Application Of Big Data In Estimating Vegetation Net Primary Productivity

Terrestrial ecosystems are the material basis for human survival.Vegetation is an important part of terrestrial ecosystems.It plays a significant role not only in material flow and energy cycle,but also in reducing greenhouse gases and maintaining climate stability.As an important part of the surface carbon cycle,the net primary productivity(NPP) of vegetation not only directly reflects the production capacity of vegetation under natural environmental conditions,but also reflects the ability of vegetation to fix CO2 in the atmosphere through photosynthesis.The net primary productivity of vegetation is considered to be an important link in the carbon cycle and carbon balance process of terrestrial ecosystems,and is the main factor and main indicator for determining the carbon accumulation and regulation of ecological processes in ecosystems.This paper taken the hot spot of global net change of vegetation as the core,and taken the Wulaga Natural Vegetation System of Xilin Gol League in Inner Mongolia as the research object.Based on the multi-source satellite remote sensing data of Beijing Remote Sensing Company and Mengcao,the Wulagai Management Area was used.The temperature data and precipitation data of each test site were backed up by the "3S" technology,using the fresh grass quality data collected in the field and the NDVI value of the corresponding point on the multi-source satellite remote sensing image and simulating grassland productivity distribution map.The results were as follows:(1)Using the relationship between the ground sample fresh grass quality data of the research area in 2018 and the corresponding NDVI value on the image,a linear,logarithmic,exponential,polynomial,and power function estimation model were established.The results showed that the coefficient of determination of the model equation had reached a very significant level,and there was a positive correlation between biomass and NDVI value.Therefore,it could be considered that it was feasible to use NDVI to establish a remote sensing model of grassland biomass.(2)The regression results between the productivity and the measured productivity obtained through the grassland productivity estimation model showed that the linear model,the exponential model and the polynomial model had higher determination coefficients.Therefore,these three models are a better model for estimating grassland productivity.(3)The monitoring results of the grassland productivity estimation model(linear model,exponential model,polynomial model)showed that the linear function model inversion results in higher precision and stability,and was a widely applicable model.(4)From the results of the grassland productivity inversion in the study area simulated by the optimal linear function model,it could be seen that the actual productivity of the grassland class productivity in the study area was consistent.(5)This study revealed the spatial distribution characteristics and productivity of grassland biomass in the study area.The research results could provide a scientific basis for rational utilization and protection of grassland resources and maintain grassland ecological balance in the region,and produce and manage local agriculture and animal husbandry.Provided technical information support services for local agriculture,animal husbandry production,management and decision-making...