale digital maps, linked to the same provincial SNF and SLF soil attribute files . ISQ94 procedures can also be easily modified to suit additional crops, or be used to perform sensitivity analysis of various attribute limits. ISQ94 requires linkage to detailed SNF and SLF databases, which are not currently available or fully correlated for all agricultural regions of Canada. Also, there are some concerns about the representation of broad, SLC level polygons by a limited number of soil types in the current SLC Component file. In many instances, SLF attribute values for individual soils are based on estimates, or only a few samples from detailed soil sampling sites in various locations . Refinements of the SLC CMP, and SNF and SLF files will increase the applicability of the ISQ94 procedures in the future . 3.4.2 Components of ISQ Procedures The ISQ procedures can be subdivided into stages (Figure 3-2) consisting of pre-rating (screening) procedures, rating procedures and post-rating or presentation procedures . 3.4.3 Pre-Rating Procedures The main purpose of the pre-rating procedures is to ensure that all GIS databases and essential rte_ rso rl ri J ~r~r~rliir=~
:1f_fl1
('JJ" ~ f ,
" Data checking " Selection of files and attributes " Exclusion " Preparation of interim items and files
llf' ~
" Available porosity " Nutrient retention " Physical rooting conditions " Chemical rooting conditions " ISQ overall rating
PJfiJ -(' J~~~fI 11 l'iJ
" Generalization for mapping at certain scale GIS query and display " Reporting
Figure 3-2. The organization of ISQ procedures
17
data fields required for the ISQ94 rating procedures are present, and that all of the mandatory data fields have acceptable attribute values . Once the ISQ crop type is selected, the threshold values for the various soil and climatic attributes can be reviewed or altered. Interim files are generated to store the output ratings and statistics . Soil map components with inappropriate data values for rating calculations are assigned "data error" codes and are excluded from further analysis . Details of these steps are presented in Appendix 4. 3.4.4 Rating Procedures Once pre-rating procedures are completed, calculation of ISQ ratings proceeds for each soil polygon component. An initial set of threshold criteria are used to identify climatic, landform, or soil conditions that are considered unsuitable for crop production . SLC polygon components that lack sufficient heat for a minimal growing season (Effective Growing Degree Days less than 1050), or are too steep (over 30% slope), or have non mineral soil types (organic soils, frozen soils, bedrock, or other non soils) are rated as "Unsuitable" for cereal crop production, and are excluded from further analysis . The soil polygon components for areas that meet minimum threshold criteria for crop production are then evaluated using the algorithms developed for the four elements of soil quality. The ISQ94 rating procedures consist of four basic sub-procedures and one overall rating sub-procedure, each of the four basic sub-procedures corresponding to one of the four elements of inherent soil quality (Table 2-1) . The general approaches and rating considerations are described in the following section. Further details of the ISQ94 rating program and procedures are provided in Appendix 4. 1) Available porosity element This procedure evaluates the capacity of the soil component to retain and supply moisture to the crop, and also its ability to provide sufficient aeration for healthy root development. The two components, aeration porosity and moisture holding porositv, are calculated and rated separately. The most limiting of the two conditions determines the overall ISQ available porosity element rating . Aeration Porosity is calculated for each soil horizon and the accumulated value is the total aeration porosity value (in cm of air). This is calculated for the surface layer (20 cm) and for the entire rooting zone (an accumulated value, "AIR TOT", for all horizons down to the maximum allowable rooting depth) . The formula used is: n
AIR- TOT _ Y, THICKNESS,*[(PDT - BD,) / PD, - 0 .01* KP33i]
where, 18
AIR-TOT = accumulated air-filled porosity (cm) from top to the maximum allowable rooting depth (horizon i = 1, .. ., n) . THICKNESS = the thickness of the soil layer i (cm). PD = Particle Density, g/cm3. (2.65 g/m3 for mineral soil layers) BD = Bulk Density, g/cm3. KP33 = % water, by volume, that is retained by the soil at 1/3 atmosphere suction. This is multiplied by 0.01 to convert from % to a ratio of the soil volume. KP33 approximates field capacity. The soil aeration value (AIR TOT, expressed in cm of air), is also calculated for the surface 20 cm by adding soil layers or portions of layers occurring within this zone. A minimum threshold value of I cm of air within the top 20 cm (5% of the surface layer volume) must be reached; if this does not occur, the soil is rated as unsuitable, and excluded from further ISQ element calculations . Values exceeding the minimum are adjusted for the moisture regime to give an aeration porosity rating value ranging from "Good" (0) to "Poor" (3). The moisture regime adjustment is based on precipitation, potential evapotranspiration, soil taxonomy, and drainage (details in Table A4-5 of Appendix 4). Moisture Holding_ Porosity is determined from an estimate of the available water holding capacity (AWHC). The AWHC is based on soil texture and ranges from 40 to 200 mm/m. It is estimated for each soil and summed over the crop rooting depth. The AWHC is modified by the moisture regime to derive a moisture holding porosity rating with a similar range of values to the aeration porosity. Low AWHC values are considered a more serious limitation to crop growth in more and climatic regimes, such as the Brown Chernozemic soil zone. See Table A4-7 of Appendix 4 for details. The overall Available Porosity element rating is determined as the `most limiting' of the two component ratings . 2) Nutrient retention element This procedure evaluates the capacity of the soil to retain plant available nutrients and release them as required for crop production . This is assessed from the cumulative cation exchange capacity (CEC) of all soil horizons within the surface rooting depth (top 20 cm), using the formula:
NUTR_ SUR =
n
CECi*BD,* BASESi*0.01* ELTHICKNESSi
where, 19
(3-2)
NUTR SUR = Accumulated nutrient retention capacity of surface 20 cm (meq/cm3) of all soil horizons or portions of horizon (i = l, . . ., n) CEC = cation exchange capacity (meq/100g) BD = Bulk Density (g/cm3) BASES = base saturation (%), This is multiplied by 0.01 to convert from % to a ratio of total CEC . ELTHICKNESS = the eligible thickness of the soil layer i considered as part of top 20 cm surface soil. If a soil horizon has a lower depth that exceeds 20 cm, only the portion to a depth of 20 cm is considered eligible . The average CEC of the top 20 on a unit volume basis is calculated as: NUTR_RATIO (meglCM 3) = NUTR SUR (meglCm2)/20 (cm)
(3-3)
A nutrient retention rating is assigned based on classes of average CEC by volume of the top 20 cm (NUTR_RATIO) . The generic rating scale, for cereal crops is shown in Table 3-2 . Table 3-2. Rating scale of ISQ nutrient retention element NUTR_RATIO of Top 20 cm (megtcm3)
< 8
8.0 to 8.9
9.0 to 15 .9
ISQNUTR 9 3 2 Ratings 0 = good; 1 = good to moderate; 2 = moderate to poor; 3 _ poor; 9 = not rated
16 to 22
> 22
1
0
3) Physical rooting conditions element This ISQ element evaluates the volume of soil material that is physically suitable for root development. For each soil component, the successive soil layers are evaluated, starting from the surface, until a physical root restriction or the maximum crop rooting depth is encountered. Rating classes are assigned based on ranges of the eligible rooting depth (THICK-TOT), as shown in Table 3-3. Table 3-3. Rating scale of ISQ physical rooting conditions element THICK TOT (cm)
< 20
20 to 29
30 to 54
55 to 79
>= 80
ISQ_ROOT Rating
9
3
2
1
0
0 = Gond_ 1=('*nnd to Mnrjerate- 7 = MnrleratP to Pnnr-'I = Pnnr- Q -_ not rntarl
4) Chemical rooting conditions element 20
The chemical rooting conditions procedure evaluates the volume of available soil that is chemically suitable for root development . For this report two natural chemical conditions, pH and soil salinity (measured as Electrical Conductivity, or EC) are each evaluated separately. The same approach could be used for anthropogenic chemical conditions such as pesticide or petroleum contamination. For each condition, both surface (0 to 20 cm) and subsurface (20 to 80 cm) depths are averaged and evaluated separately, using threshold tolerance values . Both conditions can vary considerably with depth, and surface conditions are considered more limiting to crop production . The final rating for each chemical rooting condition is determined by combining surface and subsurface ratings . Both pH and salinity are assessed in a similar fashion, and the most restricting of the two conditions is used in the overall ISQ chemical rooting element rating . In general terms, an average (depth weighted) pH and EC of the top 20 cm of the soil (SURPH and SUREC, respectively), is calculated for each soil component, based on Soil Layer File data; n
SURPH=,E PHCA, * ELTHICKNESSil20)
i=i n
SUREC=1:ECi * ELTHICKNESSiI20)
(3-5)
i=i
where,
(3-4)
SURPH = depth weighted average pH of surface 20 cm of all eligible soil
(i = 1, ..., n) PHCA = soil pH in 0.01M calcium chloride SUREC = depth weighted average EC of surface 20 cm (mS/cm) of all eligible soil horizons (i = 1, ..., n) horizons
EC = electrical conductivity (mS/cm) ELTHICKNESS = the eligible thickness of the soil layer i, considered as part of the top 20cm of soil
The surface chemical rooting condition (Table 3-4), ISQ_SURCHEM, is determined based the most restricting value of classes SURPH and SUREC values. Table 3-4. Rating scale of ISQ chemical rooting conditions element SURPH or SURPH SUREC or SUREC (ms/cm)
< 4.0 or > 9.5
4.0 to 5 .0 or 8 .1 to 9.5
5.0 to 5.5 or 7.7 to 8.1
5.5 to 6.0 or 7.3 to 7.7
6.0 to 7.3
> 12 .0
8 .1 to 12.0
4.1 to 8.0
2.1 to 4 .0
= 2). 24
W; S
= weighting coefficient based on area proportion of component or polygon i at k-flevel within polygonj at k level. = numerical constant used to scale up the magnitude of rating points.
2) Aggregation of areas of different rating classes Instead of aggregating the actual rating points, this method aggregates the area of different rating classes, i.e. `Poor', `Poor to Moderate', `Moderate to Good' and `Good". The area of each rating class is summed up respectively as expressed by the following formula: n
ISQ)m(AP,Apm,Amg,A,)=~ ISQjmf(AP,Apn,Amg,Ag)
(3-10)
Where, ISQjkn (AP,Apm,Amg,Ag)
= Areal proportion ofeach ISQ rating class (`poor', `poor to moderate', `moderate to good' and `good') ofISQ element m of polygonj at k level. (m = 1, 2, 3, 4 and represents the 4 ISQ elements respectively; j = 1,2,...,n and can be any polygon in the study area; k = 1, 2, 3, 4 and represents SLC, ecodistrict, ecoregion, and ecozone polygonal levels) .
SQ; . (A,,Apm,A,ng,Ag
= Areal proportion of each ISQ rating class of ISQ element m of component or polygon i at k-flevel (i = 1,2,...,n, components or polygons within polygonj at k level receiving same rating class, when aggregation without skipping interim levels, f=1 ; when aggregation skipping interim levels, f>= 2).
The rating class for ISQ element m of polygon j at k level is determined by the class which has the largest area proportion . This can be expressed as: ISQjm =max(ISQjm(AP,Apm,Amg,Ag))
(3-11)
For example, if the area proportion (%) of ISQ rating classes of a polygon, ISQ (AP,Apm,Amg,Ad is equal to ISQ ( 10, 20, 15, 55), the rating class ofthe polygon is "Good" (Ad . An alternative to this is to use a pie chart map to represent the areal proportion ofthe four ISQ classes within the polygon, as illustrated in Figure 3-3 . This is most effective where the total number of map polygons is limited, and each polygon covers a relatively large portion ofthe map. It is useful to report ISQ assessment results at higher level spatial units, such as ecoregions or ecozones. Ecodistricts were selected as the most appropriate spatial units for the broad regional mapping of inherent soil quality in this report. The first method, the area weighted aggregation of rating points from SLC components to ecodistrict units, was used to provide generalized ISQ ratings for 25
ecodistrict polygons . 3.5 SQS Indicators and Spatial Identification Soil quality susceptibility (SQS) to change could be assessed qualitatively, by use of selected indicators, or quantitatively, with specific procedures similar to those used for rating the inherent soil quality elements . Development of quantitative, or semi-quantitative procedures would involve investigation of a range of parameters and specific models for various types of potential soil degradation . Research on the development, calibration, and verification of various degradation models is a major area of ongoing research. The mandate of the current project was to investigate soil quality, and soil quality change, for broad geographic areas of Canada, using existing digital GIS data sources . At this level, it is more appropriate to portray soil quality susceptibility to change in a qualitative manner, using currently available indicators that are applicable at broad regional and national scales . Various potential data sources were investigated for this purpose. In order to be considered as indicators of soil quality susceptibility (SQS) for change, data had to meet the basic criteria outlined in Table 3-6.
Table 3-6 . Typical criteria for indicator selection Criteria of Indicator Selection Sensitive Diagnostic Integrative Interpretable Not redundant Appropriate scale Broadly applicable (After the Council of Great Lakes
Research Managers, 1991) The specific SQS indicators were selected based in part on the soil modifying processes and in part on the available information sources (Table 3-1) . Two distinct aspects of SQS were recognized, those due to biophysical conditions, and those due to land use and management conditions (Table 3-7).
