Abstract: Data scarcity on the relations between vegetation types and soil properties have hindered the biological rehabilitation of degraded lands, particularly in the degraded sandy lands of Songnen Plain with overburden human actives. With six vegetation types, farmland, grassland, poplar forest(Populus spp.), larch forest(Larix gmelinii), pine forest(Pinus sylvestrys var. mongolica), and elm forest(Ulmus pumila), around the Xindian forest farm, we measured the soils from 0-20, 20-40 and 40-60 cm for soil bulk density, soil moisture, soil pH, soil electrical conductance(EC), soil organic carbon(SOC), alkali-hydrolyzed nitrogen(AN), total nitrogen(TN), available phosphorus(AP), total phosphorus(TP), available potassium(AK), total potassium(TK) and root density, and root water content. Despite large differences in vertical soil profile, root density, soil bulk density, soil moisture, soil pH, EC, TN, root water content, and SOC were significantly influenced by six vegetation types. Besides vegetation-dependent significant differences of root density markedly different from different soil layers(depth×vegetation interaction, P<0.05), the other seven parameters showed the same vegetation-related differences throughout whole 60 cm soil profile. Much higher fine root biomass in grass land, compared with forest(1.4-fold on average) and farmland(1.8 fold), indicated its greater advantage in fixing sandy soil. Much lower pH, EC, soil bulk density and soil moisture in different forests, compared with the peak values in grassland, manifested the afforestation effects in improving soil physics and decreasing soil alkali-degree with more water consumption. By correlation analysis, SOC accumulation and N were related with root density, bulk density and soil moisture, while P supply was related with soil pH and EC. Our results would provide the basic data to support soil remediation through revegetation in Songnen Plain.

Key words: Songnen Plain, vegetation types, soil, physicochemical properties, root