The effects of climate data precision on fitting and projecting species niche models

By Darren J. Kriticos, Agathe Leriche. 2010

89 citations recorded by the source corpus.

Biosecurity agencies are particularly concerned to know the potential distribution of invasive alien species under present, and to a lesser extent, future climates; expensive decisions can hinge upon the degree of perceived threat a pest species poses. Climate-based niche modelling techniques are available to inform these decisions. These tools now regularly employ gridded climate datasets of moderate spatial resolution (0.5 degree), though biosecurity decision-makers continually seek greater spatial precision in the risk map products. Various splining techniques are capable of generating gridded climate datasets approaching the precision limits imposed by the availability of digital elevation model data. As the spatial precision of climate datasets increases, more detailed effects of topographic relief become apparent in the climatic data. When these datasets are used to develop and apply species niche models, the climate data is spatially intersected with species location data to infer relationships between the climate and the species' geographic distribution. Here we investigate the effect of changing climate precision on projections of species' niche models developed with CLIMEX, including the effect of upscaling and downscaling the outputs. We found that there were noticeable increases in sensitivity in models developed using more precise climate datasets. The largest differences in projections were noted where species range limits coincided with regions of strong climatic gradients such as where there was marked topographic relief in relation to the spatial precision of the climatic dataset. Upscaling (fitting a model with a fine resolution dataset and then projecting the results with a coarser grid), tended to produce smaller potential ranges for a species, albeit at the cost of model sensitivity. Downscaling had the opposite effect, identifying additional, mostly marginally climatically suitable habitat. It remains unclear how sensitive the fine resolution results are to the number and spatial arrangement of input location records used to build the model. The results indicate some benefits of improving the spatial resolution of climate datasets, though not at the expense of climatic data accuracy. Decision-makers should be mindful of the inherent uncertainties in these models, and modellers have a responsibility to identify and convey these uncertainties to their intended audience.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/23cc39778e6204e009d3fff59e0fe1011b3bc591

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecological Modeling

In Ecology

In Nature and Landscape Conservation

Citations in the corpus

Dispersal ability rather than ecological tolerance drives differences in range size between lentic and lotic water beetles (Coleoptera: Hydrophilidae) (2012, background): “In order to understand species distributions, it is also necessary to consider the effects of scale (Hui et al., 2010; Kriticos & Leriche, 2010).”
Evaluating correlative and mechanistic niche models for assessing the risk of pest establishment (2014, methodology): “The CLIMEX model is considered a process-based mechanistic niche model (Kriticos and Leriche 2010); however, it can also be a correlative model if species-specific phenological thresholds are not used in defining model parameters, but instead parameters are estimated using species’ known…”
CliMond: global high‐resolution historical and future scenario climate surfaces for bioclimatic modelling (2012, background): “Methods in Ecology and Evolution 2011 British Ecological Society, Methods in Ecology and Evolution, 3, 53–64 data can affect the model parameterization, as can the source and treatment of the foundation climatological data (Kriticos & Leriche 2010).”
Predicting distributions of invasive species (2013, methodology): “ Authors refer to this as a process-‐oriented or mechanistic model (e.g. Kriticos & Leriche, 2010) because (1) the model components consider environmental impacts on the species in a growth and stress framework, similar to process-‐based population models;…”
Review of CLIMEX and MaxEnt for studying species distribution in South Korea (2018, background): “By using “Compare Locations,” researchers can predict the potential geographical distribution of target species with regard to climatic conditions (Jung et al 2016; Kriticos and Leriche 2010; Kriticos et al 2015; Stephens et al 2007).”
The uncertain nature of absences and their importance in species distribution modelling (2010, background): “Here we assume that is more likely that these absences correspond to lack of knowledge than to the actual absence of populations, although this is also scale-dependent (Hortal 2008, Kriticos and Leriche 2010).”
Understanding (insect) species distributions across spatial scales (2010, background): “In order to describe species distributions, and understand the effect of different factors that shape them, it is necessary to explicitly consider the effect of scale (Hui et al. 2010, Kriticos and Leriche 2010).”