基于MODIS-NDVI的内蒙古植被变化遥感监测
|
摘要
植被在地球系统中扮演着重要的角色,植被影响地气系统的能量平衡,在气候、水文和生化循环中起着重要作用,是气候和人文因素对环境影响的敏感指标。由于植被具有明显的年际变化和季节变化的特点,并且是联结土壤、大气和水分的自然“纽带”,在一定的程度上能代表土地覆盖的变化,所以在全球变化研究中充当“指示器”的作用。因此对植被的动态监测可以从一定程度上反映气候变化的趋势,动态监测植被状况和植被覆盖变化也一直是生态环境研究的热点之一。
本文利用2002-2006年5-8月内蒙古地区的MODIS数据进行NDVI数据的反演,分析其变化规律及其与气候因子中温度和降水的关系。同时选取了森林、草地、农耕地、沙地四种不同植被覆盖类型区域的10个子区作为研究区,分析了不同植被类型区域NDVI的月际、年际变化规律及其与气候因子的关系,得出的相关结论如下:
1)内蒙古地区植被覆盖的基本特征是由西南向东北逐渐增加,与该区降水量分布相同;适中植被区的分布面积最大,裸土区、稀少植被区、很密植被区和茂密植被区分布面积基本相当;在NDVI的数值上,林地>农耕地>草地>沙地;
2)月际变化上,5-8月NDVI不断增加,但有地区和植被类型上的差异,总的来说,NDVI变化率5-6月>6-7月>7-8月;林地NDVI的增加主要集中在5-6月,其他月份变化不大;草地NDVI的增加主要集中在6-7月;耕地的NDVI值5-8月期间稳定增长;沙地的NDVI值几乎没有增长;
3)年际变化上,2002-2006年间,锡林郭勒盟草地的波动性最大,呈增加-减少-增加-减少的变化趋势;呼伦贝尔林区除了2004年丰水年NDVI有所增加外,其他年份都有所减少;内蒙古西部地区2002-2005年间NDVI不断增加,但2006年下降;且2005-2006年期间,各种类型植被普遍有退化现象;
4)NDVI变化与月均温有很好的相关性,东北地区是月均温与NDVI相关的高值区,中部为低值区;NDVI变化与月降水量的相关性不强,但与滞后1月降水和滞后2月降水的相关性良好,与降水相关的高值区分布于内蒙古中部的草地;对于几种气候因子与NDVI的相关性来说,一般是滞后2月降水>滞后1月降水>月均温>月降水量;对于林地和草地来说,各种相关系数高纬高于低纬,对于农耕地来说,各种相关系数基本相当;对于沙地来说,各种相关系数均不高,这与其植被稀少且几乎没有变化有关。
vegetation play a important role in the Earth system,it's have impact on energy balance of earth—atmosphere system, and have significant effect on climate and hydrology and biochemistry cycle. At the same time, Vegetation is a sensitive indicator of climate and human factors influencing environment. The vegetation has obvious inter-annual change and seasonal changes, and has being a natural "bond" linking the soil, air and water, and it can represent changes in land cover to a certain extent, so that vegetation act as a "indicator " in the study of global changes. Because of that vegetation dynamics monitoring can reflect the trend of climate change in certain extent, monitoring vegetation conditions and vegetation coverage in dynamic mode also has been a hotspot on the ecological environment research.
In this paper, we use year 2002-2006 MODIS data from May to August do NDVI data of inner Mongolia region inversion. After that we analyze its variety regulation and relationship with temperature and precipitation in climatic factors. At the same time, we selected 10 sub-regional areas of four different types of vegetation coverage area- forests, grassland, farmland, sandy land- as a study area, we analyze different vegetation type areas NDVI changes by year and by month and its relationship of climatic factors, The conclusions are as follows:
1)The basic characteristics of the Inner Mongolia vegetation coverage is gradually increasing from southwest to northeast, Equally with precipitation distribution of the same region, the distribution of moderate vegetation region are the largest area; bare soil area, sparse vegetation area, very dense vegetation area and dense vegetation area have equal Distribution proportion; in the number of NDVI,forests> farmland>grassland>sandy land;
2)On the occasion of changes by month, in May to August NDVI is overall increase, but there are differences in region and vegetation types. In general, the rate of change is May-June> June-July> July-August; NDVI of forests increased mainly concentrated in May to June, the others is in little change; and NDVI of lawn increased mainly concentrated in the June-July; NDVI of farmland grow steadily in May to August; sandy land almost have no growth.
