Most years in the period from 2018 to 2022 have been exceptionally dry in Central Europe. In Germany’s forests, this long-lasting drought has caused unprecedented tree mortality. Systematic ground-based surveys, such as the annual Crown Condition Survey, provide information on the vitality status of the different tree species and their mortality rates. However, models are needed to be able to map the spatial patterns of mortality for each tree species based on cause-effect relationships derived from field observations. In this study, logistic regression models were used to identify the most important drivers of mortality for the most important tree species in Germany. For this purpose, the dead and surviving trees from the Crown Condition Survey were combined with a large set of potential predictor variables from the domains of climate, topography, soil, land cover and deposition. After feature selection, the models were evaluated using the area under the curve (AUC) statistic. Norway spruce (Picea abies; AUC = 0.9) showed by far the greatest increase in mortality, with the country-wide average observed and predicted rates approaching almost 10% per year from 2020 to 2022, and much higher predicted rates at the regional level. Much of the spruce mortality was explained by the climatic water balance of the driest summer in previous years. The other main tree species also showed clear mortality responses to the drought conditions. However, in the case of European beech (Fagus sylvatica; AUC = 0.94) and Pedunculate and Sessile oak (Quercus robur and petraea; AUC = 0.88), the peaks in the time series of the country-wide mortality rates stayed below 1%. For these broadleaved species, mortality was more dependent on a range of site conditions, i.e., soil and topography. For Scots pine (Pinus sylvestris; AUC = 0.76), for which the observed mortality rate peaked at 1.2% in 2020, the given drivers could explain mortality only to a lesser degree than for the other species. The regression models were used for spatial prediction to produce country-wide maps of species-specific mortality rates at annual temporal and 100-m spatial resolution, covering all years from 1998 to 2022. The maps visualize the spatial patterns of mortality over time. The regions in western and central Germany, which were most seriously affected by spruce dieback can clearly be identified. The models and maps presented can be used for risk assessment, forest planning, and tree species selection, providing decision support for forest practitioners.