Previous fires moderate burn severity of subsequent wildland fires in two large western US wilderness areas

Wildland fire is an important natural process in many ecosystems. However, fire exclusion has re-duced frequency of fire and area burned in many dry forest types, which may affect vegetation structure and composition, and potential fire behavior. In forests of the western U.S., these effects pose a challenge for fire and land managers who seek to restore the ecological process of fire to ecosystems. Recent research suggests that landscapes with unaltered fire regimes are more ‘‘self-regulating’’ than those that have experienced fire-regime shifts; in self-regulating systems, fire size and severity are moderated by the effect of previous fire. To deter-mine if burn severity is moderated in areas that re-cently burned, we analyzed 117 wildland fires in 2 wilderness areas in the western U.S. that have experienced substantial recent fire activity. Burn severity was measured using a Landsat satellite-based metric at a 30-m resolution. We evaluated (1) whether pixels that burned at least twice since 1984 experienced lower burn severity than pixels that burned once, (2) the relationship between burn severity and fire history, pre-fire vegetation, and topography, and (3) how the moderating effect of a previous fire decays with time. Results show burn severity is significantly lower in areas that have re-cently burned compared to areas that have not. This effect is still evident at around 22 years between wildland fire events. Results further indicate that burn severity generally increases with time since and severity of previous wildfire. These findings may assist land managers to anticipate the consequences of allowing fires to burn and provide rationale for using wildfire as a ‘‘fuel treatment’’. Key words: burn severity; dNBR; fire history; interacting fires; reburn; wilderness; wildland fire use