Font Size: a A A

The Extraction Of Crop Planting Structure And Its Dynamic Monitoring Of Huangshui Basin From 2000 To 2015 Combined With STARFM Model

Posted on:2019-12-27Degree:MasterType:Thesis
Country:ChinaCandidate:L TianFull Text:PDF
GTID:2429330566980032Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Huangshui basin is one of the important crop planting area in Tibetan Plateau.Because of climate change and the governmental policys,such as Western Great Development and the Project of Returning farmland to forestland or grassland,land use in the basin had undergone tremendous changes.The purpose of this paper were to study the agricultural planting structure and how it evolved under the background of the land use change.In this paper,MODIS-NDVI data and Landsat data(TM/OLI)and the site data of crops in Huangshui basin were used as data sources.Combining adaptive spatio-temporal data fusion model(STARFM),asymmetric Gauss function(AGFF)for time series data smoothing and spectral similarity theory,this study extracted the distribution informations of three kinds of main crops(spring wheat,highland barley and rape)in Huangshui basin from 2000 to2015 with 5 years' interval,and discussed the spatial distribution and temporal evolution of the crop planting structure in the basin.The results were showed as following.(1)The adaptive spatial data fusion model(STARFM),used to fuse MODIS-NDVI and Landsat-NDVI(TM/OLI-NDVI)data,can improve obviously the spatial resolution of MODIS-NDVI data,enhancing the ability for monitoring the land cover change."Double high" NDVI data,stemming from the model,makes it possible to extract crop planting structure of valley,like Huangshui basin.This,meanwhile,provides data processing ideas for the application of MODIS series data to small scale areas and one owning dramastical spatial heterogeneity.(2)The asymmetric Gauss function fitting(AGFF)method has good data fidelity in the long time series NDVI data fitting,while smoothing the original data noise,it also has a good inhibitory effect on the noise introduced by the STARFM model,which contributed significantly positive effect to the further study.(3)With high Google Earth history images as a reference,the overall accuracy of extracting result of crop planting structure in Huangshui basin was more than 87% and Kappa coefficient was above 0.80.Generally speaking,this method is effecitive in differentiating different corps in Huangshui basin.In 2000,2005,2010 and 2015,spring wheat extraction user accuracy were 97%,91%,99% and 95% respectively;for each year of highland barley extraction user accuracy were 88%,96%,92% and 86% respectively;the highland barley extraction user accuracy was 70%,80%,85%,88% respectively.It could be seen that the method proposed in this paper had the same precision characteristics for each crop species extraction in different years,which provd that the method was robust.(4)According to the statistical analysis of crop planting area extracted by the method of this article,we found that from 2000 to 2010,the rank of 3 main crops in absolute acreage was relatively stable,that the area of rape was the top one,spring wheat occupied the second position,and barley was the last one.During period from 2010 to 2015,however,a drastic change of agricultural planting structure had occured.A substantial expansion of non-wide-area crop type had been croped highland barley,which made the planting area of highland barley become the major crop in the basin;wide-area spring wheat acreage change slightly;rape planting area had been shrunk dramatically,from the biggest one during period from 2000 to 2010 to the last one in 2015.The agricultural planting structure in the basin in 2015 was that the highland barley had biggest area,followed by spring wheat,and rape was the smallest.Generally,the changes of agricultural planting structure of the Huangshui River basin were obvious during period from 2000 to 2015,especially the variety of the non-wide area crop species,Highland barley,and wide-area oilseed crops,rape.(5)Through the transfer matrix of distribution in 2000,2005,2010 and 2015 in the basin constructed by unused land,cultivated land and three main crops,we found that there were significant relationships between the evolution of basin agricultural planting structure,agriculture policies and national economic development within the region.After joining the WTO,the national economy had entered a new development stage,and non-agricultural economy was boosting sharply.Subsequently,the probability of farmers' participation in non-agricultural economic activities increased parellelly,which made the non-planted farmland expand distinctly during the period from 2000 to 2005,expecially spring wheat with a large need of irrigation and labour.The abolishment of agricultural tax and the subsidies for agriculture and other agricultural policies implemented in 2006,which increase the investment in agriculture,returned a large number of non-planted farmand into farming land which make spring wheat planting area returned to the level in 2000 generally.With the the development of the market economic system,wide–area crops,including wheat and rape,whose liquidity in market has been increased.However,compared with other area suited for them,Huangshui basin lacked of comparative advantage.Therefore,the advantages of regional natural environment were turned to make full use of to cultivate characteristic agriculture.Spreading of selective breeding,providing subsidies for grain,means of production and purchase of agricultural machinery and so on,were advised to make a substantial expansion of barley planting.
Keywords/Search Tags:Huangshui basin, planting structure, dynamic monitoring, spatial and temporal adaptive reflectance fusion model
PDF Full Text Request
Related items