Host: The Association of Japanese Geographers
1.Introduction
The amount of surface runoff is influenced by land use and land cover, especially in forested areas. Shi et al. (2007) also pointed out that human activities, land use changes, and urban growth affect runoff characteristics. Mohd &Mansorr (1999) used data from the Landsat Thematic Mapper (TM) to map predicted flood areas and determine the amount of runoff of each sub watershed for the Klang Valley in Malaysia. The SCS TR55 model has been widely used but not for watershed management in Myanmar. We applied the model using the curve number method to estimate surface runoff of the Bago watershed. In the past decade, the Bago River basin has experienced rapid changes in land cover. The main purpose of our study is clarification for the surface runoff characteristics based on land cover change analysis.
2.Methodology
Landsat5 of 1990 and the landsat 7ETM+(2000) was used for surface runoff for the same period was simulated using the U.S. SCS TR55 Model.
The SCS TR55 model can be represented as
Q=(P-0.2S)2/P-0.8S P>0.2S (1)
Q=0 if P<0.2S
Q is runoff (m3/s), P is the amount of rainfall (mm), and S is the potential maximum retention after runoff begins. The watershed storage, S, and the curve number,CN,are related as follows,
S=25400/CN-254 (2)
When using the SCS TR-55 model, a curve number is assigned to portion of a watershed based on soil type, land use/cover type, hydrological soil classification, and antecedent moisture condition.
3. Results
On the basis of our visual interpretation of the satellite imagery, we classified land use/cover of the study area into six categories: closed forest, open forest, scrub and grass, agriculture, settlement areas and bodies of water. From 1990 to 2000, the percentage of forested area (closed forest) to the total area fell from 35.55% to 9.18%. The percentage of degraded forest area (open forest) to the total area increased from 14.76% and 31.7% over the same time period. Father more we compared maximum flood peak discharge at 1%, 2%, 5% and 50% rainfall probability. Our analysis showed the total surface runoff volume increased from 478.41*10000 m3 in 1990 to 516.25 *10000 m3 in 2000. The study area experiences annual flooding in June, July, and August. The mean monthly rainfall of these three months is 3057 mm, or 61.26% of the yearly total. We calculated the surface runoff volume during June, July, and August of 1990 to be 665.02 * 10000 m3 (63.28% of the total) and that of 2000 to be 628.86 *10000m3 (69.9%). Monthly and total surface runoff volume has gradually increased, and it is directly related to the changing land cover conditions.