Biophysical aspects are inherent properties of the soil and landscape that increase the susceptibility or likelihood of soil quality change . The most susceptible soils are those with a shallow topsoil, low levels of organic carbon, steep slopes, highly erodible surface textures, or shallow effective rooting depths. The assumption is that soils with one or more of these characteristics are more likely to lose topsoil due to wind or water erosion, and that these losses will result in a more significant decline in soil quality than for deeper soils with larger reserves of soil organic matter. Land use and management SQS indicators are primarily related to intensive cultivation practices, that leave the soil exposed with limited protective cover for significant periods of time. Intensive cultivation and exposure of the soil were considered likely to result in a more rapid decline in organic carbon and overall lower soil quality. The extent of particular susceptible land management practices and farming systems varies between different ecoregions. In the Prairies ecoregion of western Canada, high percentages of summer fallow result in a higher susceptibility of degradation . In eastern Canada, a high 26
percentage of row crops, and low levels of conservation tillage practices are more common susceptibility indicators . Each of these indicators identified in Table 3-7 was considered to be relatively independent . The collective effects of different SQS indicators can be spatially identified by using a GIS overlap operation . The level of susceptibility of soil quality change from water or wind erosion is estimated to be highest where biophysical and land use and management SQS indicators were both identified on the same parcels of land. Maps of these conditions can be used to identify study areas where soil quality change is most likely. Table 3-7. Selected SQS indicators and the criteria and threshold values SQS Aspects
A) Biophysical
B) Land Use and Management
SQS Indicator
Threshold Values .
Modifying Processes Affected
" Shallow topsoil
A horizon thickness 9%
Water erosion
" High erodible surface texture
Surface texture = silt or silt loam
Wind and water erosion
" Shallow effective depth
Depth to impenetrable layer Land area (notincluding water) calculated based on 1:1 million SLC data in Arc/Info format 3) Area meets minimum ISQ requirements
83.0
Table 4-4. Summary ofISQ assessment in the Mixedwood Plains Ecozone Rating class
Ontario Portion (S. ON.)
Quebec Portion (SW. QU.)
The Mixedwood Plains Ecozone
10' ha
103 ha
10 3 ha
T
%d)
%1)
1 %)
Available porosity Good
1,504
21 .6
872
38.7
2,376
25 .8
Good to moderate
5,158
74.0
702
31.2
5,860
63.6
Moderate to poor
292
4.2
295
13.1
587
6.3
14
0.2
382
17.0
396
4.3
Poor
Nutrient retention Good
3,246
46.6
1,581
70.2
4,827
52.4
Good to moderate
2,919
41 .9
550
24.4
3,469
37.6
39
Moderate to poor
612
8.8
86
3.8
698
7.6
Poor
191
2.7
34
1.5
225
2 .4
Physical rooting conditions Good
4,566
65.6
157
7.0
4,723
51 .2
Good to moderate
2,078
29.8
1,032
45.9
3,111
33 .7
Moderate to poor
324
4.6
1,057
46.9
1,380
15 .0
0
0.0
5
0.2
5
0.1
Poor
Chemical rooting conditions Good
5,349
76 .8
1,183
52.6
6,532
70.9
Good to moderate
1,619
23.2
933
41.4
2,552
27.7
Moderate to poor
0
0.0
135
6.0
135
1 .4
Poor
0
0.0
0
0.0
0
0.0
Overall rating Good Good to moderate Moderate to poor Poor ') As % of total area rated
612
8.9
47
2.1
659
7.2
5,291
75 .9
866
38.5
6,157
66.7
860
12.3
919
40.8
1,779
19.3
205
2.9
419
18.6
624
6.8
Summary statistics for each element of soil quality in Table 4-4 can also be compared to the generalized maps for the Mixedwood Plains Ecozone (Figures 4-5a,b,c,d,e) to indicate areas with particular soil quality conditions.
Available porosity
Nutrient retention
Physical rooting conditions
Chemical rooting conditions
au aa o-a..e: axa.a.oe ro moe..w: Ma- nr,a-m r.oc r-roor
o-o.oa o-M- o.ero moa..w; n+ .r-aaea-mr. . .; r-roor
Overall ISQ rating
Figure 4-4. Regional differences of ISQ ratings in the Mixedwood Plains Ecozone
Figure 4-5 Inherent soil quality (ISQ) element map of the Mixedwood Plains Ecozone : a) available porosity 42
FIgure 4-5 Inherent soil quality (ISQ) element map of the Mixedwood Plains Ecozone: b) nutrient retention 43
Good to Moderate Moderate to Poor Poor water Ecozone boundaries Montreal
hto
j1p
London
0
75
150
225
KM
Results are presented at Ecodistrict level
Figure 4-5 Inherent soil quality (ISQ) element map of the Mixedwood Plains Ecozone: c) physical rooting conditions 44
FIgure 4-5 Inherent soil quality (ISQ) element map of the Mixedwood Plains Ecozone : d) chemical rooting conditions 45
ISQ Rating
0
A/
Good Good to Moderate Moderate to Poor Poor water Ecozone boundaries
Results are presented at Ecodistrict level
FIgure 4-5 Inherent soil quality (ISQ) map of the Mixedwood Plains Ecozone: e) overall rating 46
4.3 ISQ and Potential Land Supply At broad national and regional levels, ISQ rating procedures provide a general overview of soil quality in each region and indicate the spatial variation of inherent soil quality from one area to another . They also provide an estimation of the potential land supply for agricultural crop production, as well as the quantity of land with each ISQ class. Only a limited area in Canada has climatic and soil conditions suitable for production of annual crops . The potential land supply for crop production in Prairies and the Mixedwood Plains Ecozone, based on a broad regional ISQ assessment, is shown in Table 4-5 and 4-6 respectively . Mineral soil areas that met this climatic criteria were then evaluated in terms of inherent soil quality, using ISQ92 and ISQ94 procedures, as described in Appendix 3 and 4. Table 4-5 . Potential land supply based on ISQ assessment in comparison to actual land use in the Prairies provinces. Alberta 103 ha Total land area
63,023
Saskatchewan %')
103 ha
100
58,971 [
%D
100
Manitoba 103 ha 55,378
Prairies
%')
103 ha
%')
100
177,373
100
Estimates ofpotential agricultural land supply Land suitable for agricultural crop production2)
24,064
38 .2
31,038
52.6
10,495
19.0
65,593
37.0
Land suitable for annual crop production'
16,442
26.1
28,921
49 .0
7,280
13.1
52,642
29.7
Estimates of actual agricultural land use Total farmland4)
20,811
33.0
26,865
45.6
7,724
13.9
55,401
31 .2
Area with crop cover)
12,570
19.9
20,845
35.3
5,910
10.7
39,325
22.2
Cultivated land4) 11,063 17.6 ')As % of total land area (water area not included) 2) Based on climate and minimum ISQ requirements With ISQ rating better than poor 4) Based on 1991 Census ofAgriculture data ') Based on 1989 AVHRR land cover data
19,172
32.5
5,058
9.1
35,293
19.9
In the Prairie provinces, over one-third (37%) of the total land area meets the minimum soil and climatic requirements for agriculture, and about one quarter has potential for annual cropping (an ISQ rating better than `Poor' class). Comparing these figures to the actual land use based on the 1991 Census of Agriculture, it indicates that nearly 85% of land suitable for agricultural use is currently farmed . Because a significant amount of land suitable for agricultural use (potential farmland) has been used by other alternative land uses, such as national and provincial parks, forest reserves, urban areas, and military reserves, most of the good agricultural land (in the top 3 47
ISQ ratings classes) and much marginal land (in the fourth or "Poor" ISQ ratings class) is already in production. Table 46. Potential land supply based on ISQ assessment in comparison to actual land use in the Mixedwood Plains Ecozone. Ontario Portion (S. ON.)
Quebec Portion (SW . QU.)
103 ha
103 ha
V
8,340
Total land area
100
2,764
%D
The Mixedwood Plains Ecozone 103 ha
%D
100
11,104
100
Estimates of potential agricultural land supply Land suitable for agricultural crop production')
6,968
83 .5
2,251
81 .4
9,219
83.0
Land suitable for annual crop production')
6,763
81.0
1,832
66 .3
8,595
77.4
Estimates ofactual agricultural land use Total farmland)
4,843
58.1
1,550
56.1
6,393
57.6
Area with crop cover')
5,727
68 .6
1,837
66.5
7,564
68 .1
Cultivated land4) 3,280 ')As % of total land area (water area not included) ')Based on climate and minimum ISQ requirements - 3) With ISQ rating better than poor 4~ Based on 1991 Census of Agriculture data s~ Based on 1989 AVHRR land cover data
35.1
989
32 .5
4,269
38 .4
In the Mixedwood Plains Ecozone, climatic and soil conditions are more favorable for agricultural production. About 83% of the total land area meets the minimum climatic and soil requirements for agriculture, and about 77% of the total land area has potential for annual cropping (an ISQ rating better than `Poor' class). Comparing with actual agricultural land use, nearly 68% of land suitable for agricultural use is currently farmed.. Considering the amount used by other competing non-agricultural land uses (for example, about 25% and 2% of total land area in the Mixedwood Plains Ecozone are under forest and urban uses respectively based on 1989 AVHRR satellite imagery), the potential land supply for agricultural production is limited . The estimates of potential land supply by ISQ methods in both Prairie provinces and the Mixedwood Plains Ecozone are close to the estimates of the first 5 classes of agricultural capability assessment of soil by the Canada Land Inventory (CLI) conducted in 1970s (Shields and Nowland, 1975) .
48
4.4 Areas Susceptible to Change in Soil Quality Agricultural areas of Canada with a high Soil Quality Susceptibility to change (SQS) were identified and mapped with the GIS procedures documented in the Chapter 3. Two major types of SQS indicators were recognized : biophysical indicators and land use and management indicators. The SQS, indicators and threshold values derived from available regional data bases are shown in Table 4-7. All SQS indicators were compiled at the component level of Soil Landscapes of Canada polygons . Areas susceptible to change in soil quality due to biophysical factors for westem Canada and eastern Canada are shown in Figures 4-6a and 4-7a respectively. Areas with relatively high susceptibility are those where more than one indicator are found, especially where the background shows the overall ISQ rating is Poor (the most limiting of the four ISQ elements) . Areas susceptible to change in soil quality due to land use and management factors for western Canada (Prairies and Boreal Plains Ecozones) and eastem Canada (Mixedwood Plains Ecozone) are shown in map form in Figures 4-6b and 4-7b respectively. The proportion of land susceptible to soil quality changes identified by the selected SQS indicators is summarized in Table 4-7 and 4-8 . Depth of topsoil is a most significant biophysical indicator . Canadian soils have been formed since last glacial period (10,000 B .P.) and are shallow . Between 8.4 and 56.3% of the potential agricultural land in individual provinces was less or equal to the chosen depth threshold of 15 cm. For most provinces, the steep slope indicator identified a small but significant proportion of the potential agricultural land (9 - 15 .5%) except for Quebec where the agricultural area assessed was predominantly marine and fluvial sediments. Western Canadian soil were formed under grassland vegetation, with high organic carbon content in the surface horizon . Eastern Canadian soils were formed under forest vegetation with lower organic carbon content, and also have a longer history of intensive crop production. Low organic carbon content of the topsoil is a more serious problem in eastern Canada than in western Canada. Soil quality susceptibility to change due to land use and management factors varies markedly between different ecozones in Canada. A large proportion of the agricultural area in the Mixedwood Plains Ecozone, especially in southwestern Ontario, is susceptible to soil quality change as it is predominantly crop land, with intensive row cropping . In the Prairies and Boreal Plains Ecozones of western Canada, row cropping is insignificant, but approximately 3-7% of the farmland area has a high percentage (>30% of farmland) of summerfallow practices . Most of the surnmerfallow area occurs in the portion of the Prairies Ecozone, where soil moisture is limiting for annual dryland crop production, and summerfallow has traditionally been used as part of the crop rotation. Areas with the greatest susceptibility to soil quality change are those with a combination of 49
biophysical and land use and management factors. For example, areas with steep slopes and a high percentage of summerfallow or row cropping have a greater susceptibility to soil quality change than areas with only a single SQS factor. About 2-3% of the potential agricultural area of the Prairies and Boreal Plains Ecozones, and 7- 9% of the Mixedwood Plains Ecozone is susceptible due to both biophysical conditions and land use and management practices . Table 47. Proportion of susceptible areas of soil quality change in the Prairies provinces SQS Indicators
Alberta
Saskatchewan
Manitoba
Prairies
A) SQS indicated by soil and landscape conditions') (%Z)) Shallow topsoil (A horizon thickness < 15 cm)
16.7
63.5
37 .0
42.1
Low organic carbon content of topsoil (A horizon OC < 1%)
0.1
0.6
1 .5
0.6
13 .5
15.5
9.0
13.7
18 .2
8.8
0.8
11.0
Steep surface slope (slope steepness > 9%) High erodible surface texture (surface texture = silt or silt loam)
B) SQS indicated by land use and management practices') (%1)) High cropping intensity ( area under crop > 70% of farmland)
3.0
3.4
32 .1
7.2
High level of `unfriendly' practices (summerfallow > 30% of farmland)
2.8
7.3
0.0
4.6
C) SQS indicated by both soil-landscape conditions and land use and management practices (% 2)) Indicators of co-occurance s> 1.9 3.0 1 .8 ') Based on SLC data z) As % of total land area assessed for ISQ (see Table 4-1 and 4-3) 3)Based on 1991 Census of Agriculture data 4) As % of total farmland s) Area where at least one of the bio-physical indicators overlaps with at least one of land use and management indicators
2.4
Table 48. Proportion ofsusceptible areas ofsoil quality change in the Mixedwood Plains Ecozone SQS indicators
Ontario Portion (S . ON.)
Quebec Portion (SW. QU.)
The Mixedwood Plains Ecozone
A) SQS indicated by soil and landscape conditions'? (%')) Shallow topsoil (A horizon thickness < 15 cm)
51 .5
32.5
46 .9
Low organic carbon content of topsoil (A horizon OC < 1%)
11 .4
16.5
13.2
12.0
1 .5
10.0
15.0
2.9
12.6
Steep surface slope (slope steepness > 9%) High erodible surface texture (surface texture = silt or silt loam)
B) SQS indicated by land use and management practices') ( 0/04) ) High cropping intensity 1( area under crop > 70% of farmland)
42.1
30.0
39.1
High cropping intensity 2 ( row crops > 60% of cropped land)
40.3
26.8
37.1
C) SQS indicated by both soil-landscape conditions and land use and management practices (%Z)) 7.3 8.7 Indicators of co-occurance ') Based on SLC data z> As % of total land area assessed for ISQ (see Table 4-1 and 4-3) ')Based on 1991 Census of Agriculture data a) As % of total farmland s) Area where at least one ofthe bio-physical indicators overlaps with at least one ofland use and management indicators
8.0
Broad scale areas with particular combinations of ISQ ratings and SQS conditions can be readily identified using GIS mapping techniques and the tools developed in this project. These areas can be targeted for more detailed analysis and monitoring or delivery of programs to promote alternative land use and management practices to enhance long term agricultural sustainability .
