本文選題：遙感 + 水稻��； 參考：《浙江大學(xué)》2016年碩士論文

【摘要】：“國以民為本,民以食為天”,在當(dāng)今風(fēng)云突變的氣候條件和經(jīng)濟(jì)形勢下,中國及全球主要產(chǎn)糧區(qū)糧食作物生長狀況監(jiān)測和產(chǎn)量估算等措施的及時有效執(zhí)行,密切關(guān)乎到糧食安全問題。我國作為世界第一人口大國,水稻第一生產(chǎn)和消費大國,為針對水稻的生長狀況的監(jiān)測研究提出了高要求。在此背景之下本研究以江蘇省水稻遙感監(jiān)測作為研究目標(biāo),利用遙感數(shù)據(jù)在客觀及時性、宏觀尺度、經(jīng)濟(jì)效益等方面上的優(yōu)勢,提取水稻面積,估算水稻產(chǎn)量,為更合理制定農(nóng)業(yè)政策、宏觀調(diào)控糧食價格、緩解糧食危機(jī)提供可能。本文利用時間序列MOD09A1數(shù)據(jù)提取江蘇2004-2010年的水稻面積,并利用中等空間分辨率Landsat 8 OLI數(shù)據(jù)提取江蘇省水稻種植區(qū)域,以其為準(zhǔn)確值,分析基于低空間分辨率MOD09A1數(shù)據(jù)江蘇省水稻播種區(qū)域識別結(jié)果空間匹配狀況的準(zhǔn)確性。同時以2004-2006三年水稻生育期EVI與產(chǎn)量統(tǒng)計數(shù)據(jù)構(gòu)建估產(chǎn)模型,估算江蘇省2007年各地級市水稻單產(chǎn),并將市域尺度的水稻單產(chǎn)轉(zhuǎn)換到單個象元點尺度上,實現(xiàn)水稻單產(chǎn)空間化；基于改進(jìn)的兩葉直角雙曲線模型,利用氣象數(shù)據(jù)最低氣溫、最高氣溫、相對濕度和向下短波輻射,結(jié)合MOD15A2葉面積指數(shù)數(shù)據(jù),計算得到水稻生育期GPP值,進(jìn)而估算江蘇省水稻產(chǎn)量。本文的主要成果和創(chuàng)新點：1)時間序列MODIS數(shù)據(jù)提取的水稻面積,在空間分布上有一定準(zhǔn)確度,但總體面積偏低；修改水稻提取算法參數(shù),使MODIS數(shù)據(jù)水稻面積提取與統(tǒng)計數(shù)據(jù)持平,空間匹配上存在較大誤差,特別是在南京和啟東等地,但主要產(chǎn)量區(qū)都有所提取。2)水稻估產(chǎn)結(jié)果在地級市尺度上,精確度維持在95%左右,在此基礎(chǔ)上,將水稻單產(chǎn)值轉(zhuǎn)化到單象元尺度上,細(xì)化了水稻估產(chǎn)結(jié)果,可以更直觀的了解稻田生產(chǎn)力信息。3)增加25℃最大羧化速率參數(shù)(Vcmax,25)以構(gòu)建了日尺度兩葉直角雙曲線模型估算每日GPP,并用于區(qū)域水稻估產(chǎn)。
[Abstract]:"the country is based on the people and the people take food as the sky." under the sudden climatic conditions and economic situation, the measures such as monitoring the growth status of grain crops and estimating the output of grain crops in major grain producing areas in China and the world are implemented in a timely and effective manner. It is closely related to food security. As the largest population country in the world and the largest producer and consumer of rice in the world, China has put forward high requirements for monitoring and research on the growth of rice. Under this background, this research takes the rice remote sensing monitoring in Jiangsu Province as the research goal, utilizes the remote sensing data in the objective timeliness, the macroscopic scale, the economic benefit and so on superiority, extracts the rice area, estimates the rice yield, etc. For a more rational formulation of agricultural policies, macro-control of grain prices, to ease the food crisis possible. The rice area of Jiangsu Province from 2004 to 2010 was extracted by time series MOD09A1 data, and the area of rice cultivation in Jiangsu Province was extracted by Landsat 8 OLI data with medium spatial resolution, which was regarded as the accurate value. Based on low spatial resolution MOD09A1 data, the accuracy of spatial matching of rice planting area recognition results in Jiangsu Province was analyzed. At the same time, based on the EVI and yield statistical data of three years of rice growth period from 2004 to 2006, the yield estimation model was constructed, and the rice yield per unit yield in Jiangsu Province in 2007 was estimated, and the rice yield in city scale was converted to a single pixel point scale to realize the spatialization of rice yield per unit yield. Based on the improved two-leaf right-angle hyperbolic model, the GPP value of rice growing period was calculated by using meteorological data, such as minimum temperature, maximum temperature, relative humidity and downward shortwave radiation, combined with MOD15A2 leaf area index data. And then estimate the yield of rice in Jiangsu Province. The main achievements and innovations of this paper are: 1) the rice area extracted by time series MODIS data has some accuracy in spatial distribution, but the total area is on the low side, and the parameters of rice extraction algorithm are modified to make the rice area extraction from MODIS data equal to the statistical data. There is a large error in spatial matching, especially in Nanjing and Qidong, but the main yield areas have been extracted. 2) the precision of rice yield estimation is about 95% on the municipal scale. On this basis, The single output value of rice was transformed to single pixel scale, and the estimated yield of rice was refined. The maximum carboxylation rate was increased by 25 鈩，

本文編號：1930723