省域尺度下不同时序景观指数集与粒度效应分析
|
摘要
[目的]景观格局变化是景观生态学中的热点问题。研究省域尺度下不同时序景观指数的粒度效应、选取计算景观指数所需的适宜粒度范围并筛选出代表性景观指数集对分析景观格局变化具有重要意义。[方法]以安徽省土地覆被为研究对象,以2000年、2005年、2010年的Landsat TM遥感影像为主要数据源,采用土地利用转移矩阵对安徽省土地覆被的转移数量与方向进行研究;通过设置不同的栅格大小研究景观指数的粒度效应并选取合适的粒度范围;基于相关性分析、因子分析与敏感性分析,筛选出代表性景观指数集。[结果](1) 2000—2010年安徽省的土地覆被始终以耕地为主,其次是林地与人工表面;人工表面的增加是建立在耕地面积减少基础之上的,在10年间共有超过3 200km2的耕地转化为人工表面;(2)景观指数存在粒度效应,粒度变化会对景观指数的计算结果产生影响;省域尺度下计算景观指数所需的合适粒度范围为100～125m;(3)研究安徽省景观格局变化的代表性景观指数有4个,分别为最大斑块指数、景观形状指数、平均斑块面积、香农多样性指数。[结论]计算安徽省景观指数的适宜粒度范围为100～125m,安徽省的代表性景观指数集为最大斑块指数、景观形状指数、平均斑块面积、香农多样性指数,为进一步研究景观格局变化及驱动机制提供了依据。
The evolution of landscape pattern is one of the most hot issues in landscape ecology. It is of great significance to study the grain effect of landscape indices in different time series at provincial scale,select the appropriate grain extents for calculating landscape indices and select representative landscape indices set for analyzing landscape pattern changes. Taking the land cover in Anhui province as experimental object and the Landsat TM remote sensing images in 2000,2005 and 2010 as main data sources,this study took the method of land use transfer matrix to intuitively analyze the quantity and direction change of land cover in Anhui province. Moreover,in order to study the grain size effect of landscape indices and determine the suitable grain extent,the grain size effect curves of landscape indices were depicted by setting different grid sizes. In addition,it used the correlation analysis,factor analysis and sensitivity analysis to select the representative minimum landscape indices set. The results were showed as follows. Firstly,the dominant landscape sequence of land cover in Anhui province had not changed. It was always dominated by cultivated land,followed by forest and artificial surface from 2000 to 2010.The increase of artificial surface was based on the decrease of cultivated land. The main change of land cover type was the conversion of cultivated land into artificial surface. From 2000 to 2010,there was more than 3 200 km2 of cultivated land converted into artificial surface. Secondly,landscape indices had obvious scale dependence. Some landscape indices had obvious grain effect,and the change of grain can affect the calculation results of landscape indices,which would affect the precise analysis of landscape pattern changes. According to the first scale domain,the grain extent for calculating the landscape indices from 100 to 125 m was more suitable than the other extents at provincial scale. Finally,the representative landscape indices set reflecting landscape pattern changes from 2000 to2010 in Anhui province contained four landscape indices,which were largest patch index,landscape shape index,mean patch area and Shannon's diversity index. The grain extent from 100 to 125 m is more suitable than the other extents at Anhui provincial scale,and the representative landscape indices are largest patch index,landscape shape index,mean patch area and Shannon's diversity index. The conclusions of this study can provide a basis for further research on landscape pattern changes and driving mechanism.
