Climate change and alpine-adapted insects: modelling environmental envelopes of a grasshopper radiation

Abstract

Mountains create steep environmental gradients that are sensitive barometers of climate change. We calibrated 10 statistical models to formulate ensemble ecological niche models for 12 predominantly alpine, flightless grasshopper species in Aotearoa New Zealand, using their current distributions and current conditions. Niche models were then projected for two future global climate scenarios: representative concentration pathway (RCP) 2.6 (1.0°C rise) and RCP8.5 (3.7°C rise). Results were species specific, with two-thirds of our models suggesting a reduction in potential range for nine species by 2070, but surprisingly, for six species, we predict an increase in potential suitable habitat under mild (+1.0°C) or severe global warming (+3.7°C). However, when the limited dispersal ability of these flightless grasshoppers is taken into account, all 12 species studied are predicted to suffer extreme reductions in range, with a quarter likely to go extinct due to a 96-100% reduction in suitable habitat. Habitat loss is associated with habitat fragmentation that is likely to escalate stochastic vulnerability of remaining populations. Here, we present the predicted outcomes for an endemic radiation of alpine taxa as an exemplar of the challenges that alpine species, both in New Zealand and internationally, are subject to by anthropogenic climate change.

Keywords: FRAGSTATS; alpine; biomod2; climate change; ecological niche modelling; ensemble modelling; fragmentation.

Figure 1.

Relief map of Aotearoa New Zealand with locations referred to in the text (left). Distribution maps (right) show the locations of presence and absence data points input into ecological niche models for each of 12 New Zealand grasshopper species (Acrididae; Catantopinae).

Figure 2.

Environmental predictor variables differ in their importance in models of presence for 12 species of New Zealand alpine grasshoppers in the genera Alpinacris, Brachaspis, Paprides and Sigaus. Relative scores (percentages) are from the EMwm. See electronic supplementary material, S8 for results of other models.

Figure 3.

The realized and predicted niche space of 12 New Zealand alpine grasshopper species generated by ENM (EMwms). Maps show ecological niche models for current climate and as predicted for two future climate change scenarios (RCP2.6 (1°C rise) and RCP8.5(3.7°C rise) trajectories). Colours indicate probability (high to low) of suitable habitat under the model.

Figure 4.

Predicted loss of suitable habitat (as a percentage of current range) for 12 grasshopper species in the New Zealand alpine genera Alpinacris, Brachaspis, Paprides and Sigaus under two future climate scenarios: approximately 1.0°C rise by 2070(RCP2.6) and approximately 3.7°C rise by 2070(RCP8.5). Percentages are based on the loss/stability of pixels between vector binary maps (current models compared with future); (a) is the predicted loss/gain of all potential habitat patches and (b) is the predicted loss of space excluding fragments that would require dispersal by these flightless grasshoppers to newly available pixels (habitat).

Figure 5.

Predicted habitat fragmentation for the endemic, flightless, alpine grasshopper Sigaus australis on Ka Tiritiri-o-te-Moana, Aotearoa (Southern Alps, New Zealand). Density distributions of habitat patches greater than 0.1 km² in current conditions and two future scenarios under anthropogenic climate change (RCP2.6 and RCP8.5). Vertical dashed lines indicate means of each distribution with their values. Density plots are scaled to sum total of habitat greater than 0.1 km² under each scenario (see electronic supplementary material, S12 for plots of other species).