COS 249-1 - Impacts of various factors on ecological network in Changsha-Zhuzhou-Xiangtan urban agglomeration, China--based on a perspective of functional performance

Abstract: Research on ecological network still suffers from a lack of species specificity in the identification of ecological sources and corridors, solidified identification ranges with errors, and neglect of stepping-stones.
The goals of this study are carrying out the evolution of ecological network components (sources, stepping-stones, corridors) of the most active urban wildlife, birds, alongside factors influencing them, from the perspective of functional performance.
Our approach to addressing our goals was constructing the level of ecological source function performance (SFP), stepping-stone function performance (SsFP), and corridor function performance (CFP) per grid with a non-fixed identification strategy to ensure that ecological network function of each grid was not overlooked. We used Moran’s Index, Geographical Detector, and regression equations to explore impacts of topography, corrected resistance, landscape indices on the ecological network components at the global and local scale.
Here we show: (1) Of the 238 grids with significant continuous changes in SFP and SsFP, Scenario16 (S₁₆: SFP continuously decreasing with SsFP continuously increasing) accounted for 33.40%, Scenario14 (S₁₄: SFP decreased and SsFP increased in the first period, and both decreased in the later period) accounted for 24.39%. More grids had a transfer in primary function performance from source to stepping-stone. The amount of change in the continuous change scenario was concentrated within the range 20%-60%. (2) Spatial correlation analysis showed that positive indicators, percentage of construction land (CLP) and nighttime lighting data (NLD), and negative indicator, NDVI, in resistance had consistently increasing negative and positive correlations with SFP, whereas the correlations with the landscape pattern factors remained stable.Topography and resistance factors contributed more to the spatial heterogeneity of SFP, whereas their contribution to SsFP demonstrated the opposite trend. In the regression relationships between SFP, SsFP and the explanatory variables, R² was higher for the curve fit and the segmented fit. (3) CFP at multiple scales was identified based on SFP, SsFP and contribution of resistance factors. Local linear regression showed significant positive and negative impacts on CFP, with landscape diversity and patchy irregularities providing more space for potential ecological corridor functions, and that regional landscape regulation strategies need to be weighted from multiple perspectives.
In conclusion, this study proposes a more flexible method for identifying ecological networks and urban ecological management strategies.