Soil Moisture Estimation Using MODIS Images (Case Study: Mashhad Plain Area)

آب و خاک

Article 15, Volume 29, Issue 6, January 2015, Pages 1735-1748   

Document Type: Research Article

DOI: 10.22067/jsw.v29i6.34978

Authors

M. Fashaee^* ; Seied Hosein Sanaei-Nejad; K. Davary

Ferdowsi University of Mashhad

Abstract

Introduction: Numerous studies have been undertaken based on satellite imagery in order to estimate soil moisture using vegetation indices such as NDVI. Previous studies suffer from a restriction; these indices are not able to estimate where the vegetative coverage is low or where no vegetation exists. Hence, it is essential to develop a model which can overcome this restriction. Focus of this research is on estimation of soil moisture for low or scattered vegetative land covers. Trapezoidal temperature-vegetation (Ts~VI) model is able to consider the status of soil moisture and vegetation condition. It can estimate plant water deficit for weak or no vegetation land cover.
Materials and Methods: Moran proposed Water Deficit Index (WDI) for evaluating field evapotranspiration rates and relative field water deficit for both full-cover and partially vegetated sites. The theoretical basis of this method is based on the energy balance equation. Penman-Monteith equation of energy balance was used to calculate the coordinates of the four vertices of the temperature-vegetation trapezoid also for four different extreme combinations of temperature and vegetation. For the (Ts−Ta)~Vc trapezoid, four vertices correspond to 1) well-watered full-cover vegetation, 2) water-stressed full-cover vegetation, 3) saturated bare soil, and 4) dry bare soil. WDI is equal to 0 for well-watered conditions and equals to 1 for maximum stress conditions. As suggested by Moran et al. to draw a trapezoidal shape, some field measurements are required such as wind speed at the height of 2 meters, air pressure, mean daily temperature, vapor pressure-temperature curve slope, Psychrometrics constant, vapor pressure at mean temperature, vapor pressure deficit, external radiation, solar radiation of short wavelength, longwave radiation, net radiation, soil heat flux and air aerodynamic resistance is included. Crop vegetation and canopy resistance should be measured or estimated. The study area is selected in the Mashhad plain in Khorasan Razavi province of I.R. Iran. Study area is about 1,200 square kilometers and is located around the Golmakan center of agricultural research. In this study, water deficit index (WDI) was zoning by MODIS images in subset of Mashhad plain during water year of 2011-2012. Then, based on the close relationship between WDI and soil moisture parameter, a linear relationship between these two parameters were fitted. Soil moisture is measured by the TDR and every 7 days at 5 depths of 5, 10, 20, 30 and 50 cm from the surface. Remote Sensing (RS) technology used as a tool for providing some of the data that is required. The moderate resolution imaging spectroradiometer (MODIS) instrument is popular for monitoring soil moisture because of its high spectral (36 bands) resolution, moderate spatial (250–1000 m) resolution and various products for land surface properties. MODIS products used in the present study include: MOD09A1 land surface albedo data, MOD11A1 land surface temperature data, and MOD13A1 vegetation data. Using ArcMap 9.2 and ERDAS IMAGINE 2010 softwares, WDI was calculated pixel by pixel for 18 days (non-cloudy days and simultaneous with measurement of soil moisture at the station).
Results and Discussion: The results showed that the northeastern region is predominantly rainfed and irrigated farmlands are under water stress. Conversely, the southwestern part of the area is mountainous with less water stress. Based on NDVI, there is also less crop cover in the southwestern part of the region during the year. The results showed that about 44% of the index values are in the range of 0.2-0.3. Then about 22% of the index values are in the range of 0.3-0.4. Thus it can be concluded that over 66% of the index values are in the range of 0.2-0.4. According to the maximum index value (WDI=0.59 on the 201th day of year) and the minimum values (WDI=0.0004 on the 129th day of year) during the time period of study, it seems that water stress in the study area in the six-month period of observation is moderate. To validate the results, changes in precipitation, relative humidity and WDI values were compared. As expected, after the occurrence of any significant rainfall, water stress is decreased and decreasing in relative humidity, coincided with increase in water stress. In the next step, the linear relationship between measured values of soil moisture and WDI values were fitted in 2 depth of 5 and 10 cm. It should be noted that the average values of WDI of four pixels surrounding the Golmakan station was used in calculation of the regression coefficients Similar research has shown that Ts~VI trapezoid based WDI can accurately capture temporal variation in surface soil moisture, but the capability of detecting spatial variation is poor for such a semi-arid region like Mashhad. The high correlation coefficient (93%) obtained from soil moisture (5 cm) and WDI regression showed the good mutual impacts of these two parameters on each other. The correlation coefficient between WDI index and soil moisture at a depth of 10 cm was equal to 83%. Reducing the value of the correlation coefficient was probably due to the delay in transferring the soil moisture changes to underlying depth.
Conclusion: The similarity of the mean values of rainfall and relative humidity of the air showed good compliance with the WDI. Good correlation coefficient (93%) between WDI and soil moisture (measured at depth of 5cm in the station) certifies the accuracy of the results obtained from WDI. The results showed that Ts~VI trapezoid based WDI can well capture temporal variation in surface soil moisture, while in this study, spatial zoning was avoided because of the lack of soil moisture data within the study area.

