In many arid ecosystems, vegetation frequently occurs in high-cover patches interspersed in a matrix of low plant cover. However, theoretical explanations for shrub patch pattern dynamics along climate gradients remain unclear on a large scale. This context aimed to assess the variance of the Reaumuria soongorica patch structure along the precipitation gradient and the factors that affect patch structure formation in the middle and lower Heihe River Basin (HRB). Field investigations on vegetation patterns and heterogeneity in soil properties were conducted during 2014 and 2015. The results showed that patch height, size and plant-to-patch distance were smaller in high precipitation habitats than in low precipitation sites. Climate, soil and vegetation explained 82.5% of the variance in patch structure. Spatially, R. soongorica shifted from a clumped to a random pattern on the landscape towards the MAP gradient, and heterogeneity in the surface soil properties (the ratio of biological soil crust (BSC) to bare gravels (BG)) determined the R. soongorica population distribution pattern in the middle and lower HRB. A conceptual model, which integrated water availability and plant facilitation and competition effects, was revealed that R. soongorica changed from a flexible water use strategy in high precipitation regions to a consistent water use strategy in low precipitation areas. Our study provides a comprehensive quantification of the variance in shrub patch structure along a precipitation gradient and may improve our understanding of vegetation pattern dynamics in the Gobi Desert under future climate change.
Yanchi County is located in the agro-pastoral ecotone and belongs to the ecologically fragile area of Northwest China.It is important to study the evolution of landscape pattern to curb its environmental degradation.In order to intuitively show how the landscape pattern of the study area changes over time,Landsat Thematic Mappers(TM)and Landsat Operational Land Imager(OLI)data of 1991,2000,2010 and 2017 were used.This paper attempts to apply niche theories and methods into landscape ecology,and constructs a niche model of landscape components by using"n-dimentional hypervolume niche theory"and landscape pattern indices.By evaluating the spatial and temporal evolution of niche from the perspective of two-dimensional space to reflect the changes of landscape pattern in the study area over the past 26 years,new theories and methods were introduced for the characterization of landscape pattern.The results indicate that:1)The larger the attribute and dominance value of landscape components,the higher the ecological niche and the stronger the control effect on the overall landscape.2)The ecological niche of each landscape component was significantly different,just as its control effect on the overall landscape.3)The dynamic change of the ecological niche of each landscape component was different,with grassland,unused land and arable land always in a high dominant position,although the ecological niche of construction land and water area was always low.In general,the introduction of niche theory into the landscape ecology provided a new method to study the changes in regional landscape pattern. 相似文献
ABSTRACT High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning. 相似文献