The role of ambient temperature in the ontogeny of endangered Oarisma poweshiek and their relative O. garita reared ex-situ at Assiniboine Park Zoo, Manitoba

Regionwide extirpations of Poweshiek skipperlings (Oarisma poweshiek) have prompted an international conservation effort to understand the causes of their decline and to recover the species. However, aspects of their basic biology remain unknown: in particular the role of temperature in their development and their sensitivity to climate change. I studied degree day (DD) accumulations in Poweshiek skipperling and its sister species, Garita skipperling (O. garita), from egg hatch to eclosure. I calculated the DDs accrued by both species reared from 2017 to 2020 at Assiniboine Park Zoo, Manitoba, and compared them between generations and species. I also compared the variability in their hatch dates, the start and end of overwintering, pupation dates, and eclosure dates for each generation and their ages (in days) at each stage. I used thermal upper and lower development thresholds of 32 ℃ and 6 ℃ using the standard and double-sine method to calculate degree-days as well as using the double-sine method without an upper threshold. I calculated similar DDs for both species, as well as similar averages within generations, hatch and eclosure dates and final ages at eclosure. However, Garita skipperlings exhibited substantially more DD accumulation variation, within and between generations, except for the pupal duration where Poweshiek skipperlings were more variable. Poweshiek and Garita skipperling DDs were more variable between generations than within generations, suggesting that variables other than temperature (such as photoperiod) may influence the synchronization of adult emergence. While Poweshiek and Garita skipperlings had similar eclosure dates, final ages, and DD accumulation across the generations, the larger variation observed in most developmental stages in Garita skipperling could suggest that they are more resilient to the effects of climate change. I also reared Garita skipperling larvae in constant, elevated temperatures and compared their growth and survivorship to larvae reared in natural, diurnal temperatures. I measured head capsules widths with an ocular micrometer and determined the total number of instars for Garita skipperling that survived to eclosure. Larvae were grouped using Dyar’s Values, without assuming Dyar’s Rule, then analyzed with k-means clustering to estimate the instar at each measurement. Garita skipperlings eclosed after five, six, or seven instars, and did not maintain a consistent size ratio between instars as assumed by Dyar’s Rule. While larvae with five, six, or seven instars were observed surviving to eclosure in both diurnal and static temperature regimes, the affect of treatment on larval instar number was indeterminable. These results cast doubt on the generalizability of Dyar’s Rule in instar determination for these species and reinforce previous authors’ conclusions that using statistical analysis without applying Dyar’s Rule may be more accurate. Experimental temperature regimes were used to investigate the affect of elevated temperatures on growth of Garita skipperling. Larvae that were reared in these trials were placed into environmental chambers held at 28 ℃, 21 ℃, or reared outdoors in the control. Two groups of larvae were added to the 28 ℃ environmental chamber: neonates added by the beginning of August, and larvae added at the end of August (late inductees, hereafter), after several weeks of development. Only neonates were placed into the 21 ℃ chamber and into the control group. I compared the survival, final ages (in days), and DD accumulation of the larvae that survived to eclosure in each treatment. The 21 ℃-trial had 15% lower and the 28 ℃ trials 6% higher (or 57% higher for late inductees) survivorship versus larvae reared outdoors. However, neonate mortality could not be calculated for late inductees to the 28 ℃-trial. Temperature appeared to influence the differences between my 28 ℃ trial and control. The results of the 21 ℃ treatment are difficult to interpret as potential equipment failure likely reduced the survival of this treatment group. The phenology of almost all larvae in the 28 ℃ chamber was extremely accelerated and almost all survivors eclosed in the year they hatched. All other larvae that survived to adulthood, regardless of treatment, eclosed at the same time as the control group. Larvae that eclosed in a single season and larvae in the control had comparable DDs. Larvae in both treatments that eclosed after overwintering had substantially higher DDs at eclosure than the control group. Photoperiod may provide cues to prepare for hibernation, and/or when to eclose, thus may be responsible for the similar eclosure dates of larvae that overwintered in all treatments. However, its role was not estimable because photoperiod was matched in the treatments and control. Poweshiek and Garita skipperlings currently have similar ontogenies in Manitoba, but the greater variability in the development of Garita skipperling seen in this study, and their longer wild flight period, suggests an increased resilience to climate change. My temperature manipulation experiments provide evidence that high static temperatures will accelerate most larvae exposed within several weeks of hatching, causing them to eclose before overwintering, and suggests that rapid climate change could be a factor in loss of Poweshiek skipperling populations as they have lower variability in their developmental rates. My results also suggest that overwintering may act to “reset” development so that individuals emerge in synchrony with conspecifics, also suggesting added resilience in Garita skipperling. Further research on the precise DD accumulation thresholds throughout development for both species is needed to determine the risk and severity of climate change induced phenological shifts in Poweshiek and Garita skipperlings.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.