SQS INDICATORS
N 0
100
200
300
Shallow topsoil (< 15 cm) Low organic carbon content oftopsoil (< 190) Steep slope (> 9%) High erodible surface texture (silt or siltloam) water Area excluded Ecozone boundaries
SOS indicators are presented at SLC polygon level
KM
Figure 4-6 Soil quality susceptibility (SQS) map of the Prairies provinces: a) biophysical 52
High intensive land use (cropland > 7096 offarmland) High level of 'unfriendly' land managementpractice (summerfallow > 3096 offarmland) SQS not indicated water 0 Area excluded Ecozone boundaries
n/
KM
SOS indicators are presented at SLC polygon level
Figure 4fi Soil quality susceptibility (SQS) map of the Prairies provinces: b) land use and management 53
SQS INDICATORS
0 N
~~r.~
pr,~wrcs
Shallow topsoil (< 15 cm) Low organic carbon content of topsoil (< 196) Steep slope (> 996) High erodible surface texture (silt or silt loam) SQS not indicated water Ecozone boundaries
t'
~ ia ~ _i~~
pW .lfl
AMA 6. 7RJ7-
1I a~z~~~r r~ is W~~ ~~w"wymawi
75
150
225
KM
SOS indicators are presented at SLC polygon level Figure 4-7 Soil quality susceptibility (SQS) map of the Mixedwood Plains Ecozone: a) biophysical 54
Figure 4-7 Soil quality susceptibility (SQS) map of the Mixedwood Plains Ecozone : b) land use and management 55
4.5 Implications of Changing Land Use and Management Practices on Soil Quality Land use and management practices, such as high intensity cultivation or summerfallow, are indicators of the potential decline in soil quality. While actual soil quality change (SQC) can not be directly measured at broad national and regional scales, changes in these land use and management indicators, can be used as an indication of the trends in soil quality change. It is useful to conduct this kind of analysis in conjunction with inherent soil quality (ISQ assessments to locate areas where changes in soil quality are most likely and where more detailed assessment is appropriate . In this study, data of the two census years (1981 and 1991) were used. The areas where changes occurred are shown in Figure 8 for the Prairies and Boreal Plains Ecozones and Figure 9 for the Mixedwood Plains Ecozone. Numbers showing the direction and magnitude of change over the 10 year period are presented in Table 4-9 and 4-10. Some conservation practices, such as conservation tillage and no-till, can halt or reverse soil degradation, or at least reduce the risks . Conservation tillage and no-till management have been adopted in the major cropping areas of Canada since 1981 . The location and extent of theses practices could be mapped (only for 1991 as data were not collected prior to this date) to indicated areas where soil quality may be improving and becoming less susceptible to change Table 4-9 and 4-10 lists the estimates of the changes of selected indicators of land use and management practices between 1981 and 1991, the last two Census of Canada periods . These areas of change in management practices are also shown in map form for the Prairies provinces (Figure 4-8) and the Mixedwood Plains Ecozone (Figure 4-9). Table 4-9. Change in selected SQS indicators in the Prairies provinces from 1981 to 199') SQS indicators High cropping intensity (area under crop > 70% of farmland)
Year
Alberta
Saskatchewan
Manitoba
Prairies (% 2))
1981
1 .2
1 .4
21 .3
4.2
1991
3.0
3.4
32.1
7.2
0.8
13.5
0.0
6.9
2.8
7.3
0.0
4.6
High level of unfriendly' 1981 practices (summerfallow > 1991 30%) '~ Based on 1981 and 1991 Census ofAgriculture data z> As % of total farmland
Table 4-10. Change in selected SQS indicators in the Mixedwood Plains Ecozone from 1981 to 1991') SQS indicators High cropping intensity 1 ( area under crop > 70% of farmland)
Year
Ontario Portion (S. ON.) (%2))
Quebec Portion (SW. QU.) (%2))
The Mixedwood Plains Ecozone (%2))
1981
37.3
23.2
33.9
1991
42 .1
30.0
39.1
43.4
18.8
37 .6
40 .3
26.8
37.1
High cropping intensity 2 ( 1981 row crops > 60% of cropped 1991 land) ') Based on 1981 and 1991 Census ofAgriculture data z) As % of total farmland
There has been about 3% and 5% increase in areas of high cropping intensity (>= 70% of farmland) in the Prairies provinces and the Mixedwood Plains Ecozone respectively over the 10 year period. The area of extensive use of summerfallow (>= 30% of farmland) decreased slightly in some areas in the Prairies provinces (Table 4-5 and Map 4-10). Also conservation tillage and no-till practices have been adopted and used by many farmers in the past decade (Table 4-11 and 4-12). This might indicate that the soil quality in some areas are improving and becoming less susceptible to major degradation processes such as decline in organic matter and erosion by wind and water. Table 4-11. Conservation tillage and no-till practices used in the Prairies provinces (1991)') Tillage Practices Conservation tillage No-till Conservation tillage and no-till ') Based on 1991 Census ofAgriculture data z) As % of total land prepared for seeding.
Alberta (%)2)
Saskatchewan (%)z)
Manitoba (%) z)
Prairies (%)z)
24 .3
25 .7
28 .7
25.8
2.7
10.2
4.6
6.9
27 .0
35 .9
33 .3
32.7
Table 4-12 . Conservation tillage and no-till practices used in the Mixedwood Plains Ecozone (1991)') SQS indicators
Ontario Portion (S. ON.) (%)Z)
Conservation tillage No-till Conservation tillage and no-till ') Based on 1991 Census of Agriculture data s) As % of total land prepared for seeding.
57
Quebec Portion (SW. QU.) (%)Z)
The Mixedwood Plains Ecozone (%)z)
17 .7
12.2
16.5
3 .3
1.3
2.9
21 .0
13.5
19.4
r
r
_
r
,
F-
r
r
r- -
CHANGE OF SUSCEPTIBLE AREAS Areas indicatedfrom 1981 data Areas indicatedfrom 1991 data Areas indicatedfrom both 1981 and 1991 data
0
SOS indicators arepresented at SLC polygon level
KM
Figure 4-8 Changes of selected SQS indicators of land use and management in the Prairies provinces from 1981 to 1991 58
r---
CHANGE OF SUSCEPTIBLE AREAS
0 0
Areas indicatedfrom 1981 data Areas indicatedfrom 1991 data Areas indicated from both 1981 and 1991 data SQS not indicated water Ecozone boundaries
r rr
onto
LDndon
0
75
150
225
KM
SOS indicators are presented at SLC polygon level Figure 49 Changes of selected SQS indicators of land use and management in the Mixedwood Plains Ecozone from 1981 to 1991 59
5.0 Discussion 5.1 The Sensitivity of the ISQ Procedures Land resource data sets are essential for the spatial analysis of Inherent Soil Quality. Broad scale regional and national data sets provide comprehensive coverages, but have coarse spatial resolution, and a limited set of attributes . More detailed scale data sets provide finer spatial resolution, and a larger number of attributes to enable more precise interpretive ratings, however, the coverage areas may be limited. Different ISQ rating algorithms have been developed to use the most commonly available Canadian data sources. An comparison of these methods and data sources is provided in this section. Two separate investigations were conducted to evaluate the sensitivity of ISQ procedures . The first study (Section 5.1 .1) is an evaluation of ISQ sensitivity at broad regional scales, using two different data sets and ISQ algorithms . The second study (Section 5.1 .2) is an evaluation of ISQ sensitivity to changes in map scale . A single ISQ rating procedure (ISQ94) was used for a comparative analysis of the same test area, using both detailed and broad scale data sets . 5.1.1 Sensitivity to Different Data Sets and Algorithms Two separate ISQ procedures were developed for ISQ assessments at broad regional scales . Both use 1 :1 million scale Soil Landscapes of Canada digital maps. The first method ("ISQ92") uses the generalized SLC attributes in the Dominant ("DOM") and Subdominant ("SUBDOM") extended legend files. These data sets are available for all of Canada . Details of the ISQ92 rating algorithms are in Appendix 3. The second ISQ method ("ISQ94") uses the more extensive set of modal soil attribute information available in standardized Soil Names and Soil Layer Files ("SNF' and "SLF' files). While initially developed for use with detailed soil maps, the more generalized Soil Landscapes of Canada map polygons are described in terms of soil series components (SLC CMP file records), and can be linked to SNF and SLF soil attribute files . ISQ94 procedures can therefore be used to evaluate both detailed soil maps and SLC maps (Appendix 4). This capability is currently limited to certain regions of Canada, such as the prairie provinces, where the SNF and SLF files have been completed for all SLC soil components. National or regional ISQ assessments may therefore use a combination of ISQ92 or ISQ94 data sets and algorithms for particular regions. The relative sensitivity and reliability of the ISQ92 and ISQ94 procedures was evaluated, using the Prairies provinces, where data sets for both methods are available. A summary of the overall ISQ classification results for the prairies, for each element of soil quality, are indicated in the bar charts in Figure 5-1 . Alternatively, this can also be expressed in terms of classification agreement or disagreement, on an areal extent basis, as summarized in Table 5-1 .
60
The ISQ92 and ISQ94 methods show a reasonable level of agreement. On an overall, area weighted basis, nearly 50% of the agricultural land area was rated in the same ISQ class by both methods, while a further 35% of the total land area was rated within one class of each other. The level of agreement varied between the different ISQ elements. Areas with either complete agreement, or agreement within one ISQ class, varied from a high of 92.5% for the Nutrient Retention element to a low of 75.2% for the Available Porosity element (Table 5-1). The Porosity Element is also the element that has the greatest difference between the ISQ92 and ISQ94 ratings algorithms . The ISQ92 algorithm rates porosity in terms of moisture supply limitation, while ISQ94 has separate subelements to evaluate both moisture and air supply limitations . Other differences between ISQ92 and ISQ94 ratings results may be attributed to differences in the dominant soils and soil properties in the different data sources, and to more strict exclusion procedures used in the ISQ94 algorithms . Both ISQ92 and ISQ92 are based on the same theoretical framework of soil quality assumptions, and use a similar set of four ISQ elements for the cereal crop production function. Although ISQ94 is a more rigorous rating procedure than ISQ92, there is a relatively high level of agreement between the two systems within the prairie region. This indicates that ISQ ratings, at broad scales, are relatively compatible, and not particularly sensitive to differences in data sources or ratings algorithms . ISQ92 procedures can therefore be used as a substitute for ISQ94 procedures, in areas where ISQ94 procedures cannot currently be used due to data limitations.
.
Available porosity
Nutrient Retention
physical rooting conditions
Chemical rooting conditions
Figure 5-1 Comparison ofthe results of ISQ92 and ISQ 94 in the Prairies provinces
61
Table 5-1 . Comparison of ISQ94 and ISQ92 ratings in the Prairies provinces
ISQ Elements
Agreement
Total Area 102 ha
0 class off
Disagreement 1 class off
3 classes off
2 classes off
%
Area 102 ha
%
Area 102 ha
%
Area 102 ha
%
Area 102 ha
Other" Area ha
102
%
%
Available Porosity
403,439
100 .0
143,576
35.6
159,449
39.6
89,411
22.2
1,501
0.3
9,467
2.4
Nutrient Retention
403,439
100 .0
204,064
50.6
168,889
41.9
30,299
7.5
185
-
-
-
Physical Root Conditions
403,439
100 .0
259,077
64.2
84,689
21.0
50,804
12.6
3,497
0.9
5,331
1.3
35,897
8.9
13,559
3.4
9,043
2.2
Chemical Root 403,439 100 .0 181,223 44.9 163,714 40.6 Conditions ') included and rated in ISQ92 but not in ISQ94, as ISQ94 has more strict excluding procedures .
