Are species distribution models validated by field trials?

by Kyle Jensen

Invasive species, especially plant species, are one of the greatest current threats to the Earth’s biodiversity. It is feared that with the advent of global warming areas favorable to such species will increase, especially for those invasives from warmer climates that have naturalized near areas of marginal temperature. This could have negative impacts on the diversity of exposed populations, so species distribution models (SDMs) have been developed to estimate possible future distributions of organisms. These models make predictions by relating occurrence data to environmental conditions, giving a general idea of how the potential threat of an invasive species may change over time, and suggesting possible mitigation activities. Such models however have rarely been tested against experiments from the field. Sheppard et al. (2014) seek to validate SDMs through field trials at varying sites based on suitability as predicted by SDMs. If the predicted success of species in the models matches those of actual field trials, then we could be more confident in ability of models to assess the risk of invasive success. The experiment also addresses the validity of the enemy release hypothesis, which is often assumed to be the case in invasive studies. The hypothesis posits that invasive species leave behind any natural enemies when they are introduced to a new environment, which would contribute to their success. This experiment questions that assumption and its use in SDMs.

The experiment was conducted in New Zealand with three recently naturalized plant species chosen for observation: Archontophoenix cunninghamiana, Psidium guajava and Schefflera actinophylla. Each of these plants is a woody, bird-dispersed species, and thus likely invaders, the perfect subjects for SDM. Three native plants were also selected as controls to create an appropriate basis for comparison to analyze the relative success of each invasive species. The alien palm Archontophoenix cunninghamiana for example was paired with the endemic Rhopalostylis sapida as both species are part of the same subtribe and live in similar habitats. All six species were then planted at six field sites across the two islands, the locations of which were chosen based on predicted habitat suitability by SDMs. The sites were as similar as possible in regards to soil quality and slope, to eliminate variables aside from climate. Half of all plants were sprayed periodically with pesticide to eliminate any natural enemies from both native and alien species.

Over 18 months non-native species grew more than their native counterparts in the ‘suitable’ areas but had a higher mortality rate in the colder areas, though one site in particular saw more seed die-off than expected. Thus the field results were mostly in line with the predictions of the SDMs, with discrepancies due possibly to the below-average temperatures during winter or the variation of limiting resources across field sites. The models, now validated to some degree by field studies, may give us a fairly accurate estimate of suitable ranges for a given species, and thus likely changes in range due to climate change. Based on this combination of field studies and SDMs, the potentially suitable environments for these invasive species seem likely to increase as warming raises the minimum temperature. Yet one assumption of invasive studies generally may need to undergo greater scrutiny as the enemy release hypothesis did not hold in this trial, with both native and alien plants being subjected to similar levels of herbivory with or without pesticide. The current acceptance of this hypothesis may now be called into question, with further testing required.

Sheppard, C. S., Burns, B. R., & Stanley, M. C., 2014. Predicting plant invasions under climate change: are species distribution models validated by field trials?. Global change biology, 20 (9), 2800-2814.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s