International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
  
  
  
  
  
  
  
  
Soil Regression Equation Multiple F 
Parameter R value 
Organic 0.49+2.9*BNI-3.4*Blue — 0.733 6.37 
Matter 0.5%H1 + 0.06*PC1 + 
% 29*SI 
Available -2844.8-866.5*Blue + 0.532 2.14 
N (ppm) 9960.5* VNIR - 
10767.2*PC2 - 
6811.1*Red + 1070.5*RI 
Sand (%) 103.9 - 19.96*RI 0.540 4.]* 
Silt (96) 40.3 - 52.8*Blue 0.599 5.6* 
Clay (%) 29.5 + 105.8*C1 0.495 33% 
  
  
  
  
  
  
*0.01<p<0.1, **0.001<p<0.01, ***p<0.001 
4.3 Variability map generation 
The soil fertility parameter variability maps were generated for 
OM and available N from the RS data using the above- 
mentioned empirical equations. After generating the maps they 
were applied with a 3x3 average filter to remove the noisiness 
in the maps and then classified into four classes. The outputs 
are presented in figure 2. The classes for OM had average 
values of 0.23, 0.24, 0.25 and 0.26 percent, which are 
represented from dark to light tones in the figure. Similarly the 
three classes for available N had average values of 100.0, 105.0 
and 120.0 ppm, again represented from dark to light tones in the 
figure. It may be mentioned that, the concerned field (4.43 ha), 
as per the conventional soil classification, had two major soil 
types such as sandy loam and clay loam. Thus the management 
practice based on conventional soil classification could have 
resulted into only two types, where as the remote sensing data, 
can identify more classes. These variability maps can be used 
for site-specific soil fertility management. 
  
Figure 2. Soil fertility parameter variability maps generated from remote sensing data a) Organic matter per cent (0.23, dark - 
0.26,light) b) Available nitrogen in ppm (100.0,dark — 120.0, light) 
5. CONCLUSIONS 
This study showed the usefulness of using high resolution 
multi-spectral remote sensing data for estimation of soil 
nutrient related parameters, and generate within field nutrient 
variability maps. For some of the parameters the poor 
correlations can be attributed to the time gap between soil 
observations and satellite pass. Thus, near synchronous data 
acquisition without time lag is a critical requirement for such 
studies. For further study, the data from LISS IV sensor (5.8. m 
resolution) on board Indian Resourcesat satellite is proposed to 
strengthen such findings. 
REFERENCES 
Baumgardner, M. F., Kristof, S.L, Johannsen, C.J., and 
Zachary, A.L., 1970. Effect of Organic Matter on multispectral 
properties of soils. Proceedings of Indian A cademy of Science, 
79, 413-422. 
Coleman, T.L., and Montgomery, O.L., 1987. Soil moisture, 
organic matter and iron content effect on spectral characteristics 
130 
of selected Vertisols and Alfisols in Alabama. Photogrammetric 
Engineering and Remote Sensing, 12, 1659- 1663. 
Dwivedi, R.S.,2001 Soil resources mapping: A remote sensing 
perspective. Remote sensing Review, 20, 89-122 
Horvath, E.H., Post, D.F., and Kelsey, J.B., 1984, The 
relationships of Landsat digital data to the properties of Arizona 
Rangeland. Soi/ Science Society American Journal, 48, 1331- 
1334. 
Leone, A.P. Wright, G.G. and Corves, C. 1995. The 
application of satellite remote sensing for soil studies in upland 
areas of southern Italy. Int. J. Remote Sensing, 16, pp.1087- 
1105. 
Muhr, G.R., Dutta, N.P., Sankarsubramaney, H., Leley, V.K. 
and Donahue,R.L., 1965. Soil Testing in India. om Edn., 
U.S.A.LD., New Delhi. pp 49. 
Mulla, D.J., A.C. Sekely, and M. Beatty. (2000). Evaluation of 
remote sensing and targeted soil sampling for variable rate 
application of lime. CD-ROM. In: P.C. Robert et al. (ed) 
Proc. 5" Intl. Conf. on Precision Agriculture, Minn., MN, Jul 
16-19, 2000. ASA, CSSA, and SSSA, Madison, WI 
Internatior 
EL IH 
Olsen, S.R 
Estimation 
sodium bic 
Ray, S. 8 
Parihar, J. 
remote sen 
ISRS Syn 
Ahmedaba 
Ray, S.S, 
Analysis 0 
data and s 
manageme 
Inform. Sy 
Subbiah, E 
estimation 
pp. 259-60 
Suk, Y.H. 
Palm, H.L