5.1.2 Sensitivity to Map Scales Inherent Soil Quality ratings can also vary with the scale or precision of the spatial databases. In this study, the same ISQ94 rating procedure was evaluated using both detailed and broad scale digital soil data bases, for the same map area. This comparison indicates how well ISQ interpretations at broad regional scales can portray or represent the more detailed information available for detailed soil maps. It is important to understand how much of the detailed variation is masked at broader scales (coarser resolution), and whether this introduces any systematic bias within the classification results. A single study area was selected in south central Manitoba between Brandon and Winnipeg (Figure 5-2) . This area covers 12 Rural Municipalities (over 1 million hectares), and represents a wide range of different soil landscape conditions within the Prairie ecozone . Soil landscapes include extensive areas of lacustrine sands, loams and clays, glaciofluvial and eohan sands, alluvial soils, marshes, and portions of three separate till plains. The broad scale analysis was conducted using the 1 :1 million scale Manitoba SLC digital map and Soil Component file. 25 SLC polygons, representing 97.44% of the total area, met the selection criteria (i.e. occupied over 4000 ha and had between 20% and 100% of their total area within the study area). The detailed soil data base coverage for the test area consisted of 12 separate Rural Municipality maps (approximately 17,400 soil polygons), at, a nominal map scale of Figure 5-2 . Location of the scale sensitivity test 1 :100,000. The "broad scale" and "detailed area in Southern Manitoba . scale" data sets were both evaluated using ISQ94 rating procedures, with soil polygon components linked via common soil codes to the same global Soil Names and Soil Layer modal soil property files. Since the technical procedures were the same, differences in ratings results can be attributed to differences in map scale and spatial resolution. A comparison of the overall broad scale and detailed scale ISQ classification results for the Manitoba test area, for each element of soil quality, are indicated in Figure 5-3. The results show a good general level of agreement between the detailed and generalized assessments . For all four ISQ elements, both detailed and broad scale methods are in agreement as to the relative ranking of the ISQ classes . The detailed and broad scale data bases are also in close agreement as to the percentages of the overall test area that does not meet the criteria of cereal crop production ; approximately 28% and 29% respectively . An alternative comparison was made of the detailed and broad scale ISQ classification results for 63
the 25 significant SLC polygons within the test area (Table 5-2). Table 5-2 . Comparison ofthe ISQ 94 ratings at different scales in Southern Manitoba . Agreement
Total SLC polygons"
ISQ Elements
0 class off
Disagreement 1 class off
2 classes off
3 or more classes off
Number
%
Number
%
Number
Available Porosity
25
100
12
48
5
20
7
28
1
4
Nutrient Retention
25
100
18
72
3
12
3
12
1
4
Physical Rooting Conditions
25
100
21
84
1
4
0
0
3
12
Chemical Rooting Conditions
25
100
12
48
8
32
1
4
4
16
Overall Rating (most limiting)
25
100
11
44
9
36
4
16
1
4
1)
%
Number
%
Number
number of SLC polygons
Available porosity
Nutrient Retention
Physical rooting conditions
Chemical rooting conditions
Figure 5-3. Comparison of the ISQ94 ratings at different scales in Southern Manitoba
64
%
Exact agreement as to the dominant (modal) overall ISQ class rating occurred in 44% (11 out of 25) of the SLC polygons, while a further 36% (9 of 25) were within one class. Overall ISQ ratings were therefore in either exact agreement, or within one class, for 80% (20 out of 25) of the SLC polygons within the test area. For the individual ISQ elements, the percentage of SLC polygons in agreement, or in approximate agreement (within one rating class) varied between 68% for Porosity, 84% for Nutrient Retention, 88% for Physical Rooting, and 80% for Chemical Rooting conditions . This indicates a good general correspondence between the detailed and broad scale ratings, for most SLC polygons . However, approximately 20% of the SLC polygons show a significant difference in classification results between the two map scales. A more detailed analysis of the ISQ ratings within selected SLC polygons in the test area is provided in Table 5-3. The ISQ ratings classes with the highest areal coverage in each SLC polygon, for both detailed and broad scale methods, are shown in bold. This illustrates more clearly the divergence of the ISQ rating results between detailed and broad scales . At detailed scales, several hundred soil polygon components are evaluated within each SLC area. ISQ ratings are typically distributed over a wide range of rating classes (from "Good" to "Not Rated"), reflecting the variety of different soil landscape conditions in the detailed soil maps. At broad scales, SLC polygons are represented in the Soil Component file by a comparatively small number of soil components, typically only 2 or 3 soils per polygon. Much of the variability in the detailed ratings is therefore masked. For many SLC polygons, such as SLC 44 and 59 (Table 5-3), the dominant ISQ interpretive ratings at both scales remain closely matched. For other SLC polygons, broad scale results may be unrealistically high or low, in comparison to detailed data bases . For example, SLC 58 is represented by only a single soil component (100%), and the soil, an eolian Regosol, received an unsuitable ISQ rating . Only 27.1% of the SLC was considered unsuitable for cereal crop production using the detailed soil data bases for the same area. Conversely, SLC 52 is represented by two soils components at broad scales, with the first component (70%) rated "Good to Moderate" and the second (30%) rated "Moderate to Poor". Both soils, representing 100% of the polygon, met the criteria for ISQ ratings . However, 44.7% of the polygon area was considered unsuitable when rated using the detailed soil data bases. Broad scale ISQ ratings for individual SLC polygons are therefore critically dependent on the soils selected in the SLC CMP files. For approximately 20% of the SLC polygons within the Manitoba test area, the soil components currently in the SLC CMP were not adequate to represent the modal ISQ conditions, as rated by the detailed map coverage . At the present time, somewhat more reliable results can be obtained using detailed soil data bases, re-aggregated to an SLC level. This option is only available in areas where detailed digital soil database coverage has been completed to national soil data base standards .
Table 5-3. Comparison of detailed and broad scale ISQ ratings for selected SLC polygons in Southern Manitoba. SLC GO 22
a) 2) b)
31
a) b)
32
Available Porosity %) G-M
M-P
P
N
42.4
14 .2
9 .2
7 .8
26 .6
Nutrient Retention (%)
GO
G-M
M-P
P
N
60 .9
2 .9
9 .7
0.0
26 .6
65 .0
0 .0
0 .0
0 .0
35 .0
0 .0
0.0
55 .0
10 .0
40 .6
1 .8
49 .2
35 .0
70.0
0 .0
30 .0
0 .0
8 .3
91 .0
0 .0
0 .6
0 .0
0 .0
0 .0
100 .0
0 .0
0 .0
3.1
8 .7
69 .7
3 .3
18 .3
0 .0
0.0
0 .0
100 .0
0 .0
0 .0
Physical Root Conditions (%)
GO
M-P
73 .4
0.0
0.0
0.0
26 .6
17 .8
65 .0
0.0
0.0
0.0
35 .0
8 .0
0.0 0 .0 0.1
8 .3
83 .6
0.0
0 .0
100.0
0.0
8 .7
86 .8
0 .0
0.0 4 .4
0 .0
100 .0
0 .0
100 .0
0 .0 0.0
0.0 0.0
100 .0
0 .0
0 .0 1 .0
a)
52.8
13 .9
21 .6
b)
100.0
0 .0
0 .0
a)
77.9
3 .2
b)
0.0
70.0
19 .0 30.0
0.0
0 .0
100 .0
0 .0
0 .0
0 .0
75 .8
0 .5
19.7
0 .0
30 .0
0.0
70.0
0 .0
44
a) b)
0.0
70.0
0 .0
0 .0
0.0
4 .1
95.4
0.0
0.5
0 .0
0 .0 4 .1
49
a)
14 .4
56.8
12.8
10.0
20 .0
80.0
0 .0
0 .0
0 .0
20 .0
b)
70 .0
30.0
0.0
16.0
25 .1
2.1
0.0
70.0
0.0
0 .0
1 .9
0.0
84.0
28 .4
0.0
16.0
a)
0.0 24.1
56.8
b)
70.0
0.0
30.0
0.9
44.7
47.1
100.0 50.0
a) b)
4 .5
16 .5
5 .1
0.0
0.0
0 .0
0.0
10.0
0.7 0 .0
a)
9 .8
13 .4
b)
0.0
30 .0
3 .7
57
a)
3.9
32 .4
0.0
58
b) a) b)
33
52 55 56
59
a) b)
61
a)
62
a) b
b)
0 .0 0 .0
94 .9 80 .0
0 .0 0 .0
P
N
Chemical Root Conditions (%)
G-M
G')
G-M M-P
P
2.4
55 .0
10.0
0.0
0 .0
8 .3
16 .8
45 .1
29.8
0 .0
8 .3
0 .0
30 .0
0.0
70.0
0 .0
0 .0
8 .7
10 .8
45 .5
35 .0
0.0
8 .7
0.0 0.0
0 .0 0 .0
0 .0
100.0
0.0
0.0
0.0
0 .0 0 .0 1 .2
69.5 0 .0
0 .0
0 .0 4 .1
0 .0
20 .0
0 .0
N
53.2
26 .6 35 .0
30.5
0 .0
0 .0
100 .0
0.0
0.0
76 .3
18 .4
0.0
4.1
0.0
20.0 16 .0
0 .0
0 .0
0 .0
0 .0
16 .0
60.7
80 .0 18 .5
0.0 4 .8
0.0
0 .0 4 .8
0 .0 0 .0
0 .0 0 .4
0 .0 44 .7
100.0 2 .8
0 .0 31 .8
0 .0
0.0
0.0
19 .7
1 .0
44.7
0 .0
100 .0
0 .0
0.0
0 .0
30.0
0.0
0.0
8 .1
0.0
0.0 44 .7
100.0
0.0
0.0
0.0
0 .0
100.0
0 .0
0 .0
0 .0
0 .0
73 .2
5 .7
0.0
21 .1
0.0
73.2
26.8
0 .0
0 .0
0 .0
73.2
8 .3
16 .2
2 .3
0.0
73 .2
90.0
10.0
0.0
0.0
0.0
90.0
10 .0
0 .0
0 .0
0 .0
90.0
0.0
0 .0
10 .0
0.0
90.0
5 .2
67 .9
10.2
1 .6
20.1
0.3
32 .1
0 .0
0.0
30 .0
0.0
0.0
0 .0
70 .0
44 .1
30.2
13 .3
12 .4
0.0
55.9
0 .0
6 .3
0 .0
44 .1
3 .3
11 .4
54.7
0 .0 0 .0
0.0 0.0
0 .0
19 .6 10 .0
70.0 44.1
30 .0
0 .0
70.0 44 .1
30.0
13 .6
67 .9
0.0
67.9
0.0
70 .0
0 .0
5 .1
0 .0
0 .0
13 .4
0 .0
67.9
90 .0
10.0
0.0
0.0
0.0
90.0
10.0
0.0
0 .0
10 .0
0.0
90.0
0.0
0 .0
0 .0
0 .0
27 .1
58 .8
2 .3
11 .9
0 .0
27 .1
57.5
15 .1
0 .3
0 .0
27 .1
60.2
17 .0
5 .4
1 .9
100 .0
0 .0
0.0
0 .0
0 .0
100 .0
0 .0
0 .0
0 .0
100 .0
80.0
20.0
0.0
0.0
51 .2
3 .1
29 .8
0 .5
15 .4
56.1
14.4
4.0
1 .8
15 .4 0 .0
80.0
0 .0
20 .0
0 .0
100.0
0.0
0.0
0.0
23 .8 0 .0
54 .0
9 .2
13 .1
44 .9
6.2
24 .0
4.4
20 .5
100 .0
0 .0
90.0
0.0
0.0
10.0
0.0
73 .2 40 .0
3 .5
0.0-
0.0
0.0 0 .0
0 .0
0.0
0 .0
0 .0
72.9
0.0
0.0
0.0
90.0 27.1
0.0
0.0
0.0
0.0
0.0
15 .4
84 .5
0.1
0.0
0 .0
100.0 15.4
7 .2
59.7
17 .2
0 .4
0 .0
100 .0
0.0
0.0
0.0
0.0
0 .0
0.0
0 .0
0 .0
0 .0
23 .8
76 .2
0.0
0.0
0.0
23 .8
60.1
100.0 13 .9
2.3
0 .0
23 .8
0 .0
0 .0
0 .0
100.0
0.0
0.0
0.0
0.0
100.0
0 .0
0 .0
0 .0
0 .0
5 .4
0 .9
0 .0
20 .5
78 .3
0.4
0.9
0 .0
20.5
43 .5
24 .4
10 .8
0 .7
20 .5
60 .0
0 .0
0 .0
0 .0
100 .0
0 .0
0.0
0.0
0.0
90 .0
10.0
0.0
0 .0
0 .0
') G = Good; G-M = Good to moderate ; M-P = Moderate to poor ; P = Poor; N = Not rated 2) a) = rating at detail scale; b) = rating at broad scale Dominant rating class for each SLC polygon at each scale is shown in bold .
49.6
At broad regional scales, from 1 :500, 000 to 1 :2 million, SLC polygons are the appropriate units for ISQ spatial analysis. Analysis using ISQ94 algorithms and SLC CMP data bases offer the best combination of precision and comprehensive coverage . This should be the method of choice, for areas where the SLC CMP files contain representative soil components, and Soil Names and Soil Layer data bases are successfully linked. GIS techniques can be used to assist in the upgrade of existing SLC CMP files, in areas where both detailed and SLC coverages coexist. When ISQ classification results were further generalized for the entire test area, it was observed that there was a high degree of similarity between the detailed and broad scale (SLC) methods (Figure 5-3). Differences between the detailed and broad scale interpretations, perhaps fortuitously, appeared to cancel each other out when averaged over all 25 SLC polygons within the test area. This indicates that, although the broad scale analysis masks some of the detailed map variations within each SLC polygon, it does not seem to introduce any systematic bias in the more generalized ratings for any of the ISQ elements . For very broad regional or national scale analysis, at scales from 1 :2 million to 1 :20 million, individual SLC polygon ratings are typically reaggregated and summarized within broader land units. These may be administrative units, such as crop reporting districts or provinces, or ecological units, such as ecodistricts, ecoregions, or ecozones . At these very broad scales, the finer resolution offered by detailed soil maps is not warranted, even in the limited areas of the country where such coverage is available . The coarser resolution SLC databases, analyzed by either ISQ92 or ISQ94 procedures and reaggregated to larger units, should provide adequate precision for such analysis. 5:2 Future Applications of ISQ and SQS Procedures ISQ and SQS procedures were developed as tools for broad level land resource analysis, using soil quality definitions and currently available Canadian GIS data bases . While intended for broad scale planning and policy applications, they may be used to indicate areas where more detailed investigations are warranted. Land resource data base coverages are expected to increase in both areal extent and precision in the future . Completion of global Soil Names and Soil Layer data bases in eastern Canada will permit use of more precise broad scale ISQ94 rating procedures . ISQ94 rating procedures can also be employed at more detailed levels in additional areas, as detailed NSDB data bases are completed. Additional Census of Agriculture questionnaire data bases will extend the time frame, and permit more accurate forecasting of Soil Quality Susceptibility to change (SQS) . Many refinements of the current ISQ rating procedures are also feasible, particularly regarding the integration of climatic information . Currently, ISQ procedures only consider two climatic factors, Effective Growing Degree Days (EGDD), and soil moisture availability, as a function of soil taxonomy. Both of these factors reflect only average long term conditions, and do not reflect probabilities . Additional national climatic data bases are now available, with a wider range of 67
climatic attributes calculated on an ecodistrict basis. In several regions, climatic data bases are being enhanced with additional stations, refinement of climatic contours to account for physiography, and more advanced probability calculations . These could all be utilized by enhanced versions of the current ISQ rating procedures . ISQ rating procedures can also be modified to match the soil and climatic criteria of other crop types. The distribution of the ISQ element ratings, and the total area that meets the minimum ISQ rating criteria, may be significantly different than soil quality ratings for cereal production. An interesting application of these ISQ techniques is the estimation of areas suitable for future agricultural production, under global climatic change scenarios . Soil conditions for Canada are relatively static, and are documented in the Soil Landscapes of Canada data base. Various climatic models predict global warming in northern areas of Canada, and possible increases in aridity in other areas, such as the southern prairies . If climatic model predictions (such as revised effective growing degree days contours) are available, ISQ procedures can be used to identify the future areas with favorable soil and climatic conditions for future types of crop production. Some regions may have a more favorable climate, but may have adverse soil conditions that limit an increase in the potential agricultural land base. Another application is the integration of ISQ GIS techniques with soil degradation models . Spatial analysis of ISQ can be done at different time periods, under a range of soil climatic, management, and degradation model scenarios . ISQ analysis can show the direction and magnitude of predicted soil quality changes over space and time, for any particular area. This can be a powerful tool for forecasting the effects of different management recommendations, and determining which alternatives are compatible with principles of sustainable development.