3) To the inter-annual changes, in the years 2002-2006, the grassland of Xilin Gole league have the largest fluctuation,it shows a Trend like increase-decrease-increase-decrease; the forests of Hulun Buir decrease at every year but 2004,because this year is a abundant water year; in Western Inner Mongolia, NDVI increased every year between 2002-2005, but in 2006,it's decreased; and from year 2004-2006,we found general degradation of vegetation;
4)NDVI changes have a good correlation with monthly averaged temperature. the northeastern region is high-value area which related to temperature and NDVI; low-value area is in the central region and the relevance of precipitation is not strong, but it's have good correlation with lag one month precipitation and lag two months precipitation, the precipitation-related high-value areas are in the central Inner Mongolia grassland area; for the relevance of NDVI and several climatic factors, the regulation is lag two months precipitation>lag one month precipitation>monthly averaged temperature>monthly precipitation in general; for the forest and grass region, the regulation of every correlation coefficients is that the value of high latitudes region is higher than the value of low latitudes region; for the Agricultural Area, the correlation coefficients are equal; for the sandy land, every correlation coefficients are not high, this is because of its scarce vegetation coverage.
本文利用2002-2006年5-8月内蒙古地区的MODIS数据进行NDVI数据的反演,分析其变化规律及其与气候因子中温度和降水的关系。同时选取了森林、草地、农耕地、沙地四种不同植被覆盖类型区域的10个子区作为研究区,分析了不同植被类型区域NDVI的月际、年际变化规律及其与气候因子的关系,得出的相关结论如下:
1)内蒙古地区植被覆盖的基本特征是由西南向东北逐渐增加,与该区降水量分布相同;适中植被区的分布面积最大,裸土区、稀少植被区、很密植被区和茂密植被区分布面积基本相当;在NDVI的数值上,林地>农耕地>草地>沙地;
2)月际变化上,5-8月NDVI不断增加,但有地区和植被类型上的差异,总的来说,NDVI变化率5-6月>6-7月>7-8月;林地NDVI的增加主要集中在5-6月,其他月份变化不大;草地NDVI的增加主要集中在6-7月;耕地的NDVI值5-8月期间稳定增长;沙地的NDVI值几乎没有增长;
3)年际变化上,2002-2006年间,锡林郭勒盟草地的波动性最大,呈增加-减少-增加-减少的变化趋势;呼伦贝尔林区除了2004年丰水年NDVI有所增加外,其他年份都有所减少;内蒙古西部地区2002-2005年间NDVI不断增加,但2006年下降;且2005-2006年期间,各种类型植被普遍有退化现象;
4)NDVI变化与月均温有很好的相关性,东北地区是月均温与NDVI相关的高值区,中部为低值区;NDVI变化与月降水量的相关性不强,但与滞后1月降水和滞后2月降水的相关性良好,与降水相关的高值区分布于内蒙古中部的草地;对于几种气候因子与NDVI的相关性来说,一般是滞后2月降水>滞后1月降水>月均温>月降水量;对于林地和草地来说,各种相关系数高纬高于低纬,对于农耕地来说,各种相关系数基本相当;对于沙地来说,各种相关系数均不高,这与其植被稀少且几乎没有变化有关。
vegetation play a important role in the Earth system,it's have impact on energy balance of earth—atmosphere system, and have significant effect on climate and hydrology and biochemistry cycle. At the same time, Vegetation is a sensitive indicator of climate and human factors influencing environment. The vegetation has obvious inter-annual change and seasonal changes, and has being a natural "bond" linking the soil, air and water, and it can represent changes in land cover to a certain extent, so that vegetation act as a "indicator " in the study of global changes. Because of that vegetation dynamics monitoring can reflect the trend of climate change in certain extent, monitoring vegetation conditions and vegetation coverage in dynamic mode also has been a hotspot on the ecological environment research.
