Relationships between fire severity and post-fire landscape pattern following a large mixed-severity fire in the Valle Vidal, New Mexico, USA

► We map landscape metrics in New Mexico, USA, using a moving-window approach. ► Moving-window generated metrics reveal spatial heterogeneity in landscape pattern. ► Moving-window maps produce spatial and quantitative distributions of metric values. ► Different levels of fire-severity fire result in distinctive landscape patterns. ► High-severity fire led to the greatest variability of post-fire landscape pattern. The predominant fire regime associated with ponderosa pine ( Pinus ponderosa) forests in the southwestern US has shifted from the historic norm of frequent, low-severity fires to less frequent mixed-severity and crown fires. This change in the severity of fire has altered ponderosa pine forests from the open stands typical of pre-settlement times to even-aged, high-density stands at increased risk of crown fire. As a result, restoration plans and post-fire management practices must consider the spatial and temporal variability of fire severity in both mixed-severity and crown fire events because fire-severity patterns strongly influence post-fire ecological conditions. This study examines the landscape pattern of fire severity in the Ponil Complex Fire and applies a moving-window approach to post-fire landscape pattern measurement. The moving-window approach allows examination of the quantitative and spatial variability of landscape pattern, producing a more nuanced description of forest pattern when compared to whole-landscape or patch-based metrics. The fire resulted in a complex mosaic of fire patches and forest-structure changes. In high-severity fire patches, mean and median values of many post-fire landscape metrics were markedly different from those in low and moderate-severity patches. Landscape pattern in high-severity patches also had the greatest variability of metric values, suggesting that high-severity fire patches require a spatially mediated management response to fire. Categorical fire-severity maps and traditional landscape-pattern assessment would not be able to identify these spatially variable post-fire conditions.