6.0 Conclusions The current systems for assessment of inherent soil quality and soil quality susceptibility to change were designed to use available databases and capabilities of existing GIS systems . The systems were designed to be applied to broad regional and national level (1 :1 - 1 :5 million). In addition, some procedures such as ISQ94, can be applied to detail-scale (1 :25000 - 1:50000) assessments . The results of the assessments may be combined with other environmental information : i) to assess sustainability of current soil and land management systems; ii) to target current and potential 'problem areas' of soil quality change and land degradation for monitoring and detailed studies; iii) for State of Environment (SOE) reporting . Technically, the systems and procedures were designed to be flexible enough to incorporate additional data layers, adjust the rating criteria and algorithm for specific crops or other land use requirements, be used at more detailed levels, e.g. watershed and farm levels, combine with other analytical tools or models, e.g. USLE or RUSLE to assess the potential effects of soil water erosion to soil quality at regional and national level. Results and models developed in other projects within SQEP may provide capabilities to enhance the ISQ and SQS assessment procedures . This system cannot be directly compared with previous evaluation systems for Canada (for example, Wang et al 1991, MacDonald and Brklacich, 1992; Brklacich and MacDonald, 1992, Pettapiece 1995), as the approach, objectives, databases and technical environments are not the same. The evaluation of the sensitivity of the system to different data sets and scales indicates that running ISQ rating program using alternative data sets or at different scales can achieve rating within reasonable range. Assessments of soil quality change requires repetitive monitoring and resampling of soil landscape attributes, at specific sites or small plots, over a long time period. `Procedures to measure actual soil quality change at broad scales were not developed, as the existing land resource data sources are not appropriate for this type of assessment. In the future it may be possible to provide assessments of predicted soil quality change, using process-based models in combination with land resource, climatic, and land use and management data sets . At the current time, Soil Quality Susceptibility to change (SQS) is offered as a means of indicating likelihood of changes in soil quality at broad regional scales . More precise estimates require detailed, large scale data as well as modeling procedures to characterize soil modifying processes. However, the sequential assessment (for years with Census of Agriculture Data) of soil quality and its spatial representation is operational at broad regional and national scales using existing databases .
69
References Acton, D. F. 1991a. Concept and criteria of soil quality in the context of western Canada, pp.2-4. In Soil quality in the Canadian context--1988 discussion papers, Technical Bulletin, 1991-1E, Research Branch, Agriculture Canada, Ottawa, Ontario, Canada. Acton, D. F. 1991b . The national soil quality evaluation project, pp.52-62. In proceedings of the ninth meeting of ECSS, Research Branch, Agriculture Canada, Ottawa, Ontario, Canada. Acton, D. F. 1992 . Development of a framework for soil quality evaluation in Canada, SQEP Working Paper (unpublished), Research Branch, Agriculture Canada. pp.1-14 Acton, D. F., MacDonald, K.B . and Pettapiece, W.W. 1992. A program to assess and monitor soil quality at regional and national scales: a Canadian experience, SQEP Working Paper (unpublished), pp. 1-17. Acton, D. and Gregorich, L .J. (ed), 1995 . The health of our soils -toward sustainable agriculture in Canada. Centre for Land and Biological Resources Research, Research Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada. Arnold, R.W., Sambroek, W .G., Szabolcs, I. and Targulian,V .0. 1990 . Future changes of pedosphere, pp. 87-89. pp . 87-89 . In R.W.Arnold, I. Szabolcs and V.O .Targulian (ed), Global soil change, Int. Inst. for Applied System Analysis, Laxenborg, Austria. Ayala, H. L. 1987. New view of environmental health : towards an environment research policy for international strategies, Environmental Management, Vol. 11, No. 2, pp. 141-147 Baumgardner, M.F. 1990. Spatial soil databases and modelling, pp. 91-102 . In R.W .Arnold, I. Szabolcs and V.O.Targulian (ed), Global soil change, Int. Inst. for Applied System Analysis, Laxenborg, Austria. Beinat, E., Nijkamp, P. and Rietveld, P. 1994. Indices of soil quality: a multicriteria value functions approach, Journal of Environmental Systems, v.23, No.l, pp.1-20. Best, E.P.H. and Haeck, J. (ed), 1983. Ecological Indicators for the Assessment of the Quality of Air, Water, Soil, and Ecosystem: paper presented at a symposium held in Utrecht, October 1982, Dordrecht, The Netherlands . Bouma, J., 1989a. Using soil survey data for quantitative land evaluation, pp. 177-213 . In B .A. Steward (ed), Advances in Soil Science Vol. 9. Springer-Verlag, New York. USA. Bouma, J., 1989b . Land quality in space and time. pp. 3-13. In J. Bouma and A.K. Bregt (ed), Land quality in space and time, Proceedings of ISSS Symposium, Wageningen, Pudoc ., Wageningen. The Netherlands . Bouma, J, 1997. Soil environmental quality : a European perspective, Journal of Environmental Quality, 1997, v.26, No. l, pp .26-31 Brklacich, M. and MacDonald, K.B . 1992. Prototype agricultural land evaluation systems for Canada: II. Selected applications and prospects, Soil Use and Management, Vol. 8, No. 1, pp.8-15. Bur-rough, P. A. 1986. Principles of Geographical Information Systems for Land Resources Assessment, Oxford University Press . UK . Burrough, P. A. 1989a. Modelling land qualities in space and time: The role of geographical information systems, pp .45-59. In J. Bouma and A.K. Bregt (ed), Land quality in space and time, Proceedings of ISSS Symposium, Wageningen, Pudoc ., Wageningen. The Netherlands . 70
Burrough, P.A. 1989b . Matching spatial databases and quantitative models in land resource assessment, Soil Use and Management, Vol. 5, No . 1, pp.3-8 . Cass, A, Gusli, S. and MacLeod, D.A.1994. Sustainability of soil structure quality in rice paddy--soya-bean cropping systems in South Sulawesi, Indonesia, Soil and Tillage Research, v .31, No.4, pp.339-352 . Coote D .R., Dumanski, J. and Ramsey, J.F. 1982. An assessment of the degradation of agricultural lands in Canada, Research Branch, Agriculture Canada, Ottawa, Canada. Council of Great Lakes Research Managers, 1991 . A proposed framework for developing indicators of ecosystem health for the Great Lakes region, International Joint Commission, pp. 12-14. Doran, J W, Sarranntonio, M. and Leibig, M.A. 1996. Soil health and sustainability, pp.1-54. In D .L. Sparks (ed). Advances in agronomy Vol.56, Academic Press, New York. USA. Doran, J.W. and Parkin, T.B. 1994. Defining and assessing soil quality, pp. 3-21 . In J.W. Doran, D.C. Coleman, D.F. Bezdicek, and B.A. Stewart (ed) . Defining Soil quality for a sustainable environment . SSSA Special Publication No. 35, SSSA, Inc. Madison, USA. Doran, J. W., Coleman, D .C., Bezdicek, D . F. and Stewart, B.A. (ed). 1994 . Defining soil quality for a sustainable environment, Soil Science Society of America Special Publication 35, SSSA, Madison WI. USA. Doran, J.W. and Parkin, T.B . 1996.Quantitative indicators of soil quality: a minimum data set, pp. 25-38, In J.W. Doran an A .J. Jones (ed) Methods for assessing soil quality. Special Publication 49, SSSA, Madison. WI, USA. Ecological Stratification Working Group, 1995. A national ecological framework for Canada. Agriculture and Agri-Food Canada, Research Branch, Centre for Land and Biological Resources Research ; and Environment Canada, State of the Environment Directorate, Ecozone Analysis Branch, Ottawa/Hull . Canada. Environmental Indicator Working Group, 1994 . Report of the consultation workshop on environmental indicators for Canadian agriculture, Agriculture and Agri-Food Canada, Policy Branch, Environment Bureau, Ottawa, Canada. Es, H.M.Van. 1995 . Soil and water quality: an agenda for agriculture - committee on long-range soil and water conservation, Board on Agriculture, National Research Council, National Academy Press, Journal of Environmental Quality, 1995, v.24, No.2, pp.386 . Fraser, W.R., Lelyk, G.W., Wang, F. and MacDonald, K.B. 1994. Inherent soil quality rating program for crop production (version 2.0) - Program documentation . SQEP-2 Technical Document. pp. 1-20. Fraser, W.R., Lelyk, G.W., MacDonald, K.B., and Wang, F. 1994.The development of GIS procedures to assess inherent soil quality at regional and national scales, pp. 20-32 . In Proceedings of the 37th annual meeting of the Manitoba Society of Soil Science, Jan. 4 and 5, 1994, Winnipeg, Manitoba . Canada. Gregorich, E. G., Carter, M. R., Angers, D. A., Monreal, C. M. and Ellert, B. H. 1994 . Towards a minimum data set to assess soil organic matter quality in agricultural soils, Canadian Journal of Soil Science, Vol.74, No.4, pp.367-386 . Haberrn, J. 1991 . A soil health index, Journal of Soil and Water Conservation, Vol. , No. 1, pp.6. 71
Hackett, C. 1990. Crop environment matrix: a simple device for coarsely assessing the ecophysiological suitability of environmental for crop production, Tropic Agriculture (Trinidad), Vol .67, No. 2, pp. 159-163 . Hakansson, I. 1994 . Soil tillage for crop production and for protection of soil and environmental quality: a Scandinavian viewpoint, Soil and Tillage Research, Vo1.30, No.2, pp. 109-124 . Halvorson, J. J., Smith, J. L. and Papendick, R.I. 1997 . Issues of scale for evaluating soil quality, Journal of Soil and Water Conservation, Vo1.52, No . l, pp. 26-30. Hamblin, A. 1991 . Sustainable agricultural systems: what are the appropriate measures for soil structure? Aust. J. Soil Res. Vol. 29, pp.709-715 . Harris, R.F., Marlen, D.L. and Mulla, D.J. 1997 . A conceptual framework for assessment and management of soil quality and health, pp. 61-82, In J.W. Doran an A.J. Jones (ed), Methods for assessing soil quality. Special Publication 49, SSSA, Madison. WI, USA. Haskell, B .D. Norton, B .G. and Costanza, R. 1992. What is ecosystem health and why would we worry about it? pp. 3-19. In R. Costanza, B .G. Norton and B.D. Haskell (ed), 1992. Ecosystem health: new goals for environmental management, Island Press, Washington. D .C. USA. Hortensius, D. and Nortcliff, S . 1991 . International standardization of soil quality measurement procedures for the purpose of soil protection, Soil Use and Management, Vol.7, No.3, pp. 163-166 . Hummer, R .D. Astroth, J.H., Henderson, G.S . and Young, F.J. 1991 . Geographic information systems for soil survey and land-use planning, pp.243-270 . In M.J. Mausbach and L.P. Wilding (ed), Spatial variabilities of soil and landforms, SSSA Special Publication No. 28, SSSA, Inc. Madison, USA, Johnson, D.L. 1985. Soil thickness processes, pp. 29-40. In P.D.Jungerius (ed), Soil and Geomorphology, Catena Supplement 6. Johnson, L. B. 1990. Analysing spatial and temporal phenomena using geographical information systems: a review of ecological applications, Landscape Ecology, Vol. 4, No . 1, pp. 31-43. Karlen, D. L., Wollenhaupt, N.C., Erbach, D.C., Berry, E.C., Swan, J. B., Eash, N.S . and Jordahl, J. L.1994 . Long-term tillage effects on soil quality, Soil and Tillage Research, 1994, Vo1.32, No.4, pp.313-328 . Karlen, D. L., Mausbach, M. J., Doran, J.W., Cline, R.G., Harris, R.F., Schuman, G.E. 1997. Soil quality: a concept, definition, and framework for evaluation, Soil Science Society of America Journal, Vol .61, No. l, pp. 4-10 Larson, W.E. and Pierce, F.J. 1991 . Conservation and enhancement of soil quality. pp. 175-203 . In Evaluation for sustainable land management in developing world, Vol 2: technical papers. IBSRAM Proceedings No. 12 (2), International Board on Soil Research and Management, Bangkok, Thailand . Larson, W.E . and Pierce, F.J. 1994 . The dynamics of soil quality as a measure of sustainable management. pp. 37-51 . In J.W. Doran, D.C. Coleman, D.F. Bezdicek, and B.A. Stewart (ed), Defining soil quality for a sustainable environment . SSSA Special Publication No. 35, SSSA, Inc. Madison, USA. Lal, R. 1998. Soil quality and sustainability, pp. 17-30. In R. Lal, W.H. Blum, C. Valentine and 72
