SUPER models: an application to Monarch butterfly conservation

Why are you here? More specifically, what made you decide to live where you are? And what would make you leave? These simple questions dictate how animals (both human and non-human) are spread throughout the landscape. It is questions like these that ecologists can use to assess conservation priorities. 

FIGURE 1. An example of a habitat suitability model (HSM).
Figure credit: Yannic et al. 2014

Biodiversity conservation and management relies on information about species’ geographic distributions as well as the environmental changes that pose the most risk to them. Species distribution (SDM) and habitat suitability (HSM) modeling are useful tools for projecting and mapping species-specific ranges as well as identifying important variables to explain their occurrence patterns [Figure 1]. Thus, these models have important applications to conservation biology and restoration efforts. One such application relates to ongoing efforts to monitor and conserve migratory insect pollinators. Policy makers and land managers need to be equipped with knowledge of the potential responses of these organisms to factors like climate change and increased human activity (urbanization, pesticide & herbicide use, etc.). Unfortunately, much of this information remains largely unknown. If the distributions of some pollinators can be modeled successfully, then citizens, decision makers, and scientists can all make more informed evaluations of conservation work.

...Similar to what’s being done for monarchs. The conservation of monarch butterflies (Danaus plexippus) has received a lot of attention in recent years. Monarch populations have significantly declined in the past few decades and there is tremendous concern about the recovery potential of this species; the need for effective responses is becoming more and more imperative. Conservationists have welcomed proactive solutions like native plantings and decreased chemical use, but how do we find out where to best implement these solutions? Oberhauser and Peterson’s 2003 study with SDMs have helped to answer this. They suggested we first ask: How will future populations of D. plexippus respond to climatic changes? 

Oberhauser and Peterson (2003) explored the effects of climate change on eastern monarch distributions in Mexico to communicate this population’s degree of vulnerability to extinction. Using occurrence data, they analyzed overwintering locations of monarchs according to variables such as topography, cloud cover, precipitation, temperature, and the presence of their winter homes: oyamel fir (Abies religiosa) trees. From these inputs, they were able to model the ecological niche of the species and project colony distributions according to (1) oyamel distributions and (2) future climate models. Their generated SDMs for both the monarchs and firs were highly predictive of the actual occurrence sites, demonstrating significant power in the reliability and accuracy of this technique. The main patterns that emerged in the modeled distributions included: a high association of monarch presence with oyamel fir distributions, indicating that both species possess similar ecological requirements [Figure 2]; temperature and precipitation are the critical factors that influence monarch overwinter survivorship; and thus, climatic changes (particularly, an increase in rainfall) render overwintering sites unsuitable, causing a northward shift of monarch distributions. The findings present some interesting points in the case of monarchs and the “spatial manifestation of their ecological niches” (Oberhauser and Peterson, 2003).

FIGURE 2. Ecological niche model generated from occurrence data on oyamel fir and monarchs. Darker shades of gray indicate greater model agreement. Upper panel: View of central Mexico model projection. Lower panel: Close-up of overwintering area, with best locations overlaid in black/ darkest gray. (Refer to Oberhauser & Peterson 2003 for more information on the figure.)

While climate change was predicted to have a negative impact on habitable areas for monarchs, to what extent are the actual environmental conditions driving this effect? As demonstrated in the study, it is easier to model species distributions as functions of environmental variables, but equally important are data on resource effects (i.e., food availability, nesting sites) on species. Especially in the context of habitat specialists like D. plexippus, their distributions may be determined more by host plant availability rather than environmental effects. And this is hinted at within the ’03 study: Oberhauser and Peterson modeled great overlap between oyamel fir occurrences and monarch occurrences. Their efforts to model climate change effects on monarch winter distributions inspired me to model climate change effects on monarch host plant distributions. Specifically, I plan to incorporate some of the modeling techniques used in their study to assess the ecological niche of monarch summer host plants, Asclepias syriaca (Common milkweed). And I hope to forecast their distributions to create an HSM for Asclepias, and ultimately, monarchs. I intend to limit my study area to the state of Iowa, so my research may assist in informing local conservation measures and milkweed planting projects. But I am also optimistic about its broader implications for conservation planning and addressing land-use/ farming practices. In order to save the monarchs, it’s time we recognize “model behavior."

REFERENCE:

Oberhauser, K., & Peterson, A. T. (2003). Modeling current and future potential wintering distributions

of eastern North American monarch butterflies. Proceedings of the National Academy of Sciences. U.S.A., 100: 14063-14068.