Mapping the spatial and temporal dynamics of species distributions is necessary for biodiversity conservation land-use planning decisions. Recent advances in remote sensing and machine learning have allowed for high resolution species distribution modelling that can inform landscape-level decision making. Here we compare the performance of three popular Sentinel-2 (10m) land cover maps including Dynamic World (DW), European land cover (ELC10) and World Cover (WC), in predicting wild bee species richness over southern Norway. The proportion of grassland habitat within 250m (derived from the land cover maps), along with temperature and distance to sandy soils, were used as predictors in both Bayesian Regularized Neural Network and Random Forest models. Models using grassland habitat from DW performed best (RMSE = 2.85; averaged across models), followed by WC (RMSE = 2.86) and ELC10 (RMSE = 2.89). All satellite-derived maps outperformed a manually mapped Norwegian land cover dataset called AR5 (RMSE = 3.02). When validating the model predictions of bee species richness against citizen science data on solitary bee occurrences using generalized linear models, we found that ELC10 performed best (AIC = 2800), followed by WC (AIC = 2939), and DW (AIC = 2973). While the differences in RMSE we observed between models were small, they may be significant when such models are used to prioritize grassland patches within a landscape for conservation subsidies or management policies. Partial dependencies in our models showed that increasing the proportion of grassland habitat is positively associated with wild bee species richness, thereby justifying bee conservation schemes that aim to enhance semi-natural grassland habitat. Our results confirm the utility of satellite-derived land cover maps in supporting high resolution species distribution modelling and suggest there is scope to monitor changes in species distributions over time given the dense time series provided by products like DW.

Radiative skin temperature is often used to examine heat exposure in multi-city studies and for informing urban heat management efforts since urban air temperature is rarely measured at the appropriate scales. Cities also have lower relative humidity, which is not traditionally accounted for in large-scale observational urban heat risk assessments. Here using crowdsourced measurements from over 40,000 weather stations in ≈600 urban clusters in Europe, we show the moderating effect of this urbanization-induced humidity reduction on heat stress during the 2019 heatwave. We demonstrate that daytime differences in heat index between urban clusters and their surroundings are weak and associations of this urban-rural difference with background climate, generally examined from the skin temperature perspective, is diminished due to moisture feedback. We also examine the spatial variability of skin temperature, air temperature, and heat indices within these clusters, relevant for detecting hotspots and potential disparities in heat exposure, and find that skin temperature is a poor proxy for the intra-urban distribution of heat stress. Finally, urban vegetation shows much weaker (~1/6th as strong) associations with heat stress than with skin temperature, which has broad implications for optimizing urban heat mitigation strategies. Our results are valid for both operational metrics of heat stress (such as apparent temperature and Humidex) and for various empirical heat indices from epidemiological studies. This study provide large-scale empirical evidence that skin temperature, used due to the lack of better alternatives, is weakly suitable for informing heat mitigation strategies within and across cities, necessitating more urban meteorological observations.