In this paper, we use year 2002-2006 MODIS data from May to August do NDVI data of inner Mongolia region inversion. After that we analyze its variety regulation and relationship with temperature and precipitation in climatic factors. At the same time, we selected 10 sub-regional areas of four different types of vegetation coverage area- forests, grassland, farmland, sandy land- as a study area, we analyze different vegetation type areas NDVI changes by year and by month and its relationship of climatic factors, The conclusions are as follows:
1)The basic characteristics of the Inner Mongolia vegetation coverage is gradually increasing from southwest to northeast, Equally with precipitation distribution of the same region, the distribution of moderate vegetation region are the largest area; bare soil area, sparse vegetation area, very dense vegetation area and dense vegetation area have equal Distribution proportion; in the number of NDVI,forests> farmland>grassland>sandy land;
2)On the occasion of changes by month, in May to August NDVI is overall increase, but there are differences in region and vegetation types. In general, the rate of change is May-June> June-July> July-August; NDVI of forests increased mainly concentrated in May to June, the others is in little change; and NDVI of lawn increased mainly concentrated in the June-July; NDVI of farmland grow steadily in May to August; sandy land almost have no growth.
3) To the inter-annual changes, in the years 2002-2006, the grassland of Xilin Gole league have the largest fluctuation,it shows a Trend like increase-decrease-increase-decrease; the forests of Hulun Buir decrease at every year but 2004,because this year is a abundant water year; in Western Inner Mongolia, NDVI increased every year between 2002-2005, but in 2006,it's decreased; and from year 2004-2006,we found general degradation of vegetation;
4)NDVI changes have a good correlation with monthly averaged temperature. the northeastern region is high-value area which related to temperature and NDVI; low-value area is in the central region and the relevance of precipitation is not strong, but it's have good correlation with lag one month precipitation and lag two months precipitation, the precipitation-related high-value areas are in the central Inner Mongolia grassland area; for the relevance of NDVI and several climatic factors, the regulation is lag two months precipitation>lag one month precipitation>monthly averaged temperature>monthly precipitation in general; for the forest and grass region, the regulation of every correlation coefficients is that the value of high latitudes region is higher than the value of low latitudes region; for the Agricultural Area, the correlation coefficients are equal; for the sandy land, every correlation coefficients are not high, this is because of its scarce vegetation coverage.
引文
[ 1 ] 王 连 喜 , 杨 有 林 , 气 候 变 化 和 植 被 关 系 研 究 方 法 探 讨 , 生 态 学 杂 志[J],2003,22(1),43-45
[2] 程麟生,宋振鑫,杨德保.青藏高原地表分类及其东北部辐射通量的二维数值模拟,高原气象[J], 2000.19(2),198-213.
[3] 李博,内蒙古自治区资源系列地图说明[M],北京:科学出版社,1993,14-24
[4] 陈述彭,遥感应用.地学探索(第三卷)[M],北京:科学出版社,1990,67-81.
[5] 徐冠华.三北防护林地区再生资源遥感的理论及其技术应用[M].北京:中国林业出版社. 1994,1-99.
[6] 盛永伟,陈维英等,利用卫星植被指数进行我国植被的宏观分类.科学通报[J],1995,40(1),68-71.
