| Forests are the mainstay of ecosystems,and their photosynthesis plays an important role in mitigating the greenhouse effect Forest biomass and net primary productivity(NPP)provide key data for the carbon cycle of forest ecosystems.They are important conditions for assessing carbon sink capacity of forest ecosystems,important functional indicators of ecosystems,the basis and important content of ecosystem economic value research.Natural ecosystems and environmental protection should be strengthened,and long-term monitoring and assessment of changes in forest ecosystems should be carried out in order to actively respond to the national policy of forest protection.In order to accurately estimate regional biomass and achieve dynamic monitoring and estimation of forest NPP.In this study,Dailing District,Yichun City,Heilongjiang Province was used as the research area,and combined with GLAS waveform data,MODIS multi-spectral data and field survey data to establish a biomass model to calculate the biomass of the study area.And then based on the 3-PG(Physiological Principles Predicting Growth)model,forest NPP is predicted and dynamically monitored.The research provides a reference for the accurate estimation of biomass and it provides a scientific basis for understanding the changes of forest vegetation NPP and the dynamic changes of ecosystem carbon storage in Northeast China.The method achieves a continuous high forest biomass in the whole region by establishing a GLAS forest average canopy height estimation model,a BP neural network estimation model for regional forest average canopy height,and a model of plot biomass and forest average canopy height.And by this method,it is possible to make up for the defect that the GLAS data cannot be estimated as the average canopy height of the continuous forest in the region due to the stripe distribution of the light spots.The 3-PG model was used to simulate the annual growth of net primary productivity of forest vegetation in the coniferous forest,broad-leaved forest and coniferous and broad-leaved mixed forest in 2014-2044 and the monthly growth of NPP in 2014-2019,thus to realize the dynamic monitoring research of forest NPP.The main findings are as follows:(1)By combining GLAS waveform data with MODIS multispectral data,the established BP neural network estimates the forest average canopy height model R2=0.8987,RMSE=0.794.It is indicated that the method can better estimate regional continuous forest canopy height and achieve high-precision estimation of the average canopy height and biomass of the regional forest.(2)The R2 of coniferous forest,broad-leaved forest,coniferous and broad-leaved mixed forest plot and forest average canopy height model were 0.826,0.803,0.794,and the RMSE were 0.613,0.731,0.344.It is indicated that the average canopy height-biomass model established in this study can estimate the regional forest biomass through the average canopy height.(3)By analyzing the results of 3-PG operation,it is found that the overall NPP of forests in the Dailing District shows an increasing trend during 2014-2044,but the growth rate decreases with the growth of the year.The monthly growth of NPP during the vegetation growth period in 2014-2019 is cyclical.From May to September,NPP is the trend of increasing first and then decreasing,but the NPP in August is lower than the NPP of two months before and after.The total NPP of the Dailing District from May to September in 2015 was 30.725×105 t.The annual and monthly values of NPP were expressed in the three forest stands:young forest>middle age forest>near mature forest>mature forest>overmature forest. |
PDF Full Text Request