14:15 〜 14:30
[AHW22-14] Effects of Forest Growth in Different Vegetation Communities to Forest Catchment Water Balance
キーワード：Forest vegetation, Forest growth, evapotranspiration, SWAT, remote sensing
Forests are very important for adjusting the dynamic balance of the hydrological cycle, and this balance will be affected by the growth process of the forest. Forest growth changes in the dynamic balance of the water cycle mainly come from two aspects, the first aspect is canopy density changed in the four seasons which have widely discussed in previous research. The second aspect is long-term forest growth which rarely discussed compare with the first aspect.
Modeling has always been a common method to study the hydrological cycle process, and the catchment hydrological model led by Soil & Water Assessment Tool (SWAT) are widely used. However, at the catchment scale, forests are usually mixed with evergreen vegetation and deciduous vegetation. The composition of vegetation communities in different mixed forests will inevitably be different but the composition difference usually underestimated or ignored. In this study, we aim to design a method and build a more credible model in catchment-scale which could reflect growth characteristics of different vegetation communities of forests in long-term forest growth and seasonal changes, and focused on study the different vegetation communities forest growth impact forest hydrological processes including long-term growth and seasonal growth aspects.
To build the model, we combined Landsat, MODIS and drone remote sensing methods to obtain the long-term and seasonal canopy change characteristics of the four forest regions with different vegetation communities in the Yamato River catchment from the 1980s to the present. We demonstrate that: (1) using remote sensing methods to supplement the SWAT plant growth database and vegetation characteristic parameters can significantly improve the efficiency coefficient of the model of simulating evapotranspiration and streamflow; (2) Forest evapotranspiration increased by about 10% during the study period, and there was a significant difference in different vegetation communities of hydrological processes.
Modeling has always been a common method to study the hydrological cycle process, and the catchment hydrological model led by Soil & Water Assessment Tool (SWAT) are widely used. However, at the catchment scale, forests are usually mixed with evergreen vegetation and deciduous vegetation. The composition of vegetation communities in different mixed forests will inevitably be different but the composition difference usually underestimated or ignored. In this study, we aim to design a method and build a more credible model in catchment-scale which could reflect growth characteristics of different vegetation communities of forests in long-term forest growth and seasonal changes, and focused on study the different vegetation communities forest growth impact forest hydrological processes including long-term growth and seasonal growth aspects.
To build the model, we combined Landsat, MODIS and drone remote sensing methods to obtain the long-term and seasonal canopy change characteristics of the four forest regions with different vegetation communities in the Yamato River catchment from the 1980s to the present. We demonstrate that: (1) using remote sensing methods to supplement the SWAT plant growth database and vegetation characteristic parameters can significantly improve the efficiency coefficient of the model of simulating evapotranspiration and streamflow; (2) Forest evapotranspiration increased by about 10% during the study period, and there was a significant difference in different vegetation communities of hydrological processes.