Selecting the minimum data set and quantitative soil quality indexing of alkaline soils under different land uses in northeastern China

Understanding the influences of land use conversions on soil quality (SQ) and function are essential to adopt proper agricultural management practices for a specific region. The primary objective of this study was to develop soil quality indices (SQIs) to assess the short-term influences of different land uses on SQ in semiarid alkaline grassland in northeastern China. Land use treatments were corn cropland (Corn), alfalfa perennial forage (Alfalfa), monoculture Lyemus chinensis grassland (MG) and successional regrowth grassland (SRG), which were applied for five years. Twenty-two soil indicators were determined at 0-20 cm depth as the potential SQ indicators. Of these, thirteen indicators exhibited treatment differences and were identified as the total data set (TDS) for subsequent analysis. Principal component analysis was used with the TDS to select the minimum data set (MDS), and four SQIs were calculated using linear/non-linear scoring functions and additive/weighted additive methods. Invertase, N:P ratio, water-extractable organic carbon and labile carbon were identified as the MDS. The four SQIs performed well, with significant positive correlations (P < 0.001, n = 16) among them. However, the SQI calculated using the non-linear weighted additive integration (SQI-NLWA) had the best discrimination under different land-use treatments due to the higher F values and larger coefficient of variance as compared to the other SQIs. The SQI value under the MG treatment was the highest, followed by that under the SRG and Alfalfa treatments, and all of these were significantly higher than that of Corn treatment. These results indicated that conversion of cropland to perennial forage or grassland can significantly improve the SQ in the Songnen grassland. In addition, SQI-NLWA can provide a better practical, quantitative tool for assessing SQ and is recommended for soil quality evaluation under different land uses in semiarid agroecosystems. (c) 2017 Elsevier B.V. All rights reserved.