本文選題：遙感監(jiān)測(cè) + 農(nóng)業(yè)土地利用變化�。� 參考：《青海師范大學(xué)》2015年碩士論文

【摘要】：農(nóng)業(yè)土地利用是影響地表水平衡的重要因素,華北平原作為我國(guó)棉糧果重要生產(chǎn)基地,在過(guò)去的幾十年中由于抽取地下水灌溉造成嚴(yán)重的水資源短缺。因此準(zhǔn)確地提取華北平原不同作物的空間分布信息可為作物耗水估算、水資源可持續(xù)化利用和農(nóng)業(yè)資源配置提供基礎(chǔ)數(shù)據(jù)。為了解決多年份、不同作物遙感監(jiān)測(cè)種植面積變化過(guò)程中,不同的訓(xùn)練樣本或分類(lèi)規(guī)則不能較好地反映作物種植面積時(shí)空格局變化的問(wèn)題。本文首先根據(jù)研究區(qū)(華北平原)農(nóng)業(yè)氣象觀測(cè)站提供的主要農(nóng)作物物候觀測(cè)資料獲得主要農(nóng)作物典型物候期特征,利用Savitzky-Golay對(duì)MODIS NDVI時(shí)間序列數(shù)據(jù)進(jìn)行重建,進(jìn)而提取不同作物的NDVI時(shí)序曲線(xiàn),結(jié)合二次差分法獲取的復(fù)種指數(shù),探討了基于不同作物生育期的差異,采用CART(Classification and Regression Tree)算法對(duì)華北平原多年份、不同作物的遙感分類(lèi)方法進(jìn)行可行性分析,進(jìn)而提取了2000—2013年華北平原主要作物冬小麥、玉米、棉花和果樹(shù)的種植面積,并參考縣級(jí)的農(nóng)業(yè)統(tǒng)計(jì)數(shù)據(jù)和實(shí)地樣本數(shù)據(jù)進(jìn)行精度評(píng)價(jià),最后分析了華北平原近14年間主要作物空間格局變化及其動(dòng)因。得出以下結(jié)論:(1)經(jīng)檢驗(yàn),近13年的冬小麥、玉米和棉花的遙感監(jiān)測(cè)種植面積與農(nóng)業(yè)統(tǒng)計(jì)面積相關(guān)系數(shù)分別為0.78、0.71和0.83(置信水平都為95%),具有較高的相關(guān)性。在參考實(shí)地樣本數(shù)據(jù)與分類(lèi)結(jié)果建立的混淆矩陣顯示,14年間的分類(lèi)結(jié)果Kappa系數(shù)在0.72~0.96之間;總體精度在79%~97%之間,85%以下年份僅出現(xiàn)2年,從整體上來(lái)看,近14年的分類(lèi)精度可行。(2)從作物的空間分布特征上看,華北平原的冬小麥-夏玉米主要分布于山東省和河南省的黃灌區(qū)及太行山山前平原區(qū);林果主要分布于晉州-趙縣-深州一帶、泊頭-滄縣一帶、樂(lè)亭-昌黎一帶和信陽(yáng)-惠民-濱州一帶以及環(huán)北京周邊地區(qū);棉花集中分布于華北平原中部。在時(shí)間變化上,華北平原冬小麥、玉米面積均呈下降趨勢(shì),分別由2000年的553.23萬(wàn)hm2縮減至2013年的530.33萬(wàn)hm2和由2000年的642.48萬(wàn)hm2縮減至2013年的569.54萬(wàn)hm2,其中以河北平原北部縮減趨勢(shì)最為顯著,天津-廊坊一帶和保定地區(qū)甚至出現(xiàn)冬小麥主要種植區(qū)“棄耕”的現(xiàn)象。華北平原棉花14年的平均種植面積約84.01萬(wàn)hm2,其中河北平原南部棉花面積占研究區(qū)棉花總面積的一半以上(53%),年際間波動(dòng)較大,整體上呈增加趨勢(shì)。林果平均種植面積約63.55hm2,四個(gè)分區(qū)在近14年間整體上均呈增加趨勢(shì),其主要影響因素是經(jīng)濟(jì)收益和政府導(dǎo)向。(3)從糧食作物/經(jīng)濟(jì)作物種植轉(zhuǎn)換的角度來(lái)看,華北平原的糧食/林果轉(zhuǎn)換區(qū)主要集中在有林果種植歷史的區(qū)域,如:環(huán)北京周邊、唐山、滄州和山東的濱州一帶,轉(zhuǎn)換區(qū)面積約為95.53萬(wàn)hm2,占華北平原耕地面積的11.15%。糧食/棉花轉(zhuǎn)換區(qū)主要集中于華北平原南部的產(chǎn)棉區(qū)周邊及吳橋-東光-南皮一帶,轉(zhuǎn)換區(qū)面積約為64.89萬(wàn)hm2,占華北平原耕地面積的7.58%。(4)基于多年的分類(lèi)結(jié)果來(lái)統(tǒng)計(jì)作物的種植概率進(jìn)而表現(xiàn)不同土地條件和其他人為因素對(duì)作物的選擇,能較好地反映作物種植的主產(chǎn)區(qū),這為農(nóng)業(yè)土地利用變化研究中分析作物種植面積變化提供一種新思路。
[Abstract]:Agricultural land use is an important factor affecting the balance of surface water. The North China Plain is an important production base of China's cotton and fruit and fruit. In the past few decades, the extraction of groundwater irrigation has caused serious water shortage. Therefore, the accurate extraction of spatial distribution information of different crops in the North China plain can be used to estimate the water consumption of crops, and the water resources can be held. In order to solve the multiple year, the different training samples or classification rules can not reflect the change of the spatial and temporal pattern of crop planting area in order to solve the multiple year. The main crop phenological data obtained the characteristics of the typical phenological phase of main crops, rebuilt the MODIS NDVI time series data by Savitzky-Golay, then extracted the NDVI time sequence curve of different crops, combined with the compound index obtained by the two difference method, and discussed the difference based on the growth period of different crops, and used CART (Classificati). On and Regression Tree) algorithm is used to analyze the feasibility of remote sensing classification of different crops in the North China Plain for many years, and then the planting area of winter wheat, corn, cotton and fruit trees of the North China Plain from 2000 to 2013 is extracted, and the precision evaluation of the agricultural statistics and field samples at the county