Fire history and its relations with land use and climate over three centuries in the central Appalachian Mountains, USA

AIM: Our aims were to: (1) reconstruct the fire history of pine–oak forests in the central Appalachian Mountains, USA, with an annual resolution over as long a time period as possible using dendroecological techniques; (2) estimate the frequency of fire in the study area before the fire‐suppression era; and (3) investigate how variations in land use and climate have affected the occurrence of fire in the study area. LOCATION: Temperate forests at three study sites within the central Appalachian Mountains, Virginia, USA. METHODS: Cross‐sections were taken (sawn) from fire‐scarred pine (Pinus L.) trees growing in pine‐dominated patches within a hardwood forest matrix. Dendroecological techniques were used to date the scars, which were used to calculate fire intervals. A variety of analyses were carried out: Pearson correlation analysis, to investigate whether fire activity varied over time (under changing land uses); Kruskal–Wallis analysis, to examine whether fire frequency varied spatially (among study sites); chi‐square analysis, to test whether scar seasonality changed temporally; and superposed epoch analysis, to explore whether fire activity was associated with interannual climatic variations in moisture, as characterized by the Palmer drought severity index (PDSI). RESULTS: Fire scars dated back to the 17th or early 18th century (depending on site). The filtered composite fire interval, considered to be a particularly reliable estimate of fire interval, averaged between 6 and 8 years. Fire frequency remained fairly constant from the beginning of the record until effective fire suppression began in the early 20th century, after which burning virtually ceased. Fire occurred more frequently at the easternmost site, which was located in the Blue Ridge province of the Appalachian Mountains, than at the other two sites, in the Ridge and Valley province. Scar seasonality showed no discernible trend over time. Fire was associated with low PDSI (i.e. dry years) at two of the study sites. MAIN CONCLUSIONS: Fire occurred frequently at these central Appalachian study sites during the period of aboriginal depopulation that preceded European colonization, and throughout the periods of European settlement and industrialization (with mining, logging and railroads) that followed. Our results match those from other fire‐history sites in the central and southern Appalachian Mountains, and suggest that fire was an important factor influencing vegetation development in the temperate forests covering this region.