Home > Geospatial Application Papers > Natural Resource Management

Overview | Mountain | Water Resources | Ocean | Coastal Zone Management
    Overview
    Ground Water Assessment
    Water Quality
>>Watershed Management
    Surface Water Management

| Printer Friendly Format |


Page 1 of 1

Analysis of Watershed Change Using Temporal Multi spectral satellite data - The Shenit-Thokalchiwadi Watershed, Ahmednagar District, Maharashtra, India.

Dr. Hrishikesh P. Samant
Department of Geology,
St. Xavier’s College,
Mumbai 400 001



Abstract
Shenit-Thokalchiwadi is located in Akola Taluka of Ahmednagar district. The area used to be, semi arid and barren of vegetation. The terrain is rugged and hilly with the highest peak in the Deccan i.e. Kalsubai quite close by. The average rainfall is around 500mm, but the area suffered from periodic drought, resulting in highly depleted ground water reserves. The agriculture in the area being primarily rain fed, was a failure (Tripathy, et.al, 1996).

In 1995, Watershed Organisation Trust (WOTR) an NGO based in Ahmednagar, Maharashtra began a systematic watershed development programme in the Shenit watershed.

Soil conservation being the main task in the successful implementation of a watershed development programme, (Sebastian, 1995) area treatments, drainage line treatments, aforestation and pasture development was extensively done in the area.

To quantify the changes in the soil cover, after the implementation of watershed management practices, remotely sensed satellite data has been studied covering the pre and post treatment period. This report brings out the remarkable changes in soil and vegetation cover, which occurred between January 1996 and December 1999.

Introduction:
Shenit-Thokalchiwadi is located in Akola Taluka of Ahmednagar district. The area used to be, semi arid and barren of vegetation. The terrain is rugged and hilly with the highest peak in the Deccan i.e. Kalsubai quite close by. The average rainfall is around 500mm, but the area suffered from periodic drought, resulting in highly depleted ground water reserves. The agriculture in the area being primarily rain fed, was a failure (Tripathy, et.al, 1996).

In 1995, Watershed Organisation Trust (WOTR) an NGO based in Ahmednagar, Maharashtra began a systematic watershed development programme in the Shenit watershed.

Soil conservation being the main task in the successful implementation of a watershed development programme, (Sebastian, 1995) area treatments, drainage line treatments, aforestation and pasture development was extensively done in the area.

To quantify the changes in the soil cover, after the implementation of watershed management practices, remotely sensed satellite data has been studied covering the pre and post treatment period. This report brings out the remarkable changes in soil and vegetation cover, which occurred between January 1996 and December 1999.

Theoretical considerations in Vegetation Indexing:
The simplest of soil conservation methods is to plant grass or other vegetation, which will drastically reduce the removal of topsoil. Hence the ideal watershed should have a minimum of barren soil as seen in the field photograph (Plate 1), where small portions of the Shenit-Thokalchiwadi watershed still remain barren of any vegetation. Grass covered and stabilised soil has a much higher water retention capacity that leads to reduced surface runoff and higher and faster recharge of the aquifer.

Analysis of vegetation cover and detection of changes in vegetation patterns are keys to watershed resource assessment and monitoring. Green vegetation has a very distinctive interaction with energy in the visible and near infrared regions of the electromagnetic spectrum. In the visible regions, plant pigments (most notably chlorophyll) cause strong absorption of energy, primarily for the purpose of photosynthesis. This absorption peak in the red and blue areas of the visible spectrum, leads to the characteristic green appearance of most leaves. In the near infrared, however, a very different interaction occurs. Energy in this region is not used in photosynthesis, and it is strongly scattered by the internal structure of most leaves, leading to a very high apparent reflectance in the near infrared. It is this strong contrast between the amount of reflected energy in the red and near-infrared regions of the electromagnetic spectrum, which has been used to develop a quantitative index of vegetation or grass cover condition using remotely sensed digital data from our Indian Remote Sensing Satellite IRS 1C LISS III sensor. The spatial resolution of this data is 23.5 m.


Plate 1: Barren Soil cover in Shenit-Thokalchiwadi Watershed.

The Shenit-Thokalchiwadi watershed area being in a semi-arid environment, any vegetation indexing will have to take into consideration the effect of soil brightness since vegetation is sparse and pixels contain a mixture of green vegetation and soil background. The Distance-Based Vegetation Indices are based on the concept of a soil line. A soil line is a linear equation that describes the relationship between reflectance values between the red and infrared bands on a sample of bare soil pixels. The line is produced by running a simple linear regression between the red and infrared bands on a sample of bare soil pixels. Once that relationship is known, all unknown pixels in an image that have same relationship in red and infrared reflectance values are assumed to be bare soils. Unknown pixels that fall far from the soil line because they have higher reflectance values in the infrared band are assumed to be vegetation (based on the characteristic spectral response pattern for vegetation where the infrared band reflectance values are relatively higher than those of the red band).

Methodology:
Digital data from the Indian Remote Sensing Satellite IRS1C LISS III sensor for two dates (January 1996 and December 1999) was obtained, for the area covered by Survey of India’s Topographical map No. 47E14, from National Remote Sensing Agency. The area of interest was extracted from the complete coverage and accurately georeferenced with the topographical map (Fig. 1). The individual bands of the satellite data were separated into files. Standard False colour composites were generated using IRS 1C LISS III bands 2,3 and 4 (Fig 2 and 3). IRS 1C LISS III Band 3 and Band 4 data was used to calculate the Normalised Difference Vegetation Index (NDVI).