The evolution of landscape pattern is one of the most hot issues in landscape ecology. It is of great significance to study the grain effect of landscape indices in different time series at provincial scale,select the appropriate grain extents for calculating landscape indices and select representative landscape indices set for analyzing landscape pattern changes. Taking the land cover in Anhui province as experimental object and the Landsat TM remote sensing images in 2000,2005 and 2010 as main data sources,this study took the method of land use transfer matrix to intuitively analyze the quantity and direction change of land cover in Anhui province. Moreover,in order to study the grain size effect of landscape indices and determine the suitable grain extent,the grain size effect curves of landscape indices were depicted by setting different grid sizes. In addition,it used the correlation analysis,factor analysis and sensitivity analysis to select the representative minimum landscape indices set. The results were showed as follows. Firstly,the dominant landscape sequence of land cover in Anhui province had not changed. It was always dominated by cultivated land,followed by forest and artificial surface from 2000 to 2010.The increase of artificial surface was based on the decrease of cultivated land. The main change of land cover type was the conversion of cultivated land into artificial surface. From 2000 to 2010,there was more than 3 200 km2 of cultivated land converted into artificial surface. Secondly,landscape indices had obvious scale dependence. Some landscape indices had obvious grain effect,and the change of grain can affect the calculation results of landscape indices,which would affect the precise analysis of landscape pattern changes. According to the first scale domain,the grain extent for calculating the landscape indices from 100 to 125 m was more suitable than the other extents at provincial scale. Finally,the representative landscape indices set reflecting landscape pattern changes from 2000 to2010 in Anhui province contained four landscape indices,which were largest patch index,landscape shape index,mean patch area and Shannon's diversity index. The grain extent from 100 to 125 m is more suitable than the other extents at Anhui provincial scale,and the representative landscape indices are largest patch index,landscape shape index,mean patch area and Shannon's diversity index. The conclusions of this study can provide a basis for further research on landscape pattern changes and driving mechanism.
引文
[1]伍星,沈珍瑶.长江上游地区土地利用/覆被和景观格局变化分析.农业工程学报,2007,23(10):86-92.
[2]井云清,张飞,陈丽华,等.艾比湖湿地土地利用/覆被—景观格局和气候变化的生态环境效应研究.环境科学学报,2017,37(9):3590-3601.
[3] Huang J,Wang R,Zhang H. Analysis of patterns and ecological security trend of modern oasis landscapes in Xinjiang,China. Environmental Monitoring&Assessment,2007,134(1-3):411.
[4] Giupponi C,Ramanzin M,Sturaro E,et al. Climate and land use changes,biodiversity and agri-environmental measures in the Belluno province,Italy. Environmental Science&Police,2006,9(2):163-173.
[5] Mwalusepo S,Muli E. land use and land cover data changes in Indian Ocean Islands:Case study of Unguja in Zanzibar Island. Data in Brief,2017,11(C):117-121.
[6]刘世梁,安南南,尹艺洁,等.广西滨海区域景观格局分析及土地利用变化预测.生态学报,2017,37(18):5915-5923.
[7]赵锐锋,陈亚宁,李卫红,等.塔里木河干流区土地覆被变化与景观格局分析.地理学报,2009,64(1):95-106.
[8]刘晓君,李占斌,李鹏,等.基于土地利用/覆被变化的流域景观格局与水沙响应关系研究.生态学报,2016,36(18):5691-5700.
[9]殷小彤.西北地区农村居民点景观格局指数的尺度效应及其最佳粒度选择分析———以西安市为例.中国农业资源与区划,2016,37(12):149-154.
[10]邬建国.景观生态学:格局、过程、尺度与等级.北京:高等教育出版社,2007.
[11]赵文武,傅伯杰,陈利顶.景观指数的粒度变化效应.第四纪研究,2003,23(3):326-333.
[12]张立平,张世文,黄元仿,等.露天煤矿区土地损毁与复垦景观指数分析.资源科学,2014,36(1):55-64.
[13]徐丽,卞晓庆,秦小林,等.空间粒度变化对合肥市景观格局指数的影响.应用生态学报,2010,21(5):1167-1173.
[14]张静雨,王天山,张景秋.空间粒度变化在时间序列上对景观指数的影响———以环洱海区域为例.地域研究与开发,2018(1):148-152.
[15]吴未,许丽萍,张敏,等.不同斑块类型的景观指数粒度效应响应———以无锡市为例.生态学报,2016,36(9):2740-2749.
[16] Saura S. Effects of remote sensor spatial resolution and data aggregation on selected fragmentation indices. Landscape Ecology,2004,19(2):197-209.
[17] Alhamad M N,Alrababah M A,Feagin R A,et al. Mediterranean drylands:The effect of grain size and domain of scale on landscape metrics.Ecological Indicators,2011,11(2):611-621.
[18]张乐,王观湧,霍习良,等.基于适宜粒度的曹妃甸新区土地利用景观格局分析.土壤,2014(6):1149-1156.
[19]蔡毅,邢岩,胡丹.敏感性分析综述.北京师范大学学报(自然科学版),2008,44(1):9-16.
