Premium
Assessment of land levelling effects on lowland soil quality indicators and water retention evaluated by multivariate and geostatistical analyses
Author(s) -
Timm Luís Carlos,
Pires Luiz Fernando,
Centeno Luaunes,
Bitencourt Dioni Glei Bonini,
Parfitt José Maria Barbat,
Campos Alexssandra Dayanne Soares
Publication year - 2020
Publication title -
land degradation and development
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 81
eISSN - 1099-145X
pISSN - 1085-3278
DOI - 10.1002/ldr.3529
Subject(s) - levelling , environmental science , soil water , topsoil , kriging , hydrology (agriculture) , mathematics , water quality , principal component analysis , multivariate statistics , soil quality , soil science , statistics , geography , geology , geotechnical engineering , ecology , cartography , biology
Studies focusing on the impact of levelling on the management of available water for rice growth and on soil hydro‐physical quality indicators cultivated in lowland soils under aerobic conditions are still scarce. The objective of this study was to evaluate the levelling impact on soil water retention and soil quality indicators by multivariate and geostatistical analyses. A 10 × 10 m grid was used to sample the 0‐ to 20‐cm layer in a 1.0‐ha experimental field, before and after levelling. Sand and clay contents, macroporosity, bulk density (Bd), organic matter (OM), weighted average diameter of aggregates, and the water retention curve were measured. Available water capacity, structural stability, and S indexes and relative field capacity were calculated. Descriptive statistics, Kolmogorov–Smirnov test, and geostatistical analysis were used. Maps of all data sets significantly correlated with Bd and OM after levelling were constructed using ordinary and indicator kriging. Kaiser–Meyer–Olkin and Bartlett tests were performed for factorial and principal component analyses which were used to reduce the dimensionality of data and of main principal component maps. Levelling negatively affected the majority of soil quality indicators, caused an increase of their spatial range, and an improved the goodness of fit of the semivariogram models used. The majority of variables were best fitted by the Gaussian model after levelling. Bd and OM probability maps were found to be useful tools for farmers wanting to utilize different strategies for topsoil management, with the goal of improving soil quality of levelled areas for future land recuperation operations.