Incorporating time into the traditional correlational distributional modelling framework: A proof‐of‐concept using the Wood Thrush Hylocichla mustelina

Abstract

Detailed spatio‐temporal information about geographic distributions of species is critical for biodiversity analyses in conservation and planning. Traditional correlative modelling approaches use species observational data in model calibration and testing in a time‐averaged framework. This method averages environmental values through time to yield a single environmental value for each location. Although valuable for exploring distributions of species at a broad level, this averaging is one of myriad factors impacting model quality and reliability. We sought to optimize traditional correlative niche model performance in distributional ecology contexts by incorporating time specificity into the existing modelling framework. We modified the existing framework to account for temporal dynamics in species' distributions to produce more robust, temporally explicit models. Using the Wood Thrush Hylostichla mustelina as our study species, we introduce a method of (a) deriving a temporally explicit pseudo‐absence dataset using kernel density estimates to replicate relative sampling of sites through time, and (b) incorporating temporally explicit covariates in model calibration. By accounting for location, and month and year of primary data collection, the time‐specific models successfully yielded dynamic predictions reflecting known distributional shifts in Hylocichla mustelina's annual movement pattern. The modified data preparation steps that we present incorporate temporal dimensions into traditional correlational modelling approaches improving predictive capacity and overall utility of these models for highly mobile, short‐lived or behaviourally complex species. With the ability to estimate species' niches in greater detail, time‐specific models will be able to address specific concerns of species‐level management and policy development for highly mobile and/or migratory species, as well as disease vectors of public health interest.

Dynamic Species Distribution Models in the Marine Realm: Predicting Year-Round Habitat Suitability of Baleen Whales in the Southern Ocean

“…Relatively recent winter sightings (after 2002) are very limited, with many of the older sightings observed in winter or south of the SIE being excluded from current models due to the lack of concurrent environmental data. Although the recognition that temporal sampling bias can result in environmental bias that may affect the model reliability, limited studies have investigated methods for correcting for the effect of temporal sampling bias (Ingenloff et al, 2020). It is challenging to correct for temporal sampling bias in the current data, as the problem is not the imbalanced sightings across time (e.g., higher sampling in summer than in winter), but the almost lack of sampling efforts from mid-March to mid-December (Supplementary Figure 1), particularly from high latitudes.…”

Potential global distribution of a temperate marine coastal predator: The role of barriers and dispersal corridors on subpopulation connectivity

“…It is also possible that differences between subpopulation niches are a result of assigning annual temperature values to presences that only occurred seasonally (or monthly) across the edges of the distribution (i.e., areas of vagrancy, Ingenloff and Peterson 2021). This is particularly relevant for highly mobile species like N. cepedianus that undertake large‐scale movements and show marked migratory patterns (Barnett et al 2011; Williams et al 2012; De Wysiecki et al 2020).…”

Phenology dictates the impact of climate change on geographic distributions of six co‐occurring North American grasshoppers

“…Due to the temporal aspect of the hypotheses tested here, I used a phenological approach to generating pseudoabsences (Ingenloff & Peterson, 2021 ). Briefly, for each species, I calculated the number of observations falling within each month.…”

“…For example, ENMs can identify areas at risk of invasion under future climates (Gong et al, 2020 ; Kistner‐Thomas, 2019 ) or identify high‐priority conservation targets (Garzon et al, 2021 ), which is critically important as the ranges of many threatened species might collapse in the near future (Lemoine, 2015 ). One shortcoming is that ENMs rarely account for phenology (Ingenloff & Peterson, 2021 ); many simply use mean annual temperature or precipitation (Booth et al, 2014 ; Title & Bemmels, 2018 ). The use of models that incorporate annual trends might miss spatiotemporal shifts in distribution because habitat suitability can change within a given year (Martinez‐Meyer et al, 2004 ).…”

“…Late‐season grasshoppers, in contrast, emerge into an arid environment of 29°C and 40 mm of rainfall. Thus, accurate predictions in ENMs require that climatological data match life history data as closely as possible (Ingenloff & Peterson, 2021 ). Using mean annual temperature or precipitation might over‐ or underestimate the sensitivity of species to climate change by mischaracterizing their environmental niches.…”