Fertility loss under thermal stress

Climate change models predict that the frequency and intensity of heatwaves are likely to increase, therefore understanding population responses to these extreme climatic events will be key in mitigating biodiversity loss. Here, using the male dimorphic bulb mite, Rhizoglyphus robini, we investigate and compare the impact of experimental heat stress on survival and fertility between females and male morphs that differ in the expression of a sexually selected weapon. We show that lethal limits are similar among all individuals, but find clear sex differences, although no difference between morphs, in sub-lethal temperatures causing sterility: female fertility remains high close to lethal limits, whereas both male morphs suffer fertility loss at more than 2.5°C below their lethal limits. Contrastingly, when thermal stress was lower, the sexes and morphs do not differ in their thermal sensitivity and declines in reproductive output are comparable, these declines appear to be additive as we observe the greatest declines when both sexes were exposed to thermal stress. In addition, during assays, we included an extra treatment that allowed individuals two extra days to recover and found that fertility loss was almost fully recovered when thermal stress was low, but only partially recovered under the most extreme temperatures. The impact of heat stress on fertility therefore appears to be temporary with individuals rapidly recovering: whether recovery can negate negative population level effects remains to be investigated. Our experiments reveal the impact of heat stress on survival and fertility, finding sex-specific fertility loss under the most extreme thermal conditions.

Methods
Individual mites were isolated from a large outbred stock population. These individuals were then exposed to varying levels of thermal stress (or control temperatures). These focal individuals were then paired with individuals of the opposite sex that has not been exposed to heat stress for 2 days and allowed to mate and produce offspring. We also included an addiontal treatment that allowed heat stressed individuals an addiontal 2 days to recover from heat stress prior to pairing. We scored fertility loss in two ways: 1) thermal fertility limits was estimated by scoring focals as fertile or not by the presence/absence or any number of larvae produced and 2) thermal sensetivity of fertility was estimated by counting the number of offspring produced by pairs. Addiionally we calculated thermal limits by scoring individuals as dead or alive after heat stress.

Thermal limits and thermal fertility limits were analysed using dose response models and thermal sensetivity of fertility was analysed using hurdle models due to zero inflation in these data.