Incorporating spatial non-stationarity of regression coefficients into predictive vegetation models

Kupfer, John A.; Farris, Calvin A.

doi:10.1007/s10980-006-9058-2

Cited by 103 publications

(65 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A consistent result of GWR comparison studies is the increased GWR local R 2 compared to OLS global R 2 [15,16,18,45]. For example, Foody [15] found that over 90% of the variation between sub-Saharan African bird species richness and a set of environmental variables (climate and NDVI) was explained at fine scales (1°), and that the explained variance decreased as the bandwidth used in the GWR increased (to 8°).…”

Section: Discussionmentioning

confidence: 83%

Exploring Spatial Scale, Autocorrelation and Nonstationarity of Bird Species Richness Patterns

Holloway

Miller

2015

IJGI

View full text Add to dashboard Cite

Abstract:In this paper we explore relationships between bird species richness and environmental factors in New York State, focusing particularly on how spatial scale, autocorrelation and nonstationarity affect these relationships. We used spatial statistics, Getis-Ord Gi*(d), to investigate how spatial scale affects the measurement of richness "hot-spots" and "cold-spots" (clusters of high and low species richness, respectively) and geographically weighted regression (GWR) to explore scale dependencies and nonstationarity in the relationships between richness and environmental variables such as climate and plant productivity. Finally, we introduce a geovisualization approach to show how these relationships are affected by spatial scale in order to understand the complex spatial patterns of species richness.

show abstract

Section: Discussionmentioning

confidence: 83%

Exploring Spatial Scale, Autocorrelation and Nonstationarity of Bird Species Richness Patterns

Holloway

Miller

2015

IJGI

View full text Add to dashboard Cite

show abstract

“…GWR can better simulate the spatial non-stationary relationship between vegetation activity and climatic change, which is difficult to predict due to the combination of different factors (Kupfer and Farris, 2007). Compared with the approach 20 based on traditional correlation analysis, GWR has a better ability to deal with regional difference in responding mechanism or processes, which is essential for making policy to suit adaptation measures to local situation.…”

Section: Spatial Heterogeneitymentioning

confidence: 99%

Past and future influence of climate change on spatially heterogeneous vegetation activity in China

Gao

Jiao

et al. 2017

Preprint

View full text Add to dashboard Cite

Abstract. Climate change is a major driver of vegetation activity, and thus their complex relationships become a frontier and 10 difficulty in global change research. In this paper, the spatial distribution and dynamic variation of climate change impacts on vegetation activity from 1980s to 2050 in China were investigated by the Geographically Weighted Regression (GWR) model, based on the combined datasets of satellite vegetation index, climate observation and projection, and future vegetation productivity simulation. Our results noted that the significant positive precipitation-vegetation relationship was and will be almost distributed in the north of China, except the northeast and northwest China with significant but varying 15 influence of temperature rising, while the regions with temperature dominated vegetation activity mainly located in the southern part of China. There will be different climatic dominators for vegetation activity in some regions such as northwest China, and even opposite correlation in the northeast China, and further the responding patterns of vegetation activity to precipitation variation will be moving southward in the future three decades. It is indicated that although high warming rate will restrain the vegetation activity, precipitation variability can mediate the hydrothermal conditions for vegetation activity, 20 for example the enhanced vegetation activity in the Tibetan Plateau and the weakened vegetation activity in the East and Middle China in the future. Furthermore, coupling the responding patterns and the dynamic variation, it can be found that during the period from 2021 to 2050, vegetation in most of north China may adapt to an arid environment, while in many southern parts it will be repressed due to water shortage. However, the continuous and dynamic responding process of vegetation activity to climate change will be determined by the spatial heterogeneity in climate change and vegetation cover. 25

show abstract

“…The use of GWR as a predictor has attracted much attention, where it has been empirically and favourably compared to: (i) alternative regressions (e.g. Zhang et al 2005;Gao et al 2006;Bitter et al 2007;Kupfer and Farris 2007) or (ii) kriging (Páez et al 2008). In contrast, GWR has been empirically and unfavourably compared to: (a) alternative regressions (Wheeler and Waller 2009;Salas et al 2010) or (b) kriging (Lloyd 2010).…”

Section: Introductionmentioning

confidence: 99%

Links, comparisons and extensions of the geographically weighted regression model when used as a spatial predictor

Harris

Brunsdon

Fotheringham

2010

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

In this study, we link and compare the geographically weighted regression (GWR) model with the kriging with an external drift (KED) model of geostatistics. This includes empirical work where models are performance tested with respect to prediction and prediction uncertainty accuracy. In basic forms, GWR and KED (specified with local neighbourhoods) both cater for nonstationary correlations (i.e. the process is heteroskedastic with respect to relationships between the variable of interest and its covariates) and as such, can predict more accurately than models that do not. Furthermore, on specification of an additional heteroskedastic term to the same models (now with respect to a process variance), locallyaccurate measures of prediction uncertainty can result. These heteroskedastic extensions of GWR and KED can be preferred to basic constructions, whose measures of prediction uncertainty are only ever likely to be globallyaccurate. We evaluate both basic and heteroskedastic GWR and KED models using a case study data set, where data relationships are known to vary across space. Here GWR performs well with respect to the more involved KED model and as such, GWR is considered a viable alternative to the more established model in this particular comparison. Our study adds to a growing body of empirical evidence that GWR can be a worthy predictor; complementing its more usual guise as an exploratory technique for investigating relationships in multivariate spatial data sets.

show abstract

Incorporating spatial non-stationarity of regression coefficients into predictive vegetation models

Cited by 103 publications

References 47 publications

Exploring Spatial Scale, Autocorrelation and Nonstationarity of Bird Species Richness Patterns

Exploring Spatial Scale, Autocorrelation and Nonstationarity of Bird Species Richness Patterns

Past and future influence of climate change on spatially heterogeneous vegetation activity in China

Links, comparisons and extensions of the geographically weighted regression model when used as a spatial predictor

Contact Info

Product

Resources

About