International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
checking or verifying outcomes, and inefficiency in measuring 
several hundreds of small sub-watersheds are considered to be 
the main delimits of manual measurements. 
Additionally, manually delineated watershed was compared 
with an automated one. An automated delineation of the 
watershed was performed by means one of the widely used 
scripts in ArcView. The comparison of an automated and 
manually delineated whole Thadokhola watershed is given in 
Figure 4. An automated delineation for the whole watershed (14 
km?) is very similar to that of manually produced watershed. 
Where brown color represents the automated Thadokhola 
watershed and Opaque represents the manually produced 
Thadokhola watershed. Unfortunately, for the small sub- 
watersheds (lower than TOWs) having an area approx. less than 
0.1 km, the script did not found worthy. The automated small 
sub-watersheds were not resembled with manually produced 
one. The script might not be compatible for the small 
watersheds. Further study on this using several such scripts 
need to be experimented. 
  
  
  
Figure 3. Manual Sub-water Delineation on 
Topographical Map 
3. RESULTS AND DISCUSSION 
3.1 Geomorphometric Parameters 
Thadokhola watershed is a four order watershed, comprises 101 
first order (total area 7.16 km”), 17 second order (5.96 sq.km) 
and 2 third order watersheds (9.14 km?). First order watersheds 
(FOW) attribute approx. 51 %, 60% and 66% of its total area to 
whole, third order watershed (TOW) and second order 
watersheds (SOW), respectively. Similarly, SOWs attributes 
42% of total area to whole and 56 % to the TOWSs, whereas, 
only 65% of total area of TOWSs attribute to the whole 
Thadokhola watershed. 
  
Figure 4. Comparison of Automated and Manually Delineated 
  
  
  
  
  
  
  
  
  
  
  
  
Watersheds 
Watersheds 
First Second | Third | Thadakola 
Order Order Order Total 
Average 0,07 0,35 4,57 
Area calculated 
in Areview [Min Q.01 | 0,07 | 4,57 
Max. 0,47 2,16 4,57 14,10 
St. Dev. 0,07 0,49 0,00 
Total 7,16 5,96 9,14 
Average 0,07 0,35 4,52 
Area Measured [Min 001 | 007 | 451 
by Planimeter | Max. 0,44 2,12 4,52 14,00 
GCPLAN G0) | st. Der 0,07 0,48 0,00 
Total 7,19 5,89 9,03 
  
  
  
  
  
  
  
  
Table 1. Comparisons on Automated and Manual 
Measurements of Watershed Area 
In comparing to other watersheds in Siwalik region, 
Thadokhola watershed has properties of larger sub-watershed 
arca, longer stream length, higher drainage density and gentle 
stream slope (Shrestha, 2001). High drainage density suggests 
the occurrence of high erosion rates in lower order watersheds. 
Larger area, longer stream length and gentle slope gradient of 
higher order watersheds infer the mitigation of sediment 
transportation. This presumption was verified by field survey 
carried out during the dry season of 2000. Sediments from 
lower order watersheds were deposited at/near the branch points 
of higher order watersheds forming small fans and talus. Thus, 
if secondary sediment transportation from the higher order 
watersheds is controlled, volume of downstream sediment 
supply will be lessened significantly. 
566 
Interne 
St 
  
Nur 
Ave 
lene 
= 
€ 
  
  
Ave 
(km 
  
Ave 
diff 
  
Ave 
den: 
  
Stre 
(nos 
  
Bifu 
  
  
Table 
3.2 Li 
The la 
study : 
floodp 
rain-fe 
within 
lands a 
respect 
due to 
soil co 
an ope 
interve 
manag 
n 
[ 
Area (HT2) 
e 
M» 
  
ü 
  
Figure 
Landus 
transfoi 
floodpl. 
There i 
taken | 
2000). 
occurre 
rainfall