B .A. Stewart (ed) Methods for assessment of soil degradation, Advances in soil science, CRC Press, New York. Lal, R., Hall, G.F. and Miller, F.P. 1989 . Soil degradation : l. basic processes, Land Degradation & Rehabilitation, Vol 1, No. l, pp. 51-69. MacDonald, K.B . and Moore, A. 1991 . Preliminary assessment of the elements of soil quality based on data in the soil landscape of Canada digital maps, SQEP-2 Discussion Paper # 1, CLBRR, Ottawa, Canada. MacDonald, K.B . 1991 . Soil quality susceptibility to change : impact of soil modifying processes, SQEP-2 Discussion Paper # 2, CLBRR, Ottawa, Canada. MacDonald, K.B. and M. Brklacich, 1992 . Prototype agricultural land evaluation systems for Canada: I. Overview of systems development, Soil Use and Management, Vol. 8, No. 1, pp. 1-8 . MacDonald, K.B . and Valentine, K.W.G. 1992 . CanSIS/NSDB : A general description, Centre for Land and Biological Resource Research, Agriculture Canada, Ottawa, Canada. MacDonald, K.B ., Wang, F., Jarvis, I. and Fraser, W. 1993 . GIS procedures to assess soil quality at regional and national scales, pp. 325-33 . In Proceedings of GIS'93 Symposium, Vol. 11, Vancouver, British Columbia, Canada . MacDonald, K.B., Wang, F., Fraser, W.R., and Lelyk, G.W., 1994 . GIS-based system to assess soil quality . Chapter 3, pp. 1-26 In D.F. Acton (ed), A program to assess and monitor soil quality in Canada. Soil Quality Evaluation Program Summary. Centre for Land and Biological Resources Research, Research Branch, Agriculture and Agri-Food Canada, Ottawa, ON, Canada. MacDonald, K.B ., Fraser, W.R., Wang, F., and Lelyk, G.W., 1995, A geographical framework for assessing soil quality, pp. 19-30. In D.F. Acton and L.J. Gregorich (ed) The health of our soils -toward sustainable agriculture in Canada. Centre for Land and Biological Resources Research, Research Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada. MacDonald, K.B ., Wang, F., Fraser, W.R., Jarvis I.E. and Lelyk, G.W., 1996, Assessment of Soil Quality in Canada - ISQ96 Summary, Internal report, Ontario Land Resource Unit, Greenhouse and Processing Crops Research Centre, Guelph, Ontario, Canada MacDonald, K.B ., Wang, F. and Fraser, W. 1997 . Soil quality susceptibility to change - an assessment of soil quality change at national and regional scales (submitted to Land Degradation and Development) . Mathur, S. P. and Wang, C. (ed). 1991 . Soil quality in the Canadian context-1988 discussion papers, Technical Bulletin, 1991-1E, Research Branch, Agriculture Canada. Ottawa, Ontario, Canada. Nowland, J.L. 1987 . Canadian concerns for soil quality, pp. 11-22. In D.W. Anderson (ed), In search of soil conservation strategies in Canada, Agricultural Institute of Canada. Nowland, J.L. and Halstead, R.L. 1986 . Land: a fragile resource. Canadian Committee on Land Resource Services. Research Branch, Agriculture Canada, Publication 5221B . Ott, R.W. 1978 . Environmental indices : theory and practice . Ann Arbor Science Publishers Inc . Michigan, USA. pp. 66-90 . Parr, J.F., Papendick, R.I., Hornick, S.B. and Meyer, R.E. 1992. Soil quality : attributes and relationship to alternative and sustainable agriculture. Journal of Alternative Agriculture 73
7 :5-11 . Pettapiece, W.W. (ed), 1995, Land Suitability Rating System (LSRS) for Agricultural Crops 1 . Spring-seeded small grains Technical Bulletin, 1995-6E, Research Branch, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada. Pennock, D. J., Anderson, D. W., Jong, E.de, 1995 . Landscape-scale changes in indicators of soil quality due to cultivation in Saskatchewan, Canada, Geoderma, Vol.64, No. l, p.1-20. Rapport, D.J. 1992. Evaluating ecosystem health. Journal of Aquatic Ecosystem Health. Vol. 1, No . 1, pp. 15-24. Rhoton, E .E., Lindbo, D. L. 1997. A soil depth approach to soil quality assessment, Journal of Soil and Water Conservation, 1997, Vol . 52 No. 1, 66-72. Rodale Institute, 1991 . Conference reports and abstracts for the International Conference on Assessment and Monitoring of Soil Quality, Allentown, PA. July 11-13, 1991 . Rodale Press, Emmaus. PA. USA . Roming, D .E., Garlynd, M.J., Harris, R.F. and McSweeney, K. 1995. How farms assess soil health and quality, Journal of Soil and Water Conservation, 1997, Vol . 50, No. 3, pp.229236. Roming, D .E., Garlynd, M.J. and Harris, R.F. 1997. Farm-based assessment of soil quality : a soil health scorecard, pp. 39-60, In J.W. Doran an A .J. Jones (ed) Methods for assessing soil quality . Special Publication 49, SSSA, Madison. WI, USA. Sarrantonio, M., Doran, J.W., Leibig, M.A. and Halvorson, J .J. 1997 . On-farm assessment of soil quality and health, pp. 83-105, In J.W. Doran an A .J. Jones (ed) Methods for assessing soil quality. Special Publication 49, SSSA, Madison . WI, USA. Schaeffer, D. J., Herricks, E. and Karster, H. 1988. Ecosystem health I : measuring ecosystem health, Environmental Management, Vol. 12, No. 4, pp. 445-455 Science Council of Canada . 1986. A growing concern: soil degradation in Canada, Science Council of Canada, Ottawa, Ontario, Canada Simard, R. R., Angers, D. A., Lapierre, C, 1994. Soil organic matter quality as influenced by tillage, lime, and phosphorus, Biology and Fertility of Soils, Vol. 18, No. 1, p.13-19. . Sims, J. T., Cunningham, S. D . and Sumner, M. E. 1997. Assessing soil quality for environmental purposes : roles and challenges for soil scientists, Journal of Environmental Quality, Vo1.26, No. l, pp.20-25. Singer, M.J. and Warkentin, B.P. Soil in an environmental context : an America perspective, Catena, Vol 27, No 3-4, pp. 179-189 . Shields, J.A . and Nowland, J.L.1975 . Additional land for crop production: Canada. pp. 45-60. In Proceedings of the 30th Annual meeting of the Soil Conservation Society of America, August, 10-13, 1975, San Antonio, USA. Shields, J.A., Tamocai, C., Valentine, K.W.G. and MacDonald, K.B . 1991. Soil landscapes of Canada - procedures manual and users handbook . Agriculture Canada Publication 1868/E . 74 pp. Thomasson, A .J. and Jones, R.J.A. 1989 . Land evaluation at regional scale, pp. 231-240 . In J. Bouma and A.K. Bregt (ed). Land quality in space and time, Proceedings of ISSS Symposium, Wageningen, Pudoc ., Wageningen. The Netherlands . van Diepen, C.A., van Keulen, H., Wolf, J. and Berkhout J.A.A. 1991 . Land evaluation : from 74
intuition to quantification, pp. 139-204 . In B .A. Stewart (ed). Advances in soil science Vol. 15. Springger-Verlag, New York. Varallyay, G., Scharpenseel, H.W. and Targulian, V.0.1990. Type of soil processes and changes, pp. 41-46. In R.W.Amold, I. Szabolcs and V.O.Targulian (ed), Global soil change, Int. Inst. for Applied System Analysis, Laxenborg, Austria . Wang, C., Coote, D.R. and Acton, D.F. 1991 . A proposed mineral soil quality classification system for arable land, pp.54-62. In S.P. Mathur and C.Wang, (ed) . Soil quality in the Canadian context--1988 discussion papers, Technical Bulletin, 1991-1E, Research Branch, Agriculture Canada, Ottawa, Ontario, Canada. Wardle, D.A. and Reganold, J. P. 1994. Statistical analyses of soil quality, Science - AAAS Weekly Paper Edition - including Guide to Scientific Information, Vo1.264, No.5156, pp.281-292 . Warkentin, B . P, 1995 . The changing concept of soil quality, Journal of Soil and Water Conservation, 1995, Vo1.50, No.3, p.226-228 .
APPENDIX 1. KEY TERMS AND ACRONYMS AFFC - Agriculture and Agri-Food Canada AML - Arc Macro Language (Arc/Info) Arc/Info - A commercial GIS produced by Environmental Systems Research Institute, CA, USA AVHRR - Advanced Very High Resolution Radiometer CanSIS - Canadian Soil Information System CoA - Census of Agriculture (from Statistics Canada) CLI - Canada Land Inventory CMP - SLC Component File (CanSIS/NSDB) DBMS - Database Management System DSM - Detailed Soil Survey Map (CanSIS/NSDB) EGDD - Effective Growing Degree Days GIS - Geographic Information System GPCRC - Greenhouse and Processing Crop Research Centre INFO - DBMS component of Arc/Info . ISQ - Inherent Soil Quality LSRS - Land Suitability Rating System (Pettapiece et al, 1995) NSCP - The National Soil Conservation Program (Canada) NSDB - National Soil Data Base (Canada) Pamap - A commercial GIS produced by PCI Pacific GeoSolutions Inc., Victoria, Canada P-PE - Moisture Index (Precipitation-Potential Evapotranspiration) SC - Soil Carbon (database, CanSIS/NSDB) SLC - Soil Landscape of Canada (map, CanSIS/NSDB) SLF - Soil Layer File (CanSIS/NSDB) SMUF - Soil Map Unit File (CanSIS/NSDB) SNF - Soil Names File (CanSIS/NSDB) SQC - Soil Quality Change SQEP - The Soil Quality Evaluation Program (under NSCP) SQS - Soil Quality Susceptibility to change
Appendix 2. CanSIS NSDB Data Structures and Definitions This appendix lists the data structures and data field names for the Canadian Soil Information System National Soil Database files referenced in this report . Data field types are C (Character), D (Date), I (Integer), or N (Numeric) . Data fields used to link to other relational files are indicated (key). Updated information is available from the CanSIS/NSDB web site at http://res .agr.ca/ecorc/program3/cansis/. Table A2-1. SLC (version 1 .0) DONDnant and SUBdominant files attributes DOM Name SUB Name PROVINCE PROVINCE POLYNUMB POLYNUMB DOMKDMAT SUBKDMAT DOMDISTR SUBDISTR GRIDLOCN GRIDLOCN DOMREGFM SUBREGFM DOMLOCSF SUBLOCSF DOMSLOPE SUBSLOPE DOMPMDEP SUBPMDEP DOMPMTEX SUBPMTEX DOMDEVEL SUBDEVEL DOMSRFTX SUBSRFTX DOMCFRAG SUBCFRAG DOMROOT SUBROOT DOMCMPLR SUBCMPLR DOMCMPDP SUBCMPDP DOMDRAIN SUBDRAIN DOMAVWAT SUBAVWAT DOMWATAB SUBWATAB DOMICETYSUBICETY DOMICECT SUBICECT DOMPERMA SUBPERMA DOMACTLR SÜBACTLR SUBPATGD DOMPATGD DOMPHCAL SUBPHCAL DOMPHWAT SUBPHWAT DOMORGAN SUBORGAN DOMNITRO SUBNITRO DONII-IUMLR SUBHLJMLR SUBCALCA DOMCALCA DINCLUS 1 SINCLUS 1 DINCLUS2 SINCLUS2 SUBVEGET DOMVEGET
Type C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C C
Width 3 4 2 3 3 1 3 1 2 4 1 4 1 3 1 1 1 1 1 I 1 1 3 2 2 2 2 1 1 1 2 2 2
Attribute Definition provincial code and ma sheet number (key) polygon number (key) kind ofrock outcrop or other material at the surface percentage distribution of rock or other surface material 'd code for locating polygons regional landform local surface form sloe gradient class parent material mode ofdeposition parent material texture soil development surface texture of mineral soil to 15 cm coarse fragment content in control section rooting depth compacted, consolidated or contrasting layer depth to compacted, consolidated or contrasting layer drainage class available water capacity in upper 120 cm. depth to water table ice type ice content permafrost occurrence active layer depth in permafrost soils patterned and kind H of upper 15 cm of soil - CaCl2 H of upper 15 cm of soil - water organic carbon ofupper 15cm nitrogen content of upper 15 cm, % b weight thickness ofhumus layer calcareous class ofparent material soil inclusion 1 soil inclusion 2 vegetative cover and/or land use 77
DOMLAKE DOMWATBD DOMRELIA DOMCOMPL DOMNAMEI DOMNAME2 DOMTEXGP AREAKHA
SUBLAKE SUBWATBD SUBRELIA SUBCOMPL SUBNAME1 SÜBNAME2 SÜBTEXGP AREAKHA
C C C C C C C N
1 1 1 1 6 6 2 7
lake size water bodies, percentage coverage of polygon reliability class complexity class soil name 1 soil name 2 parent material texture group area ofpolygon (kilohectares)
Table A2-2. SLC (version 1 .0) Component (CMP) file attributes Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Name POLYNUMB COMPNT NUMB PERCENT KINDMAT VEGET PMDEP CFRAG ROOTDP DRAIN DEVEL CALO LOCSF SLOPE SOILCODE MODIFIER
Type I C C 1 C C C C C C C C C C C C
Width 4 1 1 3 2 2 2 1 3 1 1 1 3 1 3 3
Attribute Definition polygon number (key) component (key) inclusion number (key) percent occurrence kind ofmaterial vegetation/land use mode of deposition coarse fragment content rooting depth drainage soil development calcareous class local surface form slope gradient soil name code (key) soil name modifier (key)
Table A2-3. SLC (version 1.0) Carbon Layer (CLYR) file attributes
r
Number 1 2 3 4 5 6 7 8 9 10 11 12
Name PROV SHEETNO POLYNUMB COMPNT NUMB LAYERNO LAYER THICK THICK-ME TEXTR TEXTR-ME BDENS
Type C 1 1 1 I I C 1 C C C N
Width 2 2 4 2 2 1 3 3 1 4 1 4.2
78
Attribute Definition
province (key) ma sheet number (key) polygon number (key) component (key) inclusion number (key) layer number (key) lay zon designator thickness thickness - reliability texture texture - reliability bulk density
13 14 15
BDENS-ME OCARB OCARB-ME
Î
C N C
T
1 4 .1 1
bulk density - reliability organic carbon organic carbon - reliability
Table A2-4. DSM Soil Map Unit File (SMUF) attributes Number I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Name PROVINCE MAPUNITNOM SOIL_CODE1 MODIFIERI EXTENT1 SOILCODE2 MODIFIER2 EXTENT2 SOILCODE3 MODIFIIER3 EXTENT3 SLOPEPI SLOPEP2 SLOPEP3 STONE1 STONE2 STONE3 DATE
Type C C C C N C C N C C N N N N C C C D
