MAPPING THE DISTRIBUTION AND BIOMASS OF BAMBOO IN THE FOREST UNDER-STOREY OF QINLING MOUNTAINS, A REMOTE SENSING APPROACH

M. Bian1, T. Wang2, Y. Liu1, T. Fei1

1 - School of Resource and Environmental Science, Wuhan University, Wuhan, China

2 - ITC, Natural Resource Management, Enschede, The Netherlands

doubtanything@gmail.com

 

In our study area -- the Foping Natural Reserve, bamboo -- the main food of Giant Panda is growing under the forest canopy. In order to investigate the Giant Panda carrying capability of the area, we need to know how bamboo is distributed in the area, and what its biomass distribution. However, mapping understorey vegetation that cannot be seen in a remote sensing image is an impossible mission for interpreting the image directly.

 

In this thesis, methods for mapping the distribution and biomass of understorey bamboo in the study area are presented. First, the relationship between the environmental factors and the corresponding bamboo density is analyzed by ground sampling data with statistical methods. We found that for Bashania fargesii, the above storey light climate, the slope gradient and the aspect influenced the bamboo density most, while for Fargesia qinlingensis, the under storey light climate, elevation and aspect make most significant contributions to the bamboo’s density.

 

Then, two multivariable models were built to calculate the density of two bamboo species from their environmental factors. From this part, we can see both of the two bamboos are light dependent. They both grow in relatively flat terrain. The Fargesia qinlingensis is distributed in high elevation area, where the environmental temperature is lower than that of Bashania fargesii.

This study employed the light climate for bamboos as a key environmental factor for bamboo species.

 

Both above storey light climate and under storey light climate in the study area is mapped. And when we integrate it with other topographical factors by a general linear model, we can map the bamboo that distributed understory by mapping the niche where bamboo grows. At the same time, by making use of our multivariable model developed in this thesis, we can predict the bamboo density and biomass across the whole study area. The result shows a satisfactory mapping accuracy (78.33% for Bashania fargesii and 78.53% for Fargesia qinlingensis.