level is carried out, and the final analysis is made. The changes in the spatial pattern of main crops in the North China Plain during the last 14 years and their reasons were obtained. (1) the correlation coefficient between the monitoring planting area and the agricultural statistical area of the winter wheat, corn and cotton in the last 13 years was 0.78,0.71 and 0.83 (the confidence level was 95%), respectively. The obfuscation matrix established with the classification results showed that the Kappa coefficient of the classification results of 14 years was between 0.72~0.96; the overall accuracy was between 79%~97% and 85% years only 2 years. From the overall point of view, the classification precision was feasible for nearly 14 years. (2) from the spatial distribution characteristics of the crops, the winter wheat and summer maize in the North China Plain were mainly distributed in Shandong. The Yellow irrigated area of province and Henan province and the plain area of Taihang Mountain in front of the mountain; the fruit is mainly distributed in Jinzhou Zhaoxian County Shenzhou area, Botou Cangxian area, Leting Changli area, Xinyang Huimin Binzhou area and periphery of Beijing. The cotton is concentrated in the middle of North China Plain. In time change, the area of Winter Wheat in North China Plain and the area of corn are all present The downward trend was reduced from 5 million 532 thousand and 300 hm2 in 2000 to 5 million 303 thousand and 300 hm2 in 2013 and from 6 million 424 thousand and 800 hm2 in 2000 to 5 million 695 thousand and 400 hm2 in 2013, among which the most significant reduction trend was in the northern Hebei plain. The phenomenon of "abandonment" in the main area of Winter Wheat in the Tianjin Langfang area and Baoding area. The North China plain cotton 14 The average planting area of the year is about 840 thousand and 100 Hm2, of which the cotton area in the southern part of the Hebei plain accounts for more than half (53%) of the total cotton area in the study area. The interannual fluctuation is larger and the overall growth trend is increasing. The average planting area of the forest fruit is about 63.55hm2, and the four zoning is increasing in the last 14 years. The main influencing factors are economic returns and the main factors. (3) from the point of view of the conversion of grain crops / crop planting, the grain / fruit conversion area of the North China Plain is mainly concentrated in the region with the history of forest fruit planting, such as the surrounding area around Beijing, Tangshan, Cangzhou and Binzhou in Shandong, the area of the conversion area is about 955 thousand and 300 Hm2, which accounts for the grain / cotton rotation of the cultivated land in the North China Plain. The change area is mainly concentrated in the vicinity of the cotton producing area in the south of the North China Plain and the Wuqiao Dongguang Nanpi area. The area of the conversion area is about 648 thousand and 900 Hm2, and the area of the cultivated land in the North China Plain is 7.58%. (4). Based on the years of classification results, the planting probability of the crops is counted and the selection of different land conditions and other human factors is better. It reflects the main production area of crop planting, which provides a new idea for analyzing the change of crop acreage in the study of agricultural land use change.
【學(xué)位授予單位】：青海師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：S127

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 潘學(xué)鵬;華北平原主要作物遙感提取及時(shí)空變化研究[D];青海師范大學(xué);2015年

，

本文編號(hào)：1996753