Warmer, drier summers brought by climate change increase the risk of frequent wildfires on the moorland of the Peak District National Park (PDNP) of northern England. Fires are costly to fight, damage the ecosystem, harm water catchments, cause erosion scars and disrupt transport. Fires release carbon dioxide to the atmosphere. Accurate forecasts of the timing of fires and high fire risk locations will aid the deployment of fire fighting resources.Both spatial modelling (identifying where risk of fire is highest, based on past fires) and temporal analysis (predicting when that risk is likely to be highest, based on preceding weather) were applied in this analysis. Firstly, multi-criteria evaluation (MCE) was used to spatially model the risk of reported wildfires in the Dark Peak area (northern part of the PDNP)), based on a 28-year record of fires from the PDNP rangers’ fire log. Fire risk was investigated using habitat and aspect maps to represent vulnerability to ignition, and distance from access features as a proxy for the likelihood of ignition sources. This showed that bare peat, eroding moorland and bilberry bog were the habitats with the most reported fires. Moorland restoration measures to revegetate bare peat and raise water tables should, therefore, also serve to reduce fire risk. Heather communities had the fewest reported fires, which suggests that management of heather, including rotational burning, is successful in reducing vulnerability to wildfire. Risk of a fire occurring and being reported is increased around access routes, with most fires occurring within 300m of roads and eroded paths, 750m of trampled paths, and within 2km of the Pennine Way. Additionally, there were significantly more reported fires on Access Land, with implications for increased fire risk since the extension of access land under CroW, but also for increased reporting. Topographic aspect has a considerable influence on the fire risk, with fires fewest on east-facing slopes.Secondly, a non-linear probit model is used to assess the chance of fires at different times of the year, days of the week and under various weather conditions. Analysis concludes that current and past rainfall damps fire risk, and the danger of fire increases with maximum daily temperature. Dry spells or recent fire activity also signal extra fire hazard. Certain days are fire prone, especially spring bank holidays, due to increased visitor numbers. Some months of the year are more risky, notably the April-May and July-August periods, reflecting the interplay between visitor numbers and the changing flammability of moorland vegetation. Flammability varies as seasonal plant phenology (the spring green wave) is superimposed upon summer soil moisture deficit. The model back-predicts earlier fires accurately. The number of fires is then forecast using future climate projections. Changes in climate variability and weather extremes generate most extra fire risk. Finally, a gradual rise in mean temperature was found to have only slight effect.The combination of climate modelling, temporal and spatial analysis is a powerful tool for predicting and managing future fire risk. There is much potential to produce a decision-making tool able to identify areas and times of highest risk, and to model the potential impact of fire risk management strategies under climate change scenarios.