[7]张旭,EOS_MODIS 影像处理及其在塔里木河下游植被变化中的应用[M],新疆大学,2006
[8]马志勇,张慧,基于 MODIS 杭州地区植被变化分析,安徽农业科学[J], 2006,34(18),4618-4619
[9]郭广猛,杨丽,泾河流域近五年来植被变化分析,资源科学[J],2005,27(4),22-25
[10]张雷,宁夏全区的植被生态变化规律及其影响因素的遥感研究,中国地质大学[J],2006
[11]郭铌,植被指数及其研究进展,干旱气象[J],2003,21(4),71-75
[12]李登科,张树誉,EOS/MODIS 遥感数据与应用前景,陕西气象[J],2003,(2),37-40
[13]郭铌,李明轩,气象卫星遥感监测甘肃河西绿洲植被变化,甘肃气象[J],2002,20(1) ,33-35
[14]徐文婷,颜长珍,1998_2002年我国陆地植被变化探测研究,郑州大学学报(理学版) [J],2004,36(2),54-57
[15]李月臣,宫鹏,北方 13 省 1982 年_1999 年植被变化及其与气候因子的关系,资源科学[J], 2006,28(2),109-117
[16]中分辨率成像光谱辐射计(MODIS)摘要,新疆气象[J],2003,26(1),30-31
[17]许辉熙,杨存建,成都市区植被变化的遥感动态监测,资源开发与市场[J],2005 ,21(1),10-16
[18]罗格平,陈嘻,基于 AVHRR_NOAA 影像的天山北坡近 10a 植被变化,冰川冻土[J],2003,25(2),237-242
[19]马丽云,刘艳,MODIS 资料分析新疆植被生长的季节变化,新疆气象[J],2006,29(5),23-24
[20]邓朝平, 近 22 年来西北地区植被变化及其气候响应研究,新疆大学,2006
[21]杨建平,丁永建,长江黄河源区高寒植被变化的 NDVI 记录,地理学报[J],2005,60(3),467-478
[22]邓朝平,郭铌,近 20 余年来西北地区植被变化特征分析,冰川冻土[J],2006,28(2),686-693
[23]Koomanoff,V.A,1989.Analysis of Global Vegetation Patterns: A Comparison Between Remotely Sensed data and a conventional map. Biaogeography Research Series, Report#890201,Department of Geography, university of Maryland.
[24] 刘玉洁,杨忠东等,MODS 遥感信息处理原理与算法[M],科学出版社,2001,2-3
[25]李月臣,陈晋,基于 NDVI 时间序列数据的土覆盖变化检测指标设计,应用基础与工程科学学报[J],2005,13(3),261-275
[26]王坚,张继贤,基于 NDVI 序列影像精化结果的植被覆盖变化研究,测绘科学[J],2005,30(6),43-45
[27]陈多,基于 NOAA/AVHRRNDVI 的西藏拉萨地区植被季节变化,高原气象(副刊)[J],2003,22,145-151
[28]李云浩,李晓兵,基于遥感的 NDVI 与气候关系图式研究,中国图像图形学报[J],2002,7A(4),332-336
[29]齐述华,王长耀,利用 NDVI 时间序列数据分析植被长势对气候因子的响应,地理科学进展[J],2004,23(3),91-100
[30]邢旗,刘爱军,应用MODIS-NDVI对草原植被变化监测研究—以锡林郭勒盟为例,草地学报(增刊)[J],2005,13,15-19
[31]毕晓丽,王辉,植被归一化指数(NDVI)及气候因子相关起伏型时间序列变化分析,应用生态学报[J],2005,16(2),284-288
[32]李晓兵,史培军,中国典型植被类型 NDVI 动态变化与气温、降水变化的敏感性分析,植物生态学报[J],2000,24(3),379-382
[33]刘先华,陈佐忠,定居放牧方式下归一化植被指数(NDVI)的空间变化特征,植物生态学报[J],2000,24(6),662-666 [ 34 ] 罗 亚 , 徐 建 华 , 基 于 遥 感 影 像 的 植 被 指 数 研 究 方 法 述 评 , 生 态 科 学[J],2005,24(1),75-79
[35]张树誉,李登科,MODIS 植被指数及其在县域生态环境监测评估中的应用,粮食安全与生态环境监测[J],2002,4011-4019
