Analysis Of Ecosystem Sustainability In Yanhe Watershed: Use Of Spatial And Temporal Information Entropy Based On Remotely Sensed Data

Ecological sustainability is currently a hot topic that has received much attention,and it is also a difficult research area.This paper depicts ecosystem as a living body with self-organizing ability.Following Schrodinger's famous quote “Life depends on negative entropy for survival”,we have studied the interdisciplinary meaning of ecological sustainability from an entropy perspective.The focal aim of this research is to explore a set of spatial and temporal analysis methods based on information entropy.The main contents and conclusions are as follows:(1)Temporal information entropy has been proposed and its ecological meaning has been discussed.Calculating the entropy of time series NDVI data often depends on basic function models.In response to this problem,this study introduces and improves a new calculation way,and then proposes a “temporal information entropy” method.The characteristic of this method is that the estimated value of entropy can be obtained directly based on the observation data,so its application range is wider compared with other methods.Then combined with ecological resilience,the ecological meaning of temporal information entropy has been discussed.We also summarize several factors that should be paid attention when analyzing ecological resilience based on the entropy of time series NDVI data.(2)Spatial information entropy has been proposed and its ecological meaning has been discussed.Due to the problem of “dimension mismatch”,information entropy,in fact,is difficult to reflect the spatial distribution characteristics of ecological patterns.This paper introduces the concept of “proximity”,which is used to quantitatively expresses spatial attributes,and then we propose a “spatial information entropy” method.Through experimental design and algorithm comparison,on the one hand,it is verified that spatial information entropy meets the general characteristics of “entropy”,on the other hand,its sensitivity(more sensitive to changes regarding spatial information)has been illustrated.Therefore,this new method can be used to characterize the degree of disorder and heterogeneity of ecological patterns.We then explore the ecological meaning of this new method combined with the concept of ecological complexity.(3)We also attempt to bulid an analysis framkework of ecological sustainability from the entropy perspective.Given that most of the current researches discuss ecological sustainability either from ecological patterns or dynamics seperately.After the calculation problems of entropy have been solved,this study attempts to combine “pattern” and“dynamic” organically: the result of spatial information entropy regarding the ecological patterns can provide information about the macro state of an ecosystem,futher more,the result of temporal information entropy combined with land-use data,its ecological meaning can be more appropriately explained which can provide a more objective basis for ecological management.(4)Finally,the ecological sustainability of Yanhe watershed has been explored.Applying the above framework to this study area,the results show that the orderliness of ecological pattern has improved and the ecological complexity has enhanced;the dynamic pattenrs of the ecosystem are mainly orderliness and less orderliness(61%),so the overall ecological resilience has improved.Based on the research results,several reference suggestions for ecological conservation and management have been put forward.