Width 2 60 3 3 3 3 3 2 3 3 2 5.1 5 .1 5 .1 1 1 1 8
Attribute Definition province (key) ma unit name (key) soil name code of soil 1 (key) soil name modifier of soil 1 (key) extent of soil I soil name code ofsoil 2 (key) soil name modifier of soil 2 (key) extent of soil 2 soil name code of soil 3 (key) soil name modifier of soil 3 (key) extent of soil 3 sloe stee ness of soil 1 (percent) sloe stee ness of soil 2 (percent) sloe stee ness of soil 3 (percent) stoniness of soil 1 stoniness of soil 2 stoniness of soil 3 date ofrevision
Width 2 24 3 3 1 1 2 1 2 2 4 4 4 2 4
Attribute Definition
Table A2-5. Soil Name File (SNF) attributes Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Name PROVINCE SOILNAME SOIL CODE MODIFIER LU KIND WATERTBL ROOTRESTRI RESTR TYPE DRAINAGE MDEP1 MDEP2 MDEP3 ORDER SGROUP
Type C C C C C C C C C C C C C C C
79
province (key) soil name soil name code (key) soil name modifier (key) land use type (key) kind of material water table presence root restricting layer number root restricting drainage class parent material mode of deposition parent material mode of deposition parent material mode of deposition soil classification - Order soil classification - Sub-group
16 17 18
GGROUP PROFILE DATE
C C D
3 14 8
soil classification - Great Group Detail II file header (key) date ofrevision
Table A2-6. Soil Layer File (SLF) attributes Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Name PROVINCE SOILCODE MODIFIER LU LAYER-NO HZN_LIT HZN MAS HZN_SUF HZN_MOD UDEVTH LDEPTH COFRAG COFRAG# DOMSAND VFSAND VFSAND# TSAND TSAND# TSILT TSILT# TCLAY TCLAY# ORGCARB ORGCARB# PHCA PHCA# PH2 PH2# BASES BASES# CEC CEC# KSAT KSAT# KPO
Type C C C C C C C C C N N N 1 C N 1 N 1 N I N 1 N 1 N I N 1 N I N 1 N I N
Width 2 3 3 1 1 1 3 5 1 3 3 3 3 2 3 3 3 3 3 3 3 3 5 .1 3 4.1 3 4.1 3 3 3 3 3 7.3 3 3
80
Attribute Definition province (key) soil code (key) modifier (key) land use (key) horizon number horizon litholo 'cal discontinuity master horizon horizon suffix horizon modifier upper depth lower depth coarse fragments dominant sand fraction very fine sand total sand total silt total clay organic carbon H in calcium chloride PH as per project report base saturation cation exchange capacity saturated hydraulic conductivity water retention
@ 0 kP
1 36 KPO# 3 37 KP10 N 3 water retention @ 10 kP 1 38 KP10# 3 39 KP33 N 3 water retention @ 33 kP KP33# 1 40 3 N water retention @ 1500 kP 41 KP1500 3 42 KP1500# 1 3 BD N 5 .2 bulk density 43 BD# 1 44 3 45 EC N 3 electrical conductivity 46 EC# 1 3 CAC03 N 2 calcium carbonate equivalent 47 48 CAC03# 1 3 49VONPOST N 2 Von Post estimate ofdecomposition 1 50 VONPOST# 3 51 WOOD N 2 wood material (percent b volume) 52 WOOD# 1 3 53 DATE D 8 date of revision Note: A field name with a trailing # indicates the number ofobservations used in determining the value. These fields are optional and may not always be present. A code of zero (0) indicates an estimated value.
Appendix 3. A Detailed Description of ISQ92 Rating Procedures The ISQ92 procedures were developed using Arc/Info GIS and SLC dominant (DOM) and subdominant (SUB) attribute files (Table 3-1 and Table A2-1). ISQ92 has four basic components ; 1) Exclusion procedures (modified in 1996), 2) Rating system and criteria, 3) Algorithms and programs, and 4) a Arc Macro Language Graphic user interface (added in 1996) . Exclusion procedures: As the current focus is on the inherent soil quality for crop production, polygon areas with unsuitable climatic and soil conditions for annual cereal crops are excluded (Table A3-1). This was done using the Arc/Info `RESELECTION' functions, prior to running the ISQ92 rating programs . Table A3-1 . Areas or soils excluded and the thresholds used in ISQ92 procedures Areas or soils excluded
SLC attributes and thresholds*
Inadequate climate for annual crops
EGDD (effective growing degree days) < 1050
Slopes too steep for crop production
DOM/SUBSLOPE (slope) > 30% (Class D, E and F)
Non mineral soil types
DOM/SUBKDMAT (kind of surface materials, such as rock, organic, mineral, water, etc .) = `OR' (organic)
Very poor surface drainage
DOM/SUBDRAIN (drainage class) = `V' (very poor)
Urban areas
DOM/SUBKDMAT = 'UR' (urban area)
Rock
DOM/SUBKDMAT= `Rl' or `R2' or `R3' (rock)
Water bodies DOM/SUBKDMAT ='WA' (water) * except EGDD, all attributes are from the SLC extended legend DOM and SUB files
Rating system and criteria Given the generalized nature of the SLC extended legend database, and the broad general nature of the SLC polygons, a four class system for inherent soil quality was considered appropriate. Accordingly, a inductive rating structure was adopted, involving a 3 class rating for each data base attribute. These were combined to calculate a 4 class 'element rating'.
Each class is assigned numeric points for quantification (Table A3-2). The integration from 'attribute rating' to 'element rating' is an additive one: ISQE
n
=E ISQA~j 11=1
Where ;
Table A3-2. Point system of ISQ92 rating ISQ attribute rating
(A3-1)
ISQEj is the total rating points of ISQ element j (j = 1,2,3, 4 for the 4 defined ISQ elements for crop production) .
Class
Point
Poor
2
Moderate
1
Good
0 ISQ element rating
ISQA;; is the rating points of the ISQ attribute i selected for ISQ element j (i = 1, 2,3, ..., n, normally 3 to 4 attributes selected for rating each ISQ element)
Class
Point
Poor
>= 3
Moderate to poor
2
Good to moderate
1
Good
0
The criteria for rating selected inherent soil quality attributes by the ISQ92 procedure are described in Table A3-3 . The SLC DOM and SUBDOM extended legend data fields considered for each of the four ISQ element are indicated, along with the data values in each of the 3 rating classes. The database fields and critical attribute values were selected based on their estimated contribution to the defined ISQ element. ISQ element definitions and ratings were limited by the available data fields, and the generalized nature of the individual data field values. Algorithms and programs The ISQ92 point system (Table A3-2) and rating criteria (Table A3-3) are implemented with a set of INFO programs which can be run on any Arc/Info system (version 5 .0 or later) with slight modification to specify the path of input data. Graphic user interface The early version of ISQ92 does not have a graphic user interface, and the output ratings of ISQ elements were directly used by Arc/Info's Arcplot module for spatial query, display and mapping . While adequate for experienced Arc/Info users, a more intuitive graphic user interface was desirable for routine operation and demonstration of the rating system. Therefore, a menu based interface was added, using Arc Macro Language (AML).
83
Table A3-3. The rating criteria of selected ISQ attribute used in ISQ92 procedures
ISQ Element
Rating criteria and points
ISQ attribute Root depth (DOMROOT/SUBROOT*)
Available Porosity
Depth to compact layer (DOMCMPDP/SUBCMPDP) Water availability in upper 120 cm (DOMAVWAT/SUBAVWAT)
Nutrient Retention
Chemical Rooting Conditions
Moderate (1)
Good (0)
< 20 cm
20 to 75 cm
> 75 cm
nla
50 cm
150 mm
Surface texture group and texture, organic carbon content (DOMTEXGPISUBTEXGP, DOMSRFTX/SÜBSRFTX and DOMORGAN/SUBORGAN)
If the texture group is loam and no coarse fragments exist in the surface layer and there is more than 2% organic carbon, deduct 1 point from the summed points ofthe ratings of above three attributes
Surface texture (DOMSRFTX/SUBSRFTX)
cobbly or gravelly or with high silt content
ifloam or clay are found in significant amounts yet high silt or coarse fragments exist.
= 2% but < 3%
>= 3%
< 20 cm
20 to 75 cm
> 75 cm
Depth to compact layer (DOMCMPDP/SUBCMPDP)
n/a
50 cm
Surface texture (DOMSRFTX/SUBSRFTX)
cobbly or gravelly with high silt content
if loam or clay are found in significant amounts yet high silt or coarse fragments are exist.
Soil development (DOMDEVEUSÜBDEVEL)
Gray Luvisol
n/a
All others
a saline soil
n/a
All others
Surface organic carbon content (DOMORGAN/SUBORGAN) Physical Rooting Conditions
Poor (2)
Root depth (DOMROOT/SUBROOT)
Water availability in upper 120 cm (DOMAVWAT/SUBAVWAT) pH ofsurface layer measured in water (DOMPHWAT/SUBPHWAT)
anything else.
>= 4.5 but 5.5 and or = 8.0 but 8.0
anything else.
Appendix 4. A Detail Description of ISQ94 Rating Procedures
The ISQ94 procedures for assessing inherent soil quality were developed to take advantage of the revised/ improved structure of attribute data associated with the SLC CMP files . ISQ94 procedures are developed in dBASEIV, and utilize databases that are also in dBASE format. The procedures described here use the SLC Component (CMP) file for national and broad regional assessments, and the Soil Map Unit File (SMUF) for detailed regional assessments . In both files, digital map polygons are described in terms of soil components and areal extent. These are linked in each case to the same set of modal soil attributes stored in Soil Names and Soil Layer files . For further information concerning the ISQ94 rating program, contact Mr. W. Fraser, Land Resource Unit, Brandon Research Centre. ISQ94 pre-rating procedures: The pre-rating procedures include data checking, selection of files and attributes to be used, exclusion and inclusion of areas/soils, and preparation of interim files to store the output ratings and statistics . The program performs an initial check to verify the location and structure of the CMP/SMUF, SNF, and SLF databases. All data fields used by the ISQ94 ratings program must be present before the main ISQ94 procedures can be run (see Table A4-1) . It is also important that the data fields contain actual data values for all soil components in the digital maps. National Soil Databases may have data deficiencies, indicated by "-9" values for numeric fields. The ISQ94 routines are programmed to skip over such records. It is highly recommended that NSDB data checking programs, such as INFOCHEKEXE, be run on all soil databases before they are used by ratings programs such as ISQ94. The ISQ94 program employs a "CROP.LOG" file to store standard exclusion and ratings threshold values. These are used by the program as input variables for various ratings calculations . All "CROP .LOG" threshold values may be reviewed, altered and stored as alternative sets of program values prior to running the actual ratings program. This is useful for sensitivity analysis testing, and development of different crop parameters for additional crop types. Final output ratings from the ISQ94 program are stored in a dBASE file ("RATINGS.dbf'), including a copy of the original input data file (the SLC CMP file or SMUF file) . A matching ASCII "History" file ("RATINGS .his") provides a summary of the ISQ ratings statistics for the map area, as well as all "CROP.LOG" settings and databases used by the ISQ94 program .
Table A4-1. Data attributes used in ISQ94 procedures* Database and attribute fields
ISQ Procedure where the attribute is used data link
exclusion
Available porosity
Nutrient retention
Physical rooting conditions
Chemical rooting conditions
1. SLC Carbon Component file (CMP) - broad scale POLYNUMB POLY AREA COMPNT NUMB PERCNT SOIL-CODE MODIFIER
2. SLC Climatic file (.CLM)** - broad scale POLYNUMB EGDD P-PE
3. DSM Soil Map Unit File (SMUF) - detail scale POLY AREA SOIL_CODE1,2,3 MODIFIER1,2,3 EXTENT1,2,3 SLOPEP1,2,3 STONE1,2,3
4. Soil Names File (SNF) - both broad and detail scale SOIL CODE MODIFIER LU KIND ROOTRESTRI RESTRI TYPE DRAINAGE ORDER S_GROUP GLGROUP
5. Soil Layer File (SLF) - both broad and detail scale SOIL_CODE MODIFER LU LAYER NO UDEPTH LDEPTH COFRAG VFSAND TSAND TSILT TCLAY ORGCARB PHCA BASES CEC KP33 BD EC
0
* See appendix 2 for details of definition of each attribute or check CanSIS/NSDB web site at http://res.agr.ca/ * *Not official NSDB file. 86
Table A4-2. Areas or soils excluded and the thresholds used in ISQ94 procedures Broad-scale
Detail-scale
SLC CMP, SNF attributes and thresholds
SMUF, SNF attributes and thresholds*
Areas with inadequate heat for annual crops
EGDD (effective growing degree days) < 1050
EGDD (effective growing degree days) < 1050
Slopes too steep
SLOPE > 30%
SLOPEP1,2,3 > 30%
Soil surface too stony
n/a (no data)
STONE1,2,3 (stoniness) >= 3
Organic soil
ORDER ='OR'
ORDER = `OR'
Very poor surface drainage
DRAINAGE (drainage class) ='VP'
DRAINAGE (drainage class) ='VP'
Urban areas
KIND = `U' (Unclassified)
KIND = `U' (Unclassified)
Rock
KIND = `N' (Non soil)
KIND = `N' (Non soil)
Water bodies
KIND = `N' (Non soil) and corresponding SOII._CODE
KIND = `N' (Non soil) and corresponding SOIL_CODE
Areas or soils excluded
* except for EGDD and SLOPE, all attributes are from the SNF .
