Avoiding verisimilitude when modelling ecological responses to climate change: the influence of weather conditions on trapping efficiency in European badgers (Meles meles)

Abstract

The signal for climate change effects can be abstruse; consequently, interpretations of evidence must avoid verisimilitude, or else misattribution of causality could compromise policy decisions. Examining climatic effects on wild animal population dynamics requires ability to trap, observe or photograph and to recapture study individuals consistently. In this regard, we use 19 years of data (1994-2012), detailing the life histories on 1179 individual European badgers over 3288 (re-) trapping events, to test whether trapping efficiency was associated with season, weather variables (both contemporaneous and time lagged), body-condition index (BCI) and trapping efficiency (TE). PCA factor loadings demonstrated that TE was affected significantly by temperature and precipitation, as well as time lags in these variables. From multi-model inference, BCI was the principal driver of TE, where badgers in good condition were less likely to be trapped. Our analyses exposed that this was enacted mechanistically via weather variables driving BCI, affecting TE. Notably, the very conditions that militated for poor trapping success have been associated with actual survival and population abundance benefits in badgers. Using these findings to parameterize simulations, projecting best-/worst-case scenario weather conditions and BCI resulted in 8.6% ± 4.9 SD difference in seasonal TE, leading to a potential 55.0% population abundance under-estimation under the worst-case scenario; 38.6% over-estimation under the best case. Interestingly, simulations revealed that while any single trapping session might prove misrepresentative of the true population abundance, due to weather effects, prolonging capture-mark-recapture studies under sub-optimal conditions decreased the accuracy of population estimates significantly. We also use these projection scenarios to explore how weather could impact government-led trapping of badgers in the UK, in relation to TB management. We conclude that population monitoring must be calibrated against the likelihood that weather conditions could be altering trap success directly, and therefore biasing model design.

Climate and anthropogenic factors determine site occupancy in Scotland's Northern‐range badger population: implications of context‐dependent responses under environmental change

“…Noonan et al. () caution that if weather conditions affect badger activity, and thus detectability, this can obscure actual population demographic responses to climate change, which we addressed here using hierarchical occupancy modelling (as recommended by Noonan et al., ).…”

“…Badgers are strongly influenced by temperature and rainfall patterns (Byrne et al, 2015;Macdonald & Newman, 2002;Macdonald et al, 2010;Nouvellet et al, 2013); therefore, variation in their aboveground activity regimes (Noonan et al, 2014) provides the most likely source of variation in detection rates. Noonan et al (2015)…”

“…This is essential when weather conditions influence activity rates and thus the ability to survey the species in question. Noonan, Rahman, Newman, Buesching, and Macdonald (2015) established that badgers tend to stay below ground in their burrows (termed setts) during poor weather (Noonan et al, 2014). We addressed this potential problem using hierarchical occupancy modelling (MacKenzie et al, 2006).…”

“…Within this framework, we used Autumn-Winter occupancy data, that is, covering a critical phase of the badger life cycle, when food availability and environmental conditions are likely to be limiting (Noonan et al, 2014;Nouvellet et al, 2013), to test how badger occupancy is influenced by: (1) climatic conditions (and associated elevation effects), which define the species' climatic niche and thus directly affect its survival and recruitment patterns (Byrne et al, 2015;Macdonald et al, 2010;Noonan et al, 2015;Nouvellet et al, 2013;Thornton, 1988); (2) development of human infrastructures, as measures of anthropogenic disturbance (Elliott, O'Brien, & Hayden, 2015;Grilo, Bissonette, & Santos-Reis, 2009;Hipólito, Santos-Reis, & Rosalino, 2016); (3) forested versus un-forested habitats, because these relate to food availability and suitable sett sites (Virgós, 2001(Virgós, , 2002; and (4) structural complexity of the landscape matrix, as a surrogate for resource richness and diversity (Santos, Rosalino, Santos-Reis, & Ustin, 2016;Virgós, 2001). We then investigated how distinct topo-climatic contexts (i.e., temperature and elevation) interact with habitat type, human infrastructural development and modification of the landscape for agriculture.…”

Seasonal variation in daily patterns of social contacts in the European badger Meles meles

“…Given the well‐documented reduction in badger activity during winter (Lindsay & Macdonald, ; Woodroffe & Macdonald, ; Noonan, Rahman, Newman, Buesching, & Macdonald, ), daily patterns of contact duration in the current study changed as expected from summer through to winter. During the summer months, there was also a substantial peak in contact duration during the daytime, while in winter contact duration remained similar throughout the daily cycle.…”