基于DInSAR的洪河国家级自然保护区沼泽水位相对变化量监测研究

水文是沼泽形成、发育和演化的决定性因素。目前,沼泽水位的监测仍然以实地观测为主,需要花费大量的人力和物力。以洪河国家级自然保护区为研究区,利用2007年6~10月星载相控阵L波段合成孔径雷达(phase array type L-band synthetic aperture radar,L-band PALSAR)数据和2015年6~11月C波段Sentinel-1A合成孔径雷达(synthetic aperture radar,SAR)数据,采用合成孔径雷达差分干涉测量(differential interferometric synthetic aperture radar,DInSAR)技术,提取出研究区沼泽的地表形变相位,计算出研究区沼泽水位相对变化量,构建了研究区沼泽水位相对变化量遥感监测经验估算公式,实现了对研究区沼泽年内水位变化量的遥感监测,研究了不同时相浅水沼泽区和深水沼泽区干涉相干性的差异;采用方差分析和回归分析方法,在水位观测站、浅水沼泽区和深水沼泽区3种尺度上,分别对利用两种数据计算出的沼泽水位相对变化量,进行了精度验证和显著性检验。研究结果表明,植物生长期和植物空间格局都对沼泽干涉相干性产生影响,8~10月研究区干涉相干性要好于6~7月,浅水沼泽区干涉相干性要高于深水沼泽区,前者的干涉相干系数比后者的大10%~35%;计算出的沼泽水位相对变化量与地面实测的水位变化量的变化趋势吻合度较高,2007年6~7月和9~10月浅水沼泽区的水位波动幅度较大,水位变化量超过了0.70 m。

Monitoring of Relative Variation of Water Level of Marshes in Honghe National Nature Reserve Using DInSAR Technique

Hydrological situation determines the formation,development and evolution of the wetlands.At present,field observation still has been a main approach for monitoring water level change of the wetlands,which usually need to spend a lot of manpower and material resources.In this paper,the technique of differential interferometry synthetic aperture radar(DInSAR)was utilized to extract the phase of surface deformation of the marshes in Honghe National Nature Reserve using L-band PALSAR images from June to October in 2007,C-band Sentinel-1A SAR images from June to October in 2015.The relative water level variation was calculated in this study,the empirical estimation formulas of water level variation were provided based on remote sensing technique,and the annual water-level variation was monitored in the study area.This paper explored the differences of interferometric coherence of shallow water marshes and deep water marshes during different periods.Variance and regression analysis were used to implement accuracy assessment and significance test of water level variation using empirical estimation formulas and field measurement from three scales of hydrological observation stations,shallow water marshes and deep water marshes,respectively.The results showed that growth phases and cluster spatial patterns of vegetation had an effect on the interferometric coherence of the marshes.The interferometric coherence of study area from August to October was better than that from June to July.The interferometric coherence of shallow water marshes was higher than that of deep water marshes,and the interferometric coherence coefficients of shallow water marshes were 10%-35%more than those of deep water marshes.The water level variation of marshes obtained by DInSAR techniquein agreement with field measurement.The larger ones of water level variation in the shallow water marshes were from June to July and from September to October,the differennce between the mmaximum and the minimum of water level was over 0.7 m.