cameras used or the film processing. The film type 
and processing house were constant throughout. 
4 COST COMPARISON 
The following figures are intended to give some 
idea of the relative costs of imagery (at April 
1986). The costs of U.K. photography were from an 
area of 900km^ and the costs for Zimbabwe are 
based on standard calculating figures. 
Table 5. Cost comparisons. 
Landsat 
TM-whole 
scene 
Landsat 
TM -1/4 
scene 
B&W aerial photos 
U.K. Zimbabwe 
1:10000 1:25000 
KILLPACK, D.P. & McCOY, R.M., 1981. An application 
of Landsat derived data to a regional 
hydrologic model, Rem. Sens. Quart. 3(2),27-33. 
SEIDEL,K.,ADE,F., LICHTENEGGER, J.,1983.Augmenting 
Landsat MSS data with Topographic Information 
for Enhanced Registration and Classification, 
IEEE Transactions on Geoscience and Remote 
Sensing, Vol. GE-21, No. 3, July 1983. 
SMART, J.S., 1972.Channel networks. Advances in 
Hydroscience 8,305-346. 
RANGO,A.,FOSTER,J.,SALOMONSON,V.V.,1975.Extraction 
and utilization of space acquired physiographic 
data for water resources development. Water 
Res. Bull. 11(6), 1245-1255. 
Sympcn 
Appli* 
develc 
(India 
A.S.Jadl 
Dept, of G 
Cost of 0.06 0.12 16.55 6.20 
imagery 
in £/km^ 
These costings of satellite imagery are for the 
whole area which in most circumstances will not be 
fully required. The cost of digital image 
processing is not included (U.K. hire rates: 
April,1986, £35-80/hour). 
5 CONCLUSIONS 
The efficiency of Landsat TM imagery in defining 
drainage networks is significantly greater than 
that of Landsat MSS imagery. This results in more 
accurate delineation of catchment boundaries and 
catchment area estimation, particularly for small 
catchments. 
Within the catchment area, Landsat TM imagery 
provides useful information on geomorphological 
parameters, with the identification of lakes as 
small as 0.6 hectares, and small streams (3-5 
metres wide) being fairly consistently recorded. 
Several different landcover types can be 
identified using Landsat TM imagery. Although 
there is more data to examine in the case of 
Landsat TM the relative ease of interpretation 
compensates for this factor. 
In terms of image enhancement the most useful 
waveband combinations for Landsat TM are 1,4,5 ; 
2,4,5 and 3,4,5. Edge enhancement using a 
Laplacian filter, Sobel operator or directional 
filtering, with a 3x3 kernel further increases the 
drainage network detail. 
ABSTRACT 
explorat 
are usee 
The main 
of grour 
water pc 
the lane 
The stuc 
field wc 
It is fc 
techniqu 
associat 
1 I NT ROC 
Whether 
drinkinc 
(surface 
critical 
of hydre 
variety 
ty, qu ar 
the wate 
A prec 
a good c 
water pc 
this stt 
aerial p 
mented v* 
and dist 
2 GEOGRP 
Consistency in image recording and processing 
should be closely monitored so as to produce 
consistent results, thus enabling more meaningful 
comparisons. 
ACKNOWLEDGMENTS 
The author would like to thank the Overseas 
Development Administration, NRSC Farnborough and 
NERC, Swindon for their assistance and the use of 
image processing facilities. 
REFERENCES 
CHIDLEY, T.R.E. & DRAYTON, R.S., 1985. The use of 
SPOT simulated imagery in hydrologic mapping, 
Int. Jour. Rem. Sens., in press. 
DRAYTON, R.S. & CHIDLEY, T.R.E., 1985. Hydrologic 
modelling using satellite imagery. Proc. Int. 
Conf.on Advanced Technology for monitoring and 
processing global environmental data, Remote 
Sensing Society/CERMA, 1985. 
HORTON, R.E.,1945. Erosional development of 
streams and their drainage basins: 
hydrophysical approach to quantitative 
morphology. Bull.of the Geological Soc.of 
America 56,275-370. 
The stuc 
27°35’ 2* 
76°58'2" 
of Harye 
of Indie 
The tc 
with moc 
main hiJ 
close ar 
the nor1 
ranges, 
ttered c 
There ai 
any majc 
beneath 
plain ar 
The rc 
charac tc 
during c 
able rar 
rainy sc 
the rair 
During 1 
weather 
rains oc 
thunder 
spells ( 
ture of 
transpii 
exceeds 
720