Fig. 1: Shenit-Thokalchiwadi Watershed: Topography and Drainage Network. (Source: SOI sheet 47E14)



Fig.2 Shenit-Thokalchiwadi Watershed: IRS 1 LISS lll FCC, January1996



Fig. 3 Shenit-Thokalchiwadi Watershed: IRS 1 LISS lll FCC, December 1999

The NDVI separates green vegetation from its background soil brightness. It is expressed as the difference between the near infrared and red bands normalised by the sum of those bands, i.e.:

NDVI = (Near Infra Red band – Red Band) / (Near Infra Red Band + Red Band). = (Band 4 – Band 3) / (Band 4 + Band 3)
Inputs to calculation of the Distance-Based VI’s are the red band, the infrared band, the slope of the soil line and the intercept of the soil line.

The first step in calculating the soil line is to identify an area of bare soil in the images. The NDVI images (created as described above) were used to develop a mask image for bare soil in the Shenit-Thokalchiwadi Watershed. Individual mask images were generated for the two years 1996 and 1999.

{Methodology adopted to generate the Mask Images:
Since Vegetated or grass covered areas have higher reflectance in the infrared than the red, the NDVI for vegetation will always be positive. Therefore, the bare soil pixels will have NDVI values less than or equal to 0.

The 1996 NDVI image has a range of values from -0.63 to 0.57 while the 1999 NDVI image has a range of values from –0.37 to 0.55. These two images were reclassified such that all values less than 0 were equated to 1 while all values more than zero were equated to 0. The two images, so obtained, bring out the areas, which are covered by barren soil and grass or vegetation (Fig. 4 And 5 ).}


Fig. 4 Shenit-Thokalchiwadi Watershed: Soil Mask image (January 1996)



Fig. 5 Shenit-Thokalchiwadi Watershed: Soil Mask image (December 1999)

The IRS 1C LISS III band 4 image was regressed against the band 3 image using the corresponding soil mask files to define the pixels from which the slope and intercept was calculated. The graphs Fig 6 for 1996 and Fig 7 for 1999 give the slope and intercept. (Note: for both the years the infrared band i.e. band 4 has been used as the independent variable in the regression). The high coefficient of determination indicates that the relationship between red and infrared reflectance for these bare soil pixels is described well by a linear equation, (Eastman, 1999).

For 1996:
Slope (b) = 0.925835
Intercept (a) = 9.879835
Coefficient of determination (r2) = 0.9313

For 1999:
Slope (b) = 0.899333
Intercept (a) = 9.586191
Coefficient of determination (r2) =0.9658

The Perpendicular Vegetation Index (PVI) developed by Perry and Lautenschlager (1987) uses the perpendicular distance from each pixel coordinate to the soil line as shown in Fig 8. Given the spectral response pattern of vegetation in which the infrared reflectance is higher than the red reflectance, all vegetation pixels will fall to the right of the soil line (e.g., pixel 2 in Fig 8 ). In some cases a pixel representing non-vegetation (e.g. Water) may be equally far from the soil line, but lies to the left of that line (e.g., pixel 1 in figure 8 ). This PVI assigns negative values to those pixels lying to the left of the soil line.

The equation is written as:
PVI = (b NIR – RED + a)/vb2 + 1

Where
NIR = reflectance in the near infrared band.
RED = reflectance in the visible band.
a = intercept of soil line.
b = slope of the soil line.


Fig. 6: Regression between Band 4 and Band 3 data of 1996



Fig. 7: Regression between Band 4 and Band 3 data of 1999


Fig. 8: Distance from the Soil line.

The two images thus generated for the two years viz. 1996 and 1999 are shown in fig 9 and fig. 10. The colours depict the percentage of vegetation in the area i.e. green is high chlorophyll content (vegetation) and as the colours shift towards yellow to yellowish-brown and finally brown the land is barren or without vegetation. The white patches are areas devoid of soil or vegetation viz. barren rock as seen in plate 2. These barren rock areas are the very steep or almost vertical cliffs seen in the northern parts of Shenit village.


Fig.9 PVI image: Shenit-Thokalchiwadi, January 1996.



Fig. 10 PVI image: Shenit-Thokalchiwadi, December 1999.



Plate 2. Barren rock cliffs in the northern part of Shenit-Thokalchiwadi watershed.



Fig. 11. Bar graph depicting the barren soil cover in 1996 and 1999



Fig. 12. Bar graph depicting the rainfall and green vegetation in 1996 and 1999

Conclusion:
On quantifying the areas under the various categories within the soil mask images for 1996 and 1999 it is evident that the barren soil cover has reduced considerably (from 115 Ha in 1996 to just 30 Ha in 1999) after the watershed management programme was implemented in Shenit-Thokalchiwadi watershed. This is very well depicted in the bar graph in fig 11. What is more laudable is that this positive change is seen even though the annual rainfall in 1999 was much less than that in 1996, and is graphically represented in fig. 12, where the green vegetation cover has increased from 773 Ha in 1996 to 858 Ha in 1999 though the rainfall decreased from 1197.45 mm in 1966 to just 819.5 mm in 1999.

At present the change detection studies for the last five years (1999-2004) is in progress.

References
  • Eastman, R.J., 1999. Guide to GIS and Image Processing, V-2, Clark Labs, Clark University, MA, USA.
  • Perry, C.Jr., and Lautenschlager, L.F., 1984. Functional Equivalence of Spectral Vegetation Indices, Remote Sensing and the Environment 14: 169-182.
  • Sebastian, M., Jayaraman, V. and Chandrasekhar. M.G., 1995. Space Technology Applications for Sustainable Development of Watersheds, Technical Report, ISRO-HQ-TR-104-95, ISRO, Bangalore.
  • Tripathy, G.K., Ghosh, T.K., and Shah, S.D., 1996. Monitoring of Desertification Process in Karnataka State of India Using Multi-Temporal Remote Sensing and Ancillary Information Using GIS, International Journal of Remote Sensing, 17(12): 2243-2257.