Post-wildfire soil erosion in the Mediterranean: Review and future research directions

Abstract Wildfires increased dramatically in frequency and extent in the European Mediterranean region from the 1960s, aided by a general warming and drying trend, but driven primarily by socio-economic changes, including rural depopulation, land abandonment and afforestation with flammable species. Published research into post-wildfire hydrology and soil erosion, beginning during the 1980s in Spain, has been followed by studies in other European Mediterranean countries together with Israel and has now attained a sufficiently large critical mass to warrant a major review. Although variations in climate, vegetation, soil, topography and fire severity cause differences in Mediterranean post-wildfire erosion, the long history of human landscape impact up to the present day is responsible for some its distinctive characteristics. This paper highlights these characteristics in reviewing wildfire impacts on hydrology, soil properties and soil erosion by water. The ‘mosaic’ nature of many Mediterranean landscapes (e.g. an intricate land-use pattern, abandoned terraces and tracks interrupting slopes) may explain sometimes conflicting post-fire hydrological and erosional responses at different sites and spatial scales. First-year post-wildfire soil losses at point- (average, 45–56 t ha − 1 ) and plot-scales (many − 1 and the majority − 1 in the first year) are similar to or even lower than those reported for fire-affected land elsewhere or other disturbed (e.g. cultivated) and natural poorly-vegetated (e.g. badlands, rangeland) land in the Mediterranean. The few published losses at larger-scales (hillslope and catchment) are variable. Thin soil and high stone content can explain supply-limited erosion preceding significant protection by recovering vegetation. Peak erosion can sometimes be delayed for years, largely through slow vegetation recovery and temporal variability of erosive storms. Preferential removal of organic matter and nutrients in the commonly thin, degraded soils is arguably just as if not more important than the total soil loss. Aspect is important, with more erosion reported for south- than north-facing slopes, which is attributed to greater fire frequency, slower vegetation recovery on the former and with soil characteristics more prone to erosion (e.g. lower aggregate stability). Post-fire wind erosion is a potentially important but largely neglected process. Gauging the degradational significance of wildfires has relied on comparison with unburnt land, but the focus for comparison should be switched to other agents of soil disturbance and/or currently poorly understood soil renewal rates. Human impact on land use and vegetation may alter expected effects (increased fire activity and post-wildfire erosion) arising from future climatic change. Different future wildfire mitigation responses and likely erosional consequences are outlined. Research gaps are identified, and more research effort is suggested to: (1) improve assessment of post-wildfire erosion impact on soil fertility, through further quantification of soil nutrient depletion resulting from single and multiple fire cycles, and on soil longevity; (2) investigate prescribed fire impacts on carbon release, air pollution and nutrient losses as well as on soil loss; (3) isolate hillslope- and catchment-scale impacts of soil water repellency under Mediterranean post-wildfire conditions; (4) test and refine application of cosmogenic radionuclides to post-wildfire hillslope-scale soil redistribution at different temporal scales; (5) use better temporal resolution of sedimentary sequences to understand palaeofire-erosion–sedimentation links; (6) quantify post-wildfire wind erosion; (7) improve the integration of wildfire into an overall assessment of the processes and impacts of land degradation in the Mediterranean; and (8) raise public awareness of wildfire impact on soil degradation.