[36]韩爱惠,用 MODIS 植被指数的时间序列分析提取荒漠化敏感区域的方法,林业资源管理[J],2004,(1),57-60
[37]宋小宁,赵英时,应用 MODIS 卫星数据提取植被-温度-水分综合指数的研究,地理与地理信息科学[J],2004,20(2),13-17
[38]曲辉,陈圣波,中分辨率成像光谱仪(MODIS)数据在地学中的应用前景,世界地质[J],2002,21(2),176-180
[39]黄家洁,万幼川,MODIS 的特性及其应用,地理空间信息[J],2003,1(4),20-28 [ 40 ] 罗 亚 , 徐 建 华 , 基 于 遥 感 影 像 的 植 被 指 数 研 究 方 法 述 评 , 生 态 科 学[J],2005,24(1),75-79
[41]徐希儒,遥感物理[M],北京大学出版社,2005,121-129
[42] http://zt.nmgnews.com.cn/zt1/dqw/nmggk/,内蒙古新闻网专题站-专题-大庆网站-内蒙古概况
[43]熊育久, 湖南省 EOS_MODIS 植被指数时空变化研究,中南林业科技大学,2006
[44] 朱文泉,潘耀忠,龙中华等, 基于 GIS 和 RS 的区域陆地植被 NPP 估算——以中国内蒙古为例,2005, (03),3-8
[2] 程麟生,宋振鑫,杨德保.青藏高原地表分类及其东北部辐射通量的二维数值模拟,高原气象[J], 2000.19(2),198-213.
[3] 李博,内蒙古自治区资源系列地图说明[M],北京:科学出版社,1993,14-24
[4] 陈述彭,遥感应用.地学探索(第三卷)[M],北京:科学出版社,1990,67-81.
[5] 徐冠华.三北防护林地区再生资源遥感的理论及其技术应用[M].北京:中国林业出版社. 1994,1-99.
[6] 盛永伟,陈维英等,利用卫星植被指数进行我国植被的宏观分类.科学通报[J],1995,40(1),68-71.
[7]张旭,EOS_MODIS 影像处理及其在塔里木河下游植被变化中的应用[M],新疆大学,2006
[8]马志勇,张慧,基于 MODIS 杭州地区植被变化分析,安徽农业科学[J], 2006,34(18),4618-4619
[9]郭广猛,杨丽,泾河流域近五年来植被变化分析,资源科学[J],2005,27(4),22-25
[10]张雷,宁夏全区的植被生态变化规律及其影响因素的遥感研究,中国地质大学[J],2006
[11]郭铌,植被指数及其研究进展,干旱气象[J],2003,21(4),71-75
[12]李登科,张树誉,EOS/MODIS 遥感数据与应用前景,陕西气象[J],2003,(2),37-40
[13]郭铌,李明轩,气象卫星遥感监测甘肃河西绿洲植被变化,甘肃气象[J],2002,20(1) ,33-35
[14]徐文婷,颜长珍,1998_2002年我国陆地植被变化探测研究,郑州大学学报(理学版) [J],2004,36(2),54-57
[15]李月臣,宫鹏,北方 13 省 1982 年_1999 年植被变化及其与气候因子的关系,资源科学[J], 2006,28(2),109-117
[16]中分辨率成像光谱辐射计(MODIS)摘要,新疆气象[J],2003,26(1),30-31
[17]许辉熙,杨存建,成都市区植被变化的遥感动态监测,资源开发与市场[J],2005 ,21(1),10-16
[18]罗格平,陈嘻,基于 AVHRR_NOAA 影像的天山北坡近 10a 植被变化,冰川冻土[J],2003,25(2),237-242
[19]马丽云,刘艳,MODIS 资料分析新疆植被生长的季节变化,新疆气象[J],2006,29(5),23-24
[20]邓朝平, 近 22 年来西北地区植被变化及其气候响应研究,新疆大学,2006
[21]杨建平,丁永建,长江黄河源区高寒植被变化的 NDVI 记录,地理学报[J],2005,60(3),467-478
[22]邓朝平,郭铌,近 20 余年来西北地区植被变化特征分析,冰川冻土[J],2006,28(2),686-693
[23]Koomanoff,V.A,1989.Analysis of Global Vegetation Patterns: A Comparison Between Remotely Sensed data and a conventional map. Biaogeography Research Series, Report#890201,Department of Geography, university of Maryland.
