中国北方沙漠化土地时空演变及其驱动力分析

发布时间：2018-07-27 21:42

【摘要】：中国北方是沙漠化土地的主要分布区,获取该地区的沙漠化土地分布及其驱动力因素对于沙漠化防治具有重要意义。沙漠化是一种土地退化现象,持续的沙漠化会导致土地贫瘠、生产力下降、生物多样性减少等诸多环境及经济问题。为了有效的大面积估算我国北方沙漠化土地的分布,本研究主要利用中分辨率成像光谱仪(Moderate Resolution Imaging Spectroradiometer,MODIS)数据的反射率产品MODIS09A1的第1波段和第2波段来计算对沙漠化土地反应敏感的荒漠化指数(Normalized Difference Desertification Index,NDDI),结合高分一号数据(GF-1)获得沙漠化土地的MODIS-NDDI时间序列图,计算研究区MODIS-NDDI时间序列图与沙漠化土地的MODIS-NDDI时间序列图之间的平均绝对距离,获得2001~2015年沙漠化土地的可能分布,对可能分布范围设定阈值得到2001~2015年沙漠化土地的分布状况,并利用交叉收敛映射(Convergent Cross Mapping,CCM)算法分析其驱动力因素。本文的主要成果如下:(1)利用多源遥感数据实现中国北方沙漠化土地的动态监测。以MODIS09A1为遥感数据源,筛选出对沙漠化土地的敏感波段(MODIS09 A1反射率数据的第1波段和第2波段)计算荒漠化指数NDDI,主要利用MODIS-NDDI时间序列数据以及高分一号(GF-1)数据来获取我国北方的沙漠化土地分布状况,总体精度达到88.38%。(2)利用多源遥感数据实现中国北方稀疏植被覆盖度的动态监测。本研究中利用MOD09A1数据和MOD15A2数据获得的归一化植被指数(Normalized Difference Desertification Index,NDVI)、光合有效辐射吸收比率(Fraction of Photosynthetically Active Radiation,FPAR)以及叶面积指数(Leaf Area Index,LAI)数据,实现中国北方2001~2015年植被覆盖度的动态监测。当1LAI3时,我们使用非稠密模型;其他情况下,我们使用稠密模型。比较TM估算结果和MODIS估算结果,结果表明二者具有高度相关性,R2=0.85,RMSE=0.50。(3)分析了中国北方近15年沙漠化土地的动态演变特征。统计近15年沙漠化土地的变化斜率可知,沙漠化土地增加的区域包括70个县,但是变化的斜率都比较小,说明沙漠化土地小幅度的加剧,沙漠化土地无变化或者减少的区域包括250个县,大部分地区的沙漠化程度都在减弱。我国北方近15年期间的沙漠化土地的重心有向高纬度低经度方向移动的趋势,15年期间沙漠化土地重心向北移动了0.06°N,向西移动了2.2°E。沙漠化土地主要分布于90~100°E经度梯度范围和30~40°N纬度梯度范围内,我国北方的沙漠化土地主要分布在海拔高度2000m以下,沙漠气候区以及草原气候区。从近15年甚至是近10年的尺度来看,我国的沙漠化状况都是有所改善的,但是近5年局部地区沙漠化程度加剧,具体表现在各大河流和湖泊附近。虽然近几年国家大力加强生态建设,总体上生态环境有所改善,但还是需要注意局部地区的环境变化。(4)分析了中国北方近15年沙漠化土地的驱动力因素。利用所获取的近15年的沙漠化土地分布及相关的降雨、温度和人口数据,采用CCM算法对中国北方沙漠化土地的驱动力因素进行分析,结果表明沙漠化土地与人口互相影响(P0.05)。同样,利用CCM算法,可知土地沙漠化受温度影响明显,但是受降雨影响不明显。另外,沙漠化土地与风速互相影响明显(P0.01)。(5)从大的尺度上看,沙漠化受自然因素影响比较大,人为活动会导致小范围的沙漠化扩张或者逆转。沙漠化土地面积如果不断扩张,会导致大量人口的迁移和社会的不稳定。中国北方是沙尘暴的巨大源头,对于该地区的沙漠化防治,预防尤为重要,我们更应该关注沙漠化潜在发生区域,政策上更应该倾向沙漠化预防。同时,我们也应该时刻关注气候变化,从结果来看,如果某些气候因子发生变化,势必导致沙漠化土地的剧烈变化,同样沙漠化土地也会对气候变化产生深远的影响。
[Abstract]:North China is the main distribution area of desertification land. The distribution of desertification land and its driving forces in this area are of great significance to the prevention and control of desertification. Desertification is a kind of land degradation phenomenon. The sustainable desertification will lead to many environmental and economic problems, such as poor land, decline in productivity, and the reduction of biodiversity. The distribution of desertification land in the north of China is estimated effectively. This study mainly uses the first and second bands of the reflectance products of Moderate Resolution Imaging Spectroradiometer (MODIS) data to calculate the desertification index (Normalized Differ) sensitive to desertification land response (Normalized Differ). Ence Desertification Index, NDDI), combined with high score 1 data (GF-1) to obtain the MODIS-NDDI time sequence diagram of desertification land, calculate the average absolute distance between the MODIS-NDDI time sequence diagram of the study area and the MODIS-NDDI time sequence diagram of the desertification land, and obtain the possible distribution of the desert land in 2015, and the possible distribution range. The thresholds were set to obtain the distribution of desertification land in 2001~2015 years, and the driving force factors were analyzed using the cross convergent mapping (Convergent Cross Mapping, CCM) algorithm. The main achievements of this paper are as follows: (1) the dynamic monitoring of desertification land in northern China is realized by using multi-source remote sensing data. MODIS09A1 is used as the remote sensing data source to screen out the sand. Desertification index NDDI is calculated on the sensitive band of desertification land (first band and second band of MODIS09 A1 reflectance data). The distribution of desertification land in northern China is obtained by using MODIS-NDDI time series data and high score 1 (GF-1) data. The overall precision reaches 88.38%. (2) using