Keywords

MODIS; Remote sensing; Soil Moisture; Water Deficit Index

References

1- Brutsaert W.H. 1982. Evaporation into the Atmosphere, D. Reidel Publ. Co., London, England, 299 p.

2- Hatfield, J.L. 1997. Plant-water interactions, Chapter 3 IN Plants for Env. Studies, CRC Lewis Publishers, New York, pp. 81-103.

3- Hiler E.A., and Clark R.N. 1971. Stress day index to characterize effects of water stress on crop yield. Trans. ASAE 14:757-761.

4- Idso S.B., R.D., Jackson P.J. Pinter Jr., Reginato R.J. and Hatfield J.L. 1981.Normalizing the stressdegree-day parameter for environmental variability, Agric. Meteorol. 24:45-55.

5- Jackson R. D. 1987. The crop water stress index: A second look, inProc, Int, Conf, on Measurement of Soil and Plant Water Status, Utah State Univ., 6-10 Jul, Logan, UT.

6- Krapez J. C. 2009. Comparison of three methods based on the Temperature-NDVI diagram for soil moisture characterization, Remote Sensing for Agriculture, Ecosystems, and Hydrology XI, edited by Christopher M. U. Neale, Antonino Maltese, Proc. of SPIE Vol. 7472, 74720Y, France

7- Liang S. 1999. Retrieval of Land Surface Albedo from Satellite Observations: A Simulation Study[J],J. Appl. Meteorol., 38, 712–725.

8- Moran M. S., Clarke T. R., Inoue Y., and Vidal. 1994. Estimating crop water deficit using the relation between surface air temperature and spectral vegetation index, Remote Sens. Environ., 49,246–263.

9- Moran M. S. 2000. Thermal infrared measurement as indicator of plant ecosystem health, USDA-ARS Southwest Watershed Research Center, Tucson, Arizona (1), 257–282.

10- Nemani R. R., and Running S. W, 1989. Estimation of regional surface resistance to evapotranspiration from NDVI and thermal IR AVHRR data, J. Appl. Meteorol., 28, 276-284.

11- Penman H.L. 1948. Natural evaporation from open water, bare soil and grass, Proc. Roy. Soc. London A193:120-146.

12- Petropoulos G., Carlson N., Wooster M. J., Islam S., and Hall A. 2009. A review of T s / VI remote sensing based methods for the retrieval of land surface energy fl uxes and soil surface moisture, 33(2), 224–250. doi:10.1177/0309133309338997.

13- Portal of Khorasan Razavi environmental protection agency, Province introduction, Environmental specifications. Available at http://www.rko.doe.ir/Portal/home (visited 24 April 2013)

14- Portal of Khorasan Razavi governer, Province introduction. Available at http://khorasan.ir (visited 24 April 2013)

15- Portal of Khorasan Razavi research, education and propagation agriculture agency, New NEWS, Golmakan Gardening research station, Propagators and gardeners. Available at http://www.areo.ir (visited 1 February 2013)

16- Sandholt I., Rasmussen K., and Andersen J, 2002. A Simple Interpretation of the Surface Temperature/Vegetation Index Space for Assessment of Surface Moisture Status, Remote Sens. Environ., 79(2), 213–224.

17- Sohrabinia M. Geostatistical analysis of surface temperature and in-situ soil moisture using LST time-series from modis, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012, XXII ISPRS Congress, 25 August – 01 September 2012, Melbourne, Australia

18- Wang W. 2010. Estimate soil moisture using trapezoidal relationship between remotely sensed land surface temperature and vegetation index, Hydrol. Earth Syst. Sci. Discuss., 7, 8703–8740, 2010

19- Wiegand C.L., and Namken L.N. 1966. Influences of plant moisture stress, solar radiation and air temperature on cotton leaf temperature, Agron. J. 58:552-556.

20- Zillman J. W. 1972. A study of some aspects of the radiation and heat budgets of the southern hemisphere oceans, Canberra: Australian Government Publishing Service, Canberra, Australia.