Land-use history in ecosystem restoration: A 40-year study in the prairie-forest ecotone

Land-use history, recent management and landscape position influence vegeta tion at the Rockefeller Experimental Tract (RET), a 40-year-old restoration experiment in northeast Kansas. RET is representative of the prairie-fores t ecotone, containing native tallgrass prairie and oak-hickory forest, but unique in having tracts of replanted prairie, seeded in 1957, that have und ergone long-term restoration treatments: burned, grazed, mowed, or untreate d. A land-use history database for RET was compiled using a geographic info rmation system to integrate historic and contemporary sources of informatio n. Restoration management on the reseeded prairie has had a profound effect on forest development: mowing or burning precluded forest establishment (< 3% forest cover), whereas portions of untreated or grazed areas became heav ily forested (>97% forest cover). Forest colonization depends upon biotic a nd edaphic conditions at the time restoration was initiated: for areas repl anted to prairie and managed by grazing, forestation was 6% on land in cult ivation prior to replanting, 20% on former pastureland, and 98% on land def orested just before replanting. Patterns of forest colonization were also s ignificantly associated with three landscape positions: near existing fores t, along water courses, and along ridge tops. Additionally, land-use histor y analyses showed that the presence of various prairie and forest species r esulted from persistence and not from colonization following restoration. B ecause of the lasting imprint of historic land use on the landscape, our re sults indicate that it is essential that restoration studies be evaluated w ithin a site-specific historical context.