c insitu c calc c

1 downloads 0 Views 734KB Size Report
Haltrin, Vladimir I. "One-parameter two-term Henyey-Greenstein phase function for light scattering in ... Leymarie, Edouard, David Doxaran, and Marcel Babin.
Improved correction method for measurement of inherent optical properties in natural waters

*Indian Institute of Technolgy Madras, Chennai 600036, Tamil Nadu, India

Poster ID: 128, RSPSoc 2017 Sayoob Vadakke-Chanat* and Palanisamy Shanmugam* Remote Sensing & Photogrammetry Society

Abstract: Light scattering and absorption due to dissolved and particulate components in natural waters are crucial for accurate observations of remotely sensed ocean color. Light

attenuation is an important parameter useful in various biogeochemical models and algorithms employed in remote sensing of the natural waters. Furthermore, the slope of the attenuation spectrum is useful in calculating the particle size distribution in natural waters which have substantial implications for ocean color remote sensing. The measurement of attenuation spectra is commonly achieved by employing absorption-attenuation meters (eg, WETLabs AC9, ACS). The difficulty in separating narrow forward scattered light from the transmitted light imposes significant uncertainty and errors in the attenuation measurement which can be as high as 20%. Though this error is widely acknowledged by the research community, a consensus on a correction procedure for the error caused remains a challenging issue. A new model based on simulated datasets and in situ measurements to correct the effect of narrow forward angle scattered light in the attenuation measurement by AC meters is developed. This will aid in a more accurate measurement of the inherent optical properties, which will, in turn, improve the remote sensing algorithms. Data & Method: A research cruise was conducted off Chennai region in April 2016. Concurrent measurements of Inherent Optical Properties like absorption(a) and attenuation(c) using WETLABS ACS instrument (from 400–700 nm with 3.8 nm interval), Temperature, Salinity and Depth using SBE FastCAT CTD sensor and Apparent Optical Properties using TriOS radiometers was performed meticulously. The absorption and scattering data were corrected for temperature (Pegau and Zaneveld, 1993) and salinity (Pegau et al., 1997) effects. Absorption coefficient was further corrected for scattering effects (Zaneveld et al., 1994). A relationship between c and the diffuse attenuation coefficient (kd) was derived using IOCCG (Lee et al, 2006) simulated datasets. The derived equation was applied to in-situ kd to model c which was compared against measured c to evaluate the difference (here, referred as c corr , which can be assumed as observed error in measured c). The difference, ccorr , was compared with b coefficients to obtain correction equation for measured c. b was calculated from bb datasets (IOCCG simulated) using Haltrin (2002) relationship and subsequently, resultant c was obtained by adding b to a. Fig 3. Correlation between ccorr and scattering coefficient b(510) from in situ dataset collected from Off Chennai Cruise (N = 7, R2 = 0.6).

The relationship between the correction factor for the attenuation coefficient and b followed a non-linear power-law function given by:

ccorr  mb

n

ccorr  fa

g

(3) Where m = 0.0888 and n = 3.2501. Here, a new relationship for obtaining the correction factor is obtained while, it must be noted that the scattering coefficient is the sole input to relationship. Alternatively, a similar relationship can also be derived with input as absorption coefficient (R2 = 0.5, N = 7) where, f = 5.8384 and g = 1.8154, and takes the form :

Fig 1. Correlation between kd(510) and beam attenuation coefficient c(510) fromt the IOCCG simulated dataset. This relation is to be used in obtaining model c(510) values from the in situ kd(510)

Results: The correlation between kd(510) and c (IOCCG derived) was developed following 4th order polynomial function given by: 4

3

2

1

c (510 )  px  qx  rx  sx  t

(1)

Here, p = -0.69023, q = 3.82728, r = -8.95887, s =,11.30360 t =-0.28516 and x = kd(510).

Fig 2. Correlation between the c obtained from IOCCG simulated data sets (using Haltrin (2002) relationships) and the modelled c of this work.

Eq. 1 was adopted to model c(510) using in-situ kd(510) values. The derived c was compared against in situ values and any difference is assumed to accrue from the measurement error in c. Equation (2) shows the relationship between modeled c (c(calc)) and in situ c (c(insitu)) based on correction factor, ccorr :

c (calc )  c (insitu )  ccorr

(2)

(4)

Conclusion & Discussion: A new model for correction of measurement errors in attenuation coefficients has been developed . The present method have a promising potential to quantify the errors caused due to scattering in narrow forward angle and acceptance angle of the instrument based on ancillary radiometric measurements. However, the relationship was developed using a limited number of datasets belonging to specific range of water types. The method can be applied to higher range of datasets which can be expected to yield a better correction procedure for measured c. The proposed approach provides an effective alternative to methods that solely rely on Monte Carlo simulations requiring high computational time for developing a correction procedure. Thus, it can be understood that present technique is a simple and effective solution for developing regional or global methods for scattering error correction based on the datasets acquired over different water types. Future work can be extended to fine tune the proposed method by using dynamic dataset from a wide variety of complex water types which include coastal, turbid and highly productive inland and estuarine waters. References: • Lee, ZhongPing. Remote sensing of inherent optical properties: fundamentals, tests of algorithms, and applications. Vol. 5. International Ocean-Colour Coordinating Group, (2006). • Haltrin, Vladimir I. "One-parameter two-term Henyey-Greenstein phase function for light scattering in seawater." Applied Optics 41.6 (2002): 1022-1028. • Leymarie, Edouard, David Doxaran, and Marcel Babin. "Uncertainties associated to measurements of inherent optical properties in natural waters." Applied optics 49.28 (2010): 5415-5436. • Gallegos, Charles L., David L. Correll, and Jack W. Pierce. "Modeling spectral diffuse attenuation, absorption, and scattering coefficients in a turbid estuary." Limnology and Oceanography 35.7 (1990): 1486-1502. • Pegau, W. Scott, Deric Gray, and J. Ronald V. Zaneveld. "Absorption and attenuation of visible and near-infrared light in water: dependence on temperature and salinity." Applied optics 36.24 (1997): 6035-6046. • Pegau, W. S., and J. Ronald Zaneveld. "Temperature-dependent absorption of water in the red and near-infrared portions of the spectrum." (1993). • Zaneveld, J. Ronald V., James C. Kitchen, and Casey C. Moore. "Scattering error correction of reflecting-tube absorption meters." Ocean Optics XII. International Society for Optics and Photonics, 1994.