Simulation Of The Dynamic Carbon Cycle And Evapotranspiration In Forest Based On Remote Sensing

Serious climate challenge spreads all over the world in these decades. According to the temperature data from 1880, the temperature of earth surface has a trend of increasing, ten top temperatures all occurred after 1998, and 2015 had the highest temperature so far. The extreme climate and nature disasters occurred frequently influencing the human activities and commercial production significantl and attracted more and more attention of many countries. Developing a low carbon economy and promoting the energy conservation and emission reduction is becoming a complex topic involving the politics, economy, international law, among others.Forest, as the main composition of terrestrial ecosystem, accounts for one third of the terrene area and stores 50%-60%of the whole bio-carbon in terrestrial ecosystem. Forest plays an important role in carbon circulation, water circulation and energy exchange between terrestrial ecosystem and atmosphere. Northeast Forest in China is one of three largest temperate forests in the world and the biggest nature forest in China. It plays a significant role in carbon circulation, water circulation and ecosystem construction in China, even the world.How to estimating the carbon circulation and water circulation in forest fast and accurately is the key to acquire the relationship between forest ecosystem and global climate. Building the process model of dynamic carbon and water circulation based on vegetation metabolism, could explain the mechanism of vegetation metabolism process and the influence of vegetation on environment. Remote sensing technology and geographic information system help to obtain the information of district vegetation. The research in this field is not so comprehensive for Northeast forest. However, these data is urgently needed to estimate the carbon and water circulation in China. To solve these problems, two issues were studied in this thesis:firstly, the method to obtain more accurate parameters of process model, then, to modify the model for specific forest and estimate the data of northeast forest. Secondly, obtain the increment of biomass in research area. The detail research content is demonstrated as follows:1 the pretreatment and strengthen of remote sensing images, including the geometric precision correction, irradiance correction and image fusion. Inputting the time course of remote sensing data to the process model to get the LAI, NNP, among the other information about we need. The pretreatment of image is the precondition of all the work.2 By data from literatures and the compatible model, the biomass of single species from 12 sample plots and the subsequent biomass of whole crop were calculated respectively. The biomass occupation of different organs of trees could be obtained then. At last, the increasing of biomass could be calculated by the difference of two years-biomass accumulation of sample plots.3 Lead Area Index (LAI) is an important index of canopy structure, reflecting the abilities of vegetation, like photosynthesis, evaporation and irradiance attenuation. LAI is one of the sensitive parameters in our process model, could influence the accuracy of the final estimated results. Obtaining a long time course of remote sensing data is the basic to study the carbon and water circulation and to estimate the LAI in a district scale. In this study, a novel instrument to monitor the dynamic LAI in forest with various terrains was invented. The instrument has a horizontal steering engine and a vertical one, which was controlled by a single chip. So the instrument could read the irradiance intensities from different angles by optical sensor and record these data to SD card. The instrument is long-term used with solar power. Then the index of canopy could be calculated according to the Bill’s Law and Miller’s Law. The detail about this instrument is described in this thesis. The performance of instrument has been compared with the commercial instrument LAI-2200 and the correlation coefficient is above 9.5.4 The parameter modification of model and improvement of algorithm. The BEPS model was used to study the district carbon and water circulation. To increase the applicability of process model to northeast forest, the parameters were modified by means of in-site measurement, referring to history data and comparison of models.5 Using the information from remote sensing data to calculate the parameter in model is an essential method to district study. But the errors of remote sensing data could be magnified in the final estimated results. To decrease this error, the model was modified by the real data from sample plots. The applicability of this model to northeast forest was also improved by this procedure.6 LAI was calculated by four dimension geometric optics model and MISR data with high temporal resolution. Obtaining time course of two-year data by remote sensing technology avoids the high load of collecting data during statistics. In this study, using four-dimensional geometric optics model to calculate the LAI was built and how to use the real data in pilot scale to detect the results of middle resolution data obtained by remote sensing was discussed.7 Inputting the results of remote sensing and corrected data of climate to process model to accomplish the estimation of carbon and water circulation.