Estimation of Soil Erosion using USLE and Landsat TM in Pakistan
Materials and methods
In this case study attempts have been made to monitor land cover distribution and evaluation of land degradation/erosion conditions using remote sensing and GIS techniques. Using data and methods are showed in figure 3 and as follows. The USLE is an empirically based model developed in the United States by using data on soil erosion rates. This equation has certain limitations but still is the best available method which is used most widely for estimating soil losses as average annual mass per unit area as a function of the major factors affecting sheet and rill erosion. It is showed as follows:
E = RKLSCP ---------------------(1)
Where, E: average annual soil loss (tones/ha/year), R: rainfall runoff erodibility factor, K : soil erodibility factor, L : slope length factor, S : slope steepness factor, C: land cover and management factor, P : erosion control practice factor.
| E: Soil loss(ton/ha/year) |
L: Slope length factor |
| R: Rainfall erodivity factor |
S: Slope steepness factor |
| K: Soil erodivity factor |
C: Cover and management factor |
| |
P: Erosion control practice factor |
Figure 3 Analysis flow of USLE model using RS & GIS
Rainfall factor (R ) is calculated as the product of storm kinetic energy time the maximum 30-minute storm depth (I30) and summed for all storms in a year. Rainfall was measured by test site locations to 10 km from study area and R was calculated from I30 and summed in monsoon season and non-monsoon (table 1). Soil erodibility factor (K) is a measure of the inherent erodibility of a given soil under the standard condition of the unit USLE plot maintained in continuous fallow. But, K value is not measured in Pakistan. We estimated K from table 2, table 3 and land capacity map. Almost soil types are categorized to moderate erodibility values (K = 0.3). Slope length factor (L) and slope steepness factor (S) are generally calculated the equation as follows:
LS = (L/22.13)m (0.065 + 0.045S + 0.0065S2)--------------(2)
where, S: slope steepness (%), L : slope length (m), m: m = 0.5. L is possible to calculate from Digital Topographic Model (DTM). But, it is impossible that the slope is continuous or no. If slope length is calculated from DTM. But, it is impossible that the slope is continuous or no. If slope length is calculated from DTM, it is sometimes very long. In this study, the slope length was fixed as 22.13 m. This value means the length of test field to decide soil erodibility factor. Slope steepness was calculated from DTM for each pixel. DTM was made from map (1/50,000) using ARC/INFO. Land cover and management factor (c ) was decided from land classification map and table 4. Land classification map was created from Landsat Thematic Mapper (TM) data acquired on 9 February 1992 by supervised classification using most likelihood methods. Erosion control practice factor (P) was decided from land classification map and table 5.
Table 1 Raifall characteristics at Fatehjang Sub-Watershed
| Parameters |
1989 |
1990 |
1991 |
1992 |
1993 |
1994 |
Mean |
| (Monsoon) |
| Total Raifall (mm) |
388 |
771 |
468 |
522 |
332 |
588 |
511.5 |
| Proportion of yearly Rainfall (%) |
74 |
64 |
74 |
67 |
66 |
76 |
70.2 |
| Max. per day (mm) |
69 |
153 |
42 |
105 |
67 |
90 |
87.7 |
| Min. per day (mm) |
2 |
1 |
5 |
5 |
3 |
2 |
3.0 |
| Rainfall events |
19 |
25 |
21 |
14 |
16 |
22 |
19.5 |
| Max. l30 (mm/hr) |
100 |
92 |
52 |
70 |
80 |
70 |
77.3 |
| Min. I30 (mm/hr) |
3 |
2 |
4 |
6 |
2 |
4 |
3.5 |
| Raifall events > 25 mm |
5 |
11 |
9 |
6 |
5 |
9 |
7.5 |
| Seasonal total EI30 |
572 |
947 |
277 |
484 |
415 |
546 |
540.2 |
| (Non Monsoon) |
| Total Raifall (mm) |
137 |
426 |
167 |
261 |
172 |
188 |
225.2 |
| Proportion of yearly raifall (%) |
26 |
36 |
26 |
33 |
34 |
24 |
29.8 |
| Max. per day (mm) |
30 |
51 |
44 |
53 |
31 |
22 |
38.5 |
| Min. per day (mm) |
2 |
2 |
1 |
3 |
2 |
2 |
2.0 |
| Rainfall events |
22 |
25 |
14 |
16 |
16 |
20 |
18.8 |
| Max. I30 (mm/hr) |
26 |
70 |
26 |
42 |
36 |
18 |
36.3 |
| Min. I30 (mm/hr) |
4 |
4 |
2 |
4 |
2 |
2 |
3.0 |
| Raifall events > 25mm |
- |
6 |
1 |
3 |
2 |
- |
3.0 |
| Seasonal Total EI30 |
84 |
216 |
45 |
77 |
57 |
35 |
85.7 |
Table 2 Soil Characteristic points to decide K value
| Soil characteristics |
division |
points |
| Soil human (%) |
<2 |
4 |
| 2-5 |
3 |
| 5-10 |
2 |
| >10 |
1 |
| Soil contents (%) |
>50 |
4 |
| 30-50 |
3 |
| 15-30 |
2 |
| <15 |
1 |
| soil structure |
blocky, planty or massive |
3 |
| medium to coarse granular |
2 |
| crumb or fine granular |
1 |
| permeability |
slow to very slow |
3 |
| moderate |
2 |
| repid to very rapid |
1 |
Table 3 K value and total points
| Total points |
division |
extent of K |
Mean of K |
| >5 |
Very low |
<0.10 |
0.05 |
| 6-8 |
Low |
0.10-1.20 |
0.15 |
| 9-11 |
Moderate |
0.20-0.40 |
0.30 |
| <11 |
High |
>0.40 |
0.45 |
Ministry of Forestry (Indonesia):
Handbook for the Preparation of Land Data Base
Forms for Regional Planning |
Table 4 C Factor Value for different classes
| Land use class |
Average C factor |
1. Primarily forest*
(canopy cover >40%) |
0.002 |
2. Secondary forest**
(canopy cover 10-40%) |
0.006 |
| 3. Shrub* |
0.014 |
| 4. Agricultural land** |
0.377 |
| 5. Grazing land** |
0.11 |
source *Adapted from Wischmeier and Smith (1978)
** Calculation from Morgan (1986) |
Table 5 Erosion Control practice factor (P)
| Land use type |
slope( %) |
P factor |
| Agricultural land |
0-5 |
0.11 |
| 5-10 |
0.12 |
| 10-20 |
0.14 |
| 20-30 |
0.19 |
| 30-50 |
0.25 |
| 50-100 |
0.33 |
| Other land |
all |
1.00 |
| source: based on interpolation from Wischmeier and Smith (1978) |
