Estimation of Net Primary Production (NPP) of Inner Mongolia in China
Abstract
1. Introduction In the latter part of 1970's, the need for more precise calculation of the fixed-quantity of global land vegetation was emphasized. This data is necessary for estimating carbon income and expenditure at a global level. Research at the Mauna Loa volcano has clearly shown that the density of CO2 in the atmosphere is increasing. This increase is caused mainly by changes in human activities and the respiration of plants and animals. At present, however, the value of CO2 income and expenditure as calculated for human activities does not agree with the value thought to be contained in the marine and terrestrial carbon sinks. Clearly the value of primary production needs to be measured more precisely on a global scale. The use of satellite data immediately enables application at a global level, leading to higher precision of estimation when analyzing ecosystem models. In this study, we analyzed and compared Hohhot and Naiman, two regions in Inner Mongolia. In situ observation data (biomass and reflection data for each type of vegetation) was collected from 1999 to 2002. The results of these ground observations were then compared to the results from wide area measurement of vegetation index utilizing Terra/MODIS data 2. Application to satellite data The MODIS Surface Reflectance product (MOD09), with resolution of 250m, was utilized from April to November of 2002. MOD09 did atmosphere correction and geometric correction. Bands 1 (RED : 620-670nm) and 2 (NIR : 841-876nm) from MOD09 were used to produce a NDVI image. In addition, to remove the influence of cloud cover, monthly vegetation index images for May to September were generated using the Temporal Window Operation method (TWO : Park et al.1999), with the mid day of each month designated as a representative day. 3. Conclusion In this study, we estimate Net Primary Production (NPP) for a semiarid region of northern China using satellite data. An area in which pasturage is prohibited was studied in 1999 to 2002. Ground observations were used to establish vegetation classes, and to measure biomass and spectrum. Then, the relation between biomass and various vegetation indexes such as NDVI, SAVI, RVI and EVI was investigated using regression analysis. The results showed the greatest coefficient of correlation to be between biomass and NDVI. We estimated a biomass of Inner Mongolia from maximum yearly-vegetation index, which is provided by satellite observation.As a result, we knew that the grassland was distributed between 50-200 g/m2 mainly.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2009
- Bibcode:
- 2009AGUFMGC23A0749P
- Keywords:
-
- 5464 PLANETARY SCIENCES: SOLID SURFACE PLANETS / Remote sensing;
- 9320 GEOGRAPHIC LOCATION / Asia