In: Wagner W., Székely. B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
167 
soil texture and were derived using soil hydraulic properties 
index defined in VIC model documentation. 
a. Landuse/Landcover Map of 1972 
b. Landuse/Landcover Map of 1985 
c. Landuse/Landcover Map of 2003 
Fig. 2 Landuse/landuse Map of 1972, 1985 & 2003 
Table l.LULC statistics of 1972,1985 & 2003 
W#» îawS* 
ÎH) 
DF Al^M D**4 
SDttrr 
{Hi 
i*2 
m. 
»72 
1.16 
27.23 
1196 
5406 
a os 
415 
134 
1*85 
1 19 
26 64 
10.92 
5641 
0.22 
383 
135 
:ooj 
162 
23.52 
996 
5&63 
0.31 
3.51 
144 
The vegetation parameter file describes the vegetative 
composition of each grid cell, and uses the same grid cell 
numbering as the soil file (latitudes and longitudes are not 
included in the file). This file cross-indexes each vegetation 
class (from any land-cover classification scheme) to the classes 
listed in the vegetation library. To prepare this file, landuse map 
was overlaid on the grid map and the no. of vegetation classes 
as well as fraction of grid covered by those classes was 
extracted. A small code in C language was used to read this 
information from crossed map and arrange it in the format 
specified by the model. Root depths for landcover types were 
accepted as recommended by Canadell et. al. (1996). It was 
assumed that the specified root zone contains all of the roots of 
a landcover type. For the selected land cover classification of 
the study area, a vegetation library file was set up. This 
describes the static (varying by month, but the same values 
year-to-year) parameters associated with each land cover class. 
LAI defines an important structural property of a plant canopy 
as the one sided leaf area per unit ground area. For derivation of 
LAI, MODIS LAI maps (MOD 15A02 product) were 
downloaded from NASA’s GSFC website (www.modis- 
land.gsfc.nasa.gov). For each landuse class, sufficient number 
of cloud free points were chosen and their LAI profile on the 
stacked image was drawn. An average monthly value of those 
points was taken as the LAI value in each month for that 
landuse class. Albedo was also derived from MODIS BRDF/ 
Albedo product in the same way. Other variables like roughness 
length, displacement height, overstory, architectural resistance, 
minimum stomatal resistance were derived from LDAS 8 th 
database (http://ldas.gsfc.nasa.gov/LDAS8th/MAPPED.VEG/ 
web. veg.monthly.table.html) and MM5 terrain dataset. 
The Meteorological forcing file contains meteorological 
variables required to force the VIC model like daily 
precipitation, daily maximum and minimum air temperature. 
Forcing data files play big role in the model input to produce all 
the outputs in both water balance and full energy balance modes 
of the model. Accurate streamflow simulation requires forcing 
input of high accuracy as it is the most influential variable 
generating runoff and driving hydrological cycle. Precipitation 
input was prepared using India Meteorological Department’s 
1°X1° gridded rainfall dataset. Daily rainfall values for each 1 
degree grid falling in the basin were extracted for 365 days in 
the year 2003. Rainfall grids were then overlaid with the basin 
grids to extract precipitation in each basin grid. VIC model 
requires one forcing file for each grid having 365 rows and 3 
columns in ASCII format. Temperature data from NCDC 
(National Climatic Data Centre, NOAA) is available for some 
selected stations in the study area. This point temperature data 
was used to derive maximum and minimum temperature of each 
basin grid using nearest neigbour and lapse rate method since it 
is assumed that temperature varies with the altitude. The 
following relation was applied in MS Excel spreadsheet: 
T grid — Tnearest point + 5.5/1000*(elevation near est point - grid 
elevation) (1) 
A Global control file where the necessary information to specify 
various user preferences and parameters was prepared. It 
contains information like N-layers, Time step, start time, end 
time, Wind H, snow temp, rain temp., Location of the input 
output files, modes which are to be activated etc. The VIC 4.0.5 
was compiled using gcc complier on Linux operating system. 
The code was compiled using the make file included in the 
archive, by typing 'make'. The compiled code creates an 
executable entitled 'vicNl'. To begin running the model, ‘vicNl - 
g (global control file name)' was written at the command 
prompt. Global control parameters were modified according to 
the input characteristics and to activate the ware balance. In 
addition to that input and output path were specified. VIC 
source code was executed in the LINUX environment to 
generate the flux files for each basin grid. These flux files