multi source remote sensing data to realize the north of China. Dynamic monitoring of sparse vegetation coverage. The normalized vegetation index (Normalized Difference Desertification Index, NDVI) obtained from the MOD09A1 data and the MOD15A2 data in this study, the photosynthetic effective radiation absorption ratio (Fraction of Photosynthetically Active Radiation, and the leaf area index) data, Dynamic monitoring of vegetation coverage in northern China during 2001~2015. When 1LAI3, we use a non dense model. In other cases, we use a dense model. Compare TM estimation results and MODIS estimation results. The results show that the two are highly correlated, R2=0.85, RMSE=0.50. (3) analysis of desertification land dynamics in northern China for nearly 15 years The change slope of desertification land in the last 15 years shows that the areas of desertification land increase include 70 counties, but the slope of the change is relatively small, which indicates that the desertification land has increased slightly, the desertification land is not changed or reduced in 250 counties, and the desertification degree of most areas is weakening. In the past 15 years, the center of desertification land has a tendency toward high latitude and low longitude direction. During the period of 15 years, the barycenter of desertification land moved northward to 0.06 N, and the desertification land moved westward 2.2 E. mainly in the range of the longitude gradient of 90~100 E and the latitude gradient of 30~40 N, and the desertification land in the north of China was mainly distributed. At the altitude of less than 2000m, the desert climate zone and the grassland climate area. From the last 15 years or even 10 years, the desertification situation in China has been improved, but the desertification degree of the local areas has intensified in the last 5 years, which is concretely represented in the large rivers and lakes. Although the state has strengthened the ecological construction in recent years, the overall situation has been greatly improved. The ecological environment has improved, but there is still a need to pay attention to the environmental changes in local areas. (4) the driving forces of desertification land in northern China for the last 15 years are analyzed. Using the desertification land distribution and related rainfall, temperature and population data obtained in the last 15 years, the driving force factors of desertification land in northern China are calculated by CCM method. The results show that desertification land and population interact with each other (P0.05). Similarly, using the CCM algorithm, it can be found that the desertification is affected by the temperature obviously, but the influence of rainfall is not obvious. In addition, the desertification land and the wind speed influence each other obviously (P0.01). (5) from the large scale, the desertification is greatly influenced by the natural factors and human activities. It will lead to a small area of desertification expansion or reversal. If the desertification land area is expanding, it will lead to a large number of population migration and social instability. Northern China is a huge source of sand dust storm, especially for the prevention and control of desertification in this area. We should pay more attention to the potential desertification area of desertification, and should be more appropriate for policy. It is a tendency to prevent desertification. At the same time, we should also pay attention to climate change at the same time. From the results, if some climate factors change, it is bound to lead to severe changes in desertification land, and the same desertification land will have a far-reaching impact on climate change.
【学位授予单位】：中国林业科学研究院
【学位级别】：博士
【学位授予年份】：2017
【分类号】：S288

【相似文献】