[20]钱晓康.安徽统计年鉴. http://www. ahtjj. gov. cn.
[21]宁雅楠,李贝,杨伟州,等.基于主成分分析法的土地利用景观分区研究———以青龙满族自治县为例.中国农业资源与区划,2016,37(2):22-28.
[22] Couvillion B R,Fischer M R,Beck H J,et al. Spatial Configuration Trends in Coastal Louisiana from 1985 to 2010. Wetlands,2016,36(2):347-359.
[23] Lam S N,Cheng W,Zou L,et al. Effects of landscape fragmentation on land loss. Remote Sensing of Environment,2018,209:253-262.
[2]井云清,张飞,陈丽华,等.艾比湖湿地土地利用/覆被—景观格局和气候变化的生态环境效应研究.环境科学学报,2017,37(9):3590-3601.
[3] Huang J,Wang R,Zhang H. Analysis of patterns and ecological security trend of modern oasis landscapes in Xinjiang,China. Environmental Monitoring&Assessment,2007,134(1-3):411.
[4] Giupponi C,Ramanzin M,Sturaro E,et al. Climate and land use changes,biodiversity and agri-environmental measures in the Belluno province,Italy. Environmental Science&Police,2006,9(2):163-173.
[5] Mwalusepo S,Muli E. land use and land cover data changes in Indian Ocean Islands:Case study of Unguja in Zanzibar Island. Data in Brief,2017,11(C):117-121.
[6]刘世梁,安南南,尹艺洁,等.广西滨海区域景观格局分析及土地利用变化预测.生态学报,2017,37(18):5915-5923.
[7]赵锐锋,陈亚宁,李卫红,等.塔里木河干流区土地覆被变化与景观格局分析.地理学报,2009,64(1):95-106.
[8]刘晓君,李占斌,李鹏,等.基于土地利用/覆被变化的流域景观格局与水沙响应关系研究.生态学报,2016,36(18):5691-5700.
[9]殷小彤.西北地区农村居民点景观格局指数的尺度效应及其最佳粒度选择分析———以西安市为例.中国农业资源与区划,2016,37(12):149-154.
[10]邬建国.景观生态学:格局、过程、尺度与等级.北京:高等教育出版社,2007.
[11]赵文武,傅伯杰,陈利顶.景观指数的粒度变化效应.第四纪研究,2003,23(3):326-333.
[12]张立平,张世文,黄元仿,等.露天煤矿区土地损毁与复垦景观指数分析.资源科学,2014,36(1):55-64.
[13]徐丽,卞晓庆,秦小林,等.空间粒度变化对合肥市景观格局指数的影响.应用生态学报,2010,21(5):1167-1173.
[14]张静雨,王天山,张景秋.空间粒度变化在时间序列上对景观指数的影响———以环洱海区域为例.地域研究与开发,2018(1):148-152.
[15]吴未,许丽萍,张敏,等.不同斑块类型的景观指数粒度效应响应———以无锡市为例.生态学报,2016,36(9):2740-2749.
[16] Saura S. Effects of remote sensor spatial resolution and data aggregation on selected fragmentation indices. Landscape Ecology,2004,19(2):197-209.
[17] Alhamad M N,Alrababah M A,Feagin R A,et al. Mediterranean drylands:The effect of grain size and domain of scale on landscape metrics.Ecological Indicators,2011,11(2):611-621.
[18]张乐,王观湧,霍习良,等.基于适宜粒度的曹妃甸新区土地利用景观格局分析.土壤,2014(6):1149-1156.
[19]蔡毅,邢岩,胡丹.敏感性分析综述.北京师范大学学报(自然科学版),2008,44(1):9-16.
[20]钱晓康.安徽统计年鉴. http://www. ahtjj. gov. cn.
[21]宁雅楠,李贝,杨伟州,等.基于主成分分析法的土地利用景观分区研究———以青龙满族自治县为例.中国农业资源与区划,2016,37(2):22-28.
[22] Couvillion B R,Fischer M R,Beck H J,et al. Spatial Configuration Trends in Coastal Louisiana from 1985 to 2010. Wetlands,2016,36(2):347-359.
[23] Lam S N,Cheng W,Zou L,et al. Effects of landscape fragmentation on land loss. Remote Sensing of Environment,2018,209:253-262.