[24] 刘玉洁,杨忠东等,MODS 遥感信息处理原理与算法[M],科学出版社,2001,2-3
[25]李月臣,陈晋,基于 NDVI 时间序列数据的土覆盖变化检测指标设计,应用基础与工程科学学报[J],2005,13(3),261-275
[26]王坚,张继贤,基于 NDVI 序列影像精化结果的植被覆盖变化研究,测绘科学[J],2005,30(6),43-45
[27]陈多,基于 NOAA/AVHRRNDVI 的西藏拉萨地区植被季节变化,高原气象(副刊)[J],2003,22,145-151
[28]李云浩,李晓兵,基于遥感的 NDVI 与气候关系图式研究,中国图像图形学报[J],2002,7A(4),332-336
[29]齐述华,王长耀,利用 NDVI 时间序列数据分析植被长势对气候因子的响应,地理科学进展[J],2004,23(3),91-100
[30]邢旗,刘爱军,应用MODIS-NDVI对草原植被变化监测研究—以锡林郭勒盟为例,草地学报(增刊)[J],2005,13,15-19
[31]毕晓丽,王辉,植被归一化指数(NDVI)及气候因子相关起伏型时间序列变化分析,应用生态学报[J],2005,16(2),284-288
[32]李晓兵,史培军,中国典型植被类型 NDVI 动态变化与气温、降水变化的敏感性分析,植物生态学报[J],2000,24(3),379-382
[33]刘先华,陈佐忠,定居放牧方式下归一化植被指数(NDVI)的空间变化特征,植物生态学报[J],2000,24(6),662-666 [ 34 ] 罗 亚 , 徐 建 华 , 基 于 遥 感 影 像 的 植 被 指 数 研 究 方 法 述 评 , 生 态 科 学[J],2005,24(1),75-79
[35]张树誉,李登科,MODIS 植被指数及其在县域生态环境监测评估中的应用,粮食安全与生态环境监测[J],2002,4011-4019
[36]韩爱惠,用 MODIS 植被指数的时间序列分析提取荒漠化敏感区域的方法,林业资源管理[J],2004,(1),57-60
[37]宋小宁,赵英时,应用 MODIS 卫星数据提取植被-温度-水分综合指数的研究,地理与地理信息科学[J],2004,20(2),13-17
[38]曲辉,陈圣波,中分辨率成像光谱仪(MODIS)数据在地学中的应用前景,世界地质[J],2002,21(2),176-180
[39]黄家洁,万幼川,MODIS 的特性及其应用,地理空间信息[J],2003,1(4),20-28 [ 40 ] 罗 亚 , 徐 建 华 , 基 于 遥 感 影 像 的 植 被 指 数 研 究 方 法 述 评 , 生 态 科 学[J],2005,24(1),75-79
[41]徐希儒,遥感物理[M],北京大学出版社,2005,121-129
[42] http://zt.nmgnews.com.cn/zt1/dqw/nmggk/,内蒙古新闻网专题站-专题-大庆网站-内蒙古概况
[43]熊育久, 湖南省 EOS_MODIS 植被指数时空变化研究,中南林业科技大学,2006
[44] 朱文泉,潘耀忠,龙中华等, 基于 GIS 和 RS 的区域陆地植被 NPP 估算——以中国内蒙古为例,2005, (03),3-8