We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
基于 MODIS 和 CLCD 数据的黄土高原土地利用 变化检测及其对植被碳吸收模拟的影响.
- Authors
吴巧丽; 张鑫阳; 蒋捷
- Abstract
Significant land-use and land-cover change (LUCC) has occurred in the Loess Plateau in recent 20 years, which has made an important contribution to the carbon sink of the terrestrial ecosystem. The accuracy of the representation of the temporal variation in LUCC is one of the most important factors affecting the accuracy of estimation of carbon exchange between terrestrial ecosystem and atmosphere. In this paper, the MODIS LULC product (i, e., MCD12Q1) and the CLCD LULC product with relatively high accuracy were introduced for comparative study. Firstly, the temporal and spatial change pattern of LUCC in the Loess Plateau from 2001 to 2019 was analyzed based on MCD12Q1 and CLCD LULC products. Then, the spatial and tempo- ral variations in GPP in the Loess Plateau from 2001 to 2019 were simulated by a process-based terrestrial biosphere model named Farquhar GPP model (FGM). Finally, the impact of classification errors of the MCD12Q1 time series data on GPP estimation was analyzed quantitatively. The results indicated that MCD12Q1 data incorrectly displayed a LUCC pattern of "increase of arable land and decrease of grassland" in the Loess Plateau region from 2001 to 2019. This could possibly be explained by the confusion in classifying arable land and grassland in MCD12Q1 data, In contrast, time-series CLCD LULC data correctly captured the LUCC variation pattern of "decrease of arable land, increase of grassland, and increase of forest land" in the Loess Plateau. The FGM simulation results based on CLCD LULC data showed that the annual total GPP of the terrestrial ecosystem in the Loess Plateau was about 5 210.2 Tg · C · a-1. From 2001 to 2019, the annual total GPP of the Loess Plateau has in- creased significantly (6.4 Tg · C · a-2. P<0.001) under the implementation of the Grain for Green Project. If the MCD12Q1 time-series data from 2001 to 2019 is used to drive the FGM model, it will cause an overestimation of approximately 9.6% of the annual total GPP and an overestimation of nearly 22.7% of the inter-annual variation trend of GPP. The research results are of great significance for scientific understanding of the changes in the structure and function of vegetation in the terrestrial eco- system of the Loess Plateau, formulation of strategies for major national ecological engineering and sustainable development.
- Subjects
CARBON cycle; FORESTS &; forestry; SUSTAINABLE engineering; ECOLOGICAL engineering; ARABLE land; BIOSPHERE
- Publication
Geography & Geographic Information Science, 2023, Vol 39, Issue 5, p30
- ISSN
1672-0504
- Publication type
Article
- DOI
10.3969/j.issn.1672-0504.2023.05.005