Mitchell, Wm. W. (Agricultural Experiment Station, School of Agriculture and Land Resources Management, University of Alaska, 1976-05)
      Revegetation studies commenced by the Alaskan Agricultural Experiment Station in 1970 on Amchitka Island culminated in 1973 with the seeding of disturbed areas associated with the nuclear testing program. Cool temperatures coupled with strong winds and a high incidence of fog and cloud cover impose a tundra aspect on Amchitka, one of Alaska's most southerly land areas. Natural revegetation is undependable for the near term. Twenty-two perennial grasses, two clovers, and four annual grasses were tested on different soil types at low to medium-high (480 ft) elevation sites. At higher elevations severe winds and frost action maintain a barren-ground aspect. Relatively humic, acidic sites were the least favorable, a test site gravel pad the most favorable. Cultivars of red fescue--Boreal, Pennlawn, and Highlight chewings— and an experimental entry of Bering hairgrass, taxa conspecific with species found on the island, and Engmo timothy performed the best. Kentucky bluegrasses and reed canarygrass grew moderately well. Wheatgrasses, wildrye, bromegrass, creeping foxtail, grandis alkaligrass, redtop bentgrass, and white and alsike clover performed unsatisfactorily at some or all of the sites. The revegetation seeding mix included Boreal red fescue, Highlight chewings fescue, Bering hairgrass, and annual ryegrass. Fertilization was necessary to establish plants on most sites. Plants responded erratically to added N on relatively humic, acidic soils, but more normally on gravelly and subsoil sites. Raising the P ration improved fertilizer response. Fertilization greatly enhanced growth on a disturbed site undergoing natural revegetation.