Home Articles Application of MSMR brightness temperature for retrieval of land surface parameters

Application of MSMR brightness temperature for retrieval of land surface parameters

Parag S. Narvekar


Parag S. Narvekar
Center of Studies in Resources Engineering (CSRE), Indian Institute of Technology (IIT) Bombay, Powai Mumbai 400 076
Email: [email protected]

Introduction
Multi Scanning Microwave Radiometer (MSMR) provides data for Brightness Temperature (TB) at four different frequencies 6.6 GHz, 10.65 GHz, 18 GHz and 21 GHz, at Vertical and Horizontal Polarizations. From the previous studies there are various models provided for estimating soil moisture, vegetation water content, surface roughness, etc. Due to the complexity of equations used, it is a tedious task to obtain these parameters with better accuracy. In the present work, 50 by 50 Km area at Bikaner Rajasthan, India is considered and the TB data of MSMR of IRS P4, SMMR of Nimbus 7 and theoretical calculations are compared. Plots of TB v/s frequency shows an unexpected fall in TB values at 18 GHz frequency of IRS P4 indicating some problem at 18 GHz of IRS P4. thus the measurements at 6.6 GHz, 10.65 GHz and 21 GHz are considered. Referring to data from “Benchmarks of India” [1], the TB is calculated for range of soil moisture, vegetation water content, surface roughness values etc., for 50 by 50 Km area of Bikaner. These results are compared with TB data of Nimbus 7 and IRS P4 satellites for the retrieval of land surface parameters with better accuracy.

Theory
Brightness Temperature as a function of soil moisture, vegetation water content, and surface roughness is given as [2]

Where
TBp is the polarization at vertical or horizontal brightness temperature.
Rsp is rough surface reflectivity.
tc is the vegetation opacity.
wp is the single scattering albedo.

Relation for Rsp and tc are same as given in [3].


where Rr is the rough surface reflectivity at look angle q, and h is


Rp is the refractivity of smooth surface and is given by


Thus for the ranges of soil moisture, vegetation water content and if required surface roughness, for the area under study, the simulations can be made and the results can be compared with satellite data sets for estimation of Land surface parameters.

Results and Discussion
On comparison the plots of TB v/s frequency of IRS P4, Nimbus 7 and Theoretical calculations it is seen that there is a decrease in the value of TB for IRS P4 at 18 GHz.


Figure 1: shows plots for TB at horizontal polarizations with frequency for (A) Theoretical calculations, (B) for Nimbus 7 and (C) for IRS P4.
Plot (C) shows after 10.65 GHz the value of TB decreases for 18 GHz and then again increases for 21 GHz. While in other two cases (A) and (B), value of 18 GHz is more than the 10.65 GHz. Thus calculation are done at other three frequencies i.e. 6.6 GHz, 10.65 GHz and 21 GHz for the month of September.

The graphs of TB v/s Days are plotted for Nimbus 7 for 1978, IRS P4 for 2001 and Theoretical calculations are made for the values from Benchmark soil of India [1].





Figure 2: plots shows the variation of brightness temperature. Curve at higher TB is for vertical polarization and one with lower TB is for horizontal polarizations at 6.6 GHz frequency for Nimbus 7(a), IRS P4(b) and Theoretical calculations ( c).
These plots are compared to see the variations of TB throughout the year for the Area of interest. Theoretical curve is drown taking average of TB over a month. It is seen that around month of may there is a decrease in the TB because of soil moisture present and due to the effect of vegetation water content. In the similar way the simulations are made for all the ranges temperatures, soil moisture, and vegetation water content. These simulations can be compared with the actual values of TB from satellite so that to get perfect match with satellite data.

Conclusion
The results obtained using above technique closer resemblance with the actual ground values of surface parameters. Thus using satellite data it is possible to get idea of the nature of the surface and vegetation presence, which can be used effectively weather forecasting, hydrologists, agriculturist, etc. It has been seen that, microwave polarization difference gives more accurate estimation of vegetation water content and soil moisture. Thus it has been tried to use this difference in inversion technique to get better results in future.

References

  • Benchmarks Soil of India, national bureau of soil survey and land used planning. (University of Agricultural Science, Bangalore India.
  • Eni G. Njoku, Retrieval of Land Surface Parameters Using Passive Microwave Measurments at 6 – 18 GHz. IEEE Transactions on Geoscience and Remote Sensing, Vol. 37, NO. 1, January 1999.
  • Myron C.Dobson, et.al, Microwave Dielectric Behavior of Wet Soil-Part II: Dielectric Mixing Model, Member IEEE.