The exclusion procedures in ISQ94 include two parts ; general and layer exclusions . The criteria of general exclusion (Table A4-2) are generally similar to ISQ92 . Note that ISQ94 uses Soil Names File data fields for most exclusion criteria, while ISQ92 uses the Soil Carbon Component File data fields. The ISQ94 layer exclusions are defined as a set of root restrictions for annual crop production, such as extremely high bulk density, salinity and extreme pH (Table A4-3). The thresholds for excluding layers with root restrictions are crop specific and can be modified in the CROPIOG file.
Table A4-3. Root restrictions and thresholds for layer exclusions Rooting conditions
Thresholds for exclusion
Rooting depth
Max. 80 cm Min . 40 cm =1.75 g/cm3
Maximum bulk density (Clay >= 40%)
>=1 .50 g/cm'
Maximum salinity (EC)
>= 16 mS/cm
Minimum pH
=10
Root restricting layers
Duric, Ortstein, Placic and Fragipan
Table A4-4. ISQ94 program variables for ISQ element rating ISQ element Available porosity
Nutrient retention
Physical rooting conditions
Chemical rooting conditions
Variable
Definition
dimension
AIR TOT
Accumulated air-filled porosity from surface to the maximum allowable rooting depth
cm
AWHC TOT
Accumulated available water holding capacity from surface to the maximum allowable rooting depth .
cm
ISQ_AIR
Rating ofISQ aeration porosity sub-element by looking up the matrix ofAIR_TOT and moisture index
arbitrary points
ISQ_AWHC
Rating of ISQ-AWHC sub-element by looking up the matrix ofAWHC_TOT and moisture index .
arbitrary points
ISQ_PORO
Final rating ofAvailable Porosity element by taking the more restricting rating ofISQ AIR subelement and ISQ AHWC sub-element
arbitrary points
NUTR_SUR
Accumulated nutrient retention capacity oftop 20 cm (for algorithm, see Formula 3)
meq/cm2
NUTR_RATIO
The ratio of NUTR-SUR to thickness, i.e. NUTR RATIO = NUTR SUR / 20
meq/cm3
ISQ NUTR
Final rating of Nutrient Retention element based on NUTR_RATIO
arbitrary points
THICK TOT
Accumulated thickness from top to the maximum allowable rooting depth.
cm
ISQ ROOT
Final rating of Physical Rooting Conditions element based on THICK TOT
arbitrary points
SURPH
Depth weighed average pH oftop 20 cm
pH scale
SUBPH
Depth weighed average pH from 20 to 80 cm
pH scale
SUREC
Depth weighed average EC of top 20 cm
mS/cm
SUBEC
Depth weighed average EC of 20 to 80 cm
ms/cm
ISQ_SURCHEM
Rating of surface chemical rooting conditions by taking the `worse' rating of SURPH and SUREC
arbitrary points
ISQ-SUBCHEM
Rating of sub surface chemical rooting conditions by taking the `worse' rating of SUBPH and SUBEC
arbitrary points
ISQ_CHEM
Final rating of Chemical Rooting Conditions element based on the matrix ofISQ SURCHEM and ISQSUBCHEM
arbitrary points
ISQ94 Rating Procedures The rating procedures consist offour sub-procedures, each corresponding to one ofthe four elements of inherent soil quality (Table 2-1). The specific criteria, algorithms, and calculation procedures for each element are given in the following sections. Program variables were used to store intermediate values calculated from particular procedures and algorithms (Table A4-4). Threshold values are used to classify the calculated program variables, and produce ratings for each ISQ element . An examination ofprogram variables is also useful for error checking and program debugging. 1) Available porosity procedure Step 1 : Calculate the air-filled porosity for each soil horizon in the Soil Layer file, and accumulate this as a total aeration porosity value (in cm ofair). This is calculated in two stages; the surface layer (20 cm) and the subsurface (an accumulated value, "AIR TOT", for all horizons down to the maximum allowable rooting depth). The formula used is: n
AIR- TOT = E THICKNESS,*[(PDi - BDi) / PDi - 0.01* KP33i] i=1
where,
(A4-1)
AIR_ TOT= accumulated air-filled porosity (cm) from top to the maximum allowable rooting depth (horizon i = 1, ..., n). THICKNESS = the thickness ofthe soil layer (cm) i. This is the absolute value ofthe upper depth minus the lower depth, [ABS(UDEPTH-LDEPTH)], as recorded in the Soil Layer file. PD = Particle Density, g/cm~. A particle density (PD) of 2.65 g/cm' is used where the organic carbon of the mineral soil material is less than 2.0% (ORGCARB field in the SLF). Where the organic carbon values are between 2% and 17%, a particle density of2.54 g/cm' is used. For organic layers (organic carbon values of >17% ), a particle density of 1 .00 g/cm' is used. BD = Bulk Density, g/cm' . KP33 = % water, by volume, that is retained by the soil at 1/3 atmosphere suction . This is multiplied by 0.01 to convert from % to a ratio of the soil volume . KP33 approximates field capacity.
The total soil aeration value (AIR TOT, expressed in cm of air), is then added for all valid soil layers within the SURFACE LAYER. The default setting for SURFACE LAYER THICKNESS is 20 cm . A minimum threshold value of 1 cm of air within the surface layer (5% of the surface layer volume) must be reached; if this does not occur, the soil is considered as NOT RATED for 0 ISQ elements .
Table A45 . Matrix for determining rating points of ISQ aeration porosity Moisture Conditions (by drainage, soil-climatic zones and approx . P-PE range) Well drained Brown Chemozems, Solonetz (-200)
All poorly drained soils
> = 15.0
0
0
0
0
1
E
10.0 to 15.0
0
0
0
0
2
O
5.0 to 9.9
0
0
0
3
3
99
1.0to4.9
0
1
2
3
3
9
9
9
Overall Rating (ISQ_AWHC)
i
< 1 .0 (within 9 9 top 20 cm) 0 = good; 1 = good to moderate; 2 = moderate to poor; 3 = poor; 9 = not rated
Step2: Assign rating points to ISQ aeration porosity subelement (ISQ AIR) by looking up the matrix of AIR TOT and moisture index class, based on Soil Taxonomy and Drainage in the Soil Names File (Table A4-5). Step 3: Estimate the available water holding capacity (AWHC). The AWHC is calculated for each soil, from the total silt, clay and very fine sand values for each soil layer in the SNF file. This is multiplied by the horizon thickness (absolute value of UDEPTH-LDEPTH), and a total value (AWHC_TOT, in mm of total available water) is accumulated to either the maximum rooting depth, or some restricting layer above the maximum rooting depth. Values for AWHC are calculated as shown in Table A4-6. Step 4: Assign rating points to ISQ available water holding capacity sub-element (ISQ-AWHC) by looking up the matrix of AWHCTOT and moisture index (Table A4-7).
90
Table A4-6. Relationship between available water holding capacity and texture Texture'
0.5*VFSAND + TSILT +TCLAY2
AWHC (mmIM)
Organic
0
500
S
1-10
40
LS
11-20
60
SL
21-40
100
L, VFSL
41-60
150
SI, SIL
61-70
170
SICL, CL
71-75
180
SIL
76-80
190
SICL 81-85 200 'USDA soil texture classes indicated here are only approximate . Several classes may overlap due to varying percentages of silt and clay. 'For soils with coarse fragments, AWHC estimates are reduced by the volume percentage ofcoarse material, a Soil Layer File attribute . ' Organic soils are recognized by tsand + tsilt + tclay = 0 in the Soil Layer File data.
Step 5: Assign overall rating points to ISQ available porosity (ISQ-PORO). The `most limiting' principle is used, namely taking the worse rating of ISQ_AIR and ISQ_AWHC as the final rating point of ISQ^PORO. Table A4-7. Matrix for determining rating points ofISQ available water holding capacity . Moisture Conditions (drainage, soil-climatic zones and approx. P-PE range) Well drained Brown Chernozems, Solonetz (< -350)
Well drained Dark Brown Chernozems, Solonetz (-350 to -300)
Well drained Black, D. Gray Chem., Luvisols, Brunisols (-300 to -200)
All other well drained and imperfectly drained soils (> -200)
All poorly drained soils
15 .0 2 1 0 = good; 1= good to moderate; 2 = moderate to poor; 3 = poor ; 9 = not rated
2) Nutrient retention procedure Step 1: Calculate the cumulative cation exchange capacity (CEC) of all eligible soil horizons within the surface rooting depth (top 20 cm). The formula used is as follows : where, NUTR_ SUR =
n
CEC*BDi* BASESi*0.01* ELTHICKNESS,
(A4-2) . .
t=1
NUTRSUR = Accumulated nutrient retention capacity of surface 20 cm (meq/cm3) of all eligible soil horizons (i = 1, .. ., n) CEC = cation exchange capacity (meq/100g) BD = Bulk Density (g(cm3) BASES = base saturation (%), This is multiplied by 0.01 to convert from % to a ratio oftotal CEC . ELTHICIKNESS = the eligible thickness of the soil layer i considered as part oftop 20 cm surface soil. If a soil horizon has a lower depth that exceeds 20cm, only the portion to a depth of 20cm is considered eligible . Ifthe total horizon thickness is less than 20cm, the next horizon is also evaluated .
91
Step 2: Calculate the ratio of accumulated CEC of the top 20 cm: NUTR RATIO (meglcm3) = NUTR SUR (meq/cm2) / 20 (cm)
(A4-3)
Step 3: Assign rating points to ISQ Nutrient retention element based on the ratio of accumulated CEC of top 20 cm (NUTR_RATIO) . The generic rating scale, for cereal crops is shown in Table A4-8. The rating scale can be modified in the CRORLOG for other crops . Table A4-8. Rating scale of ISQ nutrient retention element NUTR_RATIO of Top 20 cm (megtcm')
< 8
8.0 to 8.9
9.0 to 15 .9
ISQ-NUTR 9 3 2 Ratings 0 = good; 1 = good to moderate; 2 = moderate to poor; 3 = poor; 9 = not rated
16 to 22
> 22
1
0
3) Physical rooting conditions procedure Step 1: Calculate the total depth of soil available to the crop roots (THICK-TOT) . Each successive soil layer is examined, starting with the surface, until a physical root restricting condition is encountered . A number of standard checks are made by the ISQ program to determine if each successive soil layer in a given soil is root restricting or not. These include checks for recognized root restricting layer types . Further checks are made on each successive soil horizon for bulk density, salinity and pH threshold conditions . If any such conditions are encountered, processing stops at the upper depth of this horizon, and the total thickness (THICK TOT) of the rooting zone is calculated accordingly . Step 2: Assign rating points to ISQ physical rooting condition element based on THICK TOT calculated from step l, as shown in Table A4-9. Table A4-9. Rating scale ofISQ physical rooting conditions element THICK-TOT (cm)
< 20
ISQ_ROOT 9 Rating at'0& 1 = gnmi tn mndPmtP'
20 to 29
30 to 54
55 to 79
>= 80
3
2
1
0
7 = TnMPrAtP tn
poor., 3 = -poor&, 0 - nnt
92
rntari
4) Chemical rooting conditions procedure Chemical conditions are defined in ISQ94 in terms of an adequate depth of soil with pH (PHCA field in the SLF) and salinity (Electrical Conductivity, EC field in the SLF) within tolerable limits. Surface (top 20 cm) and subsurface (20 to 80 cm) conditions for both pH or salinity can vary considerably for each soil, and are therefore assessed separately. Where the rooting zone is restricted to less than 80cm by adverse soil conditions, the chemical conditions are evaluated within the restricted rooting zone. Chemical conditions are considered more limiting within the surface zone for annual crop production . The final rating of pH and salinity is determined by combining both surface and subsurface ratings. Both pH and salinity ratings are assessed in a similar fashion, and the most restricting of the two conditions is used in the overall ISQ chemical rooting element rating . Step 1 : Calculate average (depth weighted) pH and EC of surface layer (top 20 cm), i.e. SURPH and SUREC respectively : n
SURPH=E PHCAi * ELTHICKNESSil20)
(A4-4)
n SUREC=I, ECi * ELTHICKNESS i 20)
(A4-5)
i=1
where,
SURPH = depth weighted average pH of surface 20 cm of all eligible soil horizons (i = 1, ..., n) PHCA = pH in Calcium chloride as recorded in SLF SUREC = depth weighted average EC of surface 20 cm (mS/cm) of all eligible soil horizons (i = 1, ..., n) EC = electrical conductivity as recorded in the SLF (mS/cm) ELTHICKNESS = the eligible thickness of the soil layer i (See explanation of formula (3)
Step 2: Assign rating points to ISQ SURCHEM based on SURPH and SUREC respectively and then take the most restricting rating as the final rating of surface chemical rooting conditions (Table A4-10). Table A4-10. Rating scale of ISQ chemical rooting conditions element
SURPH or SURPH SUREC or SUBEC (MS/CM)
9.5
4.0to5 .0 or 8.1 to 9.5
5.0to5 .5 or 7.7 to 8.1
5 .5to6.0 or 7.3 to 7.7
6.0 to 7 .3
> 12.0
8.1 to 12.0
4.1 to 8.0
2.1 to 4.0
