The gene diversity or expected heterozygosity (HE) is based on the allele frequency and is often used as a measure of genetic variability of populations. Knowing the pattern of spatial distribution of HE can be useful for determining strategies of conservation and sampling of collections of individuals. In addition, it can allow one to detect genetic boundaries in a landscape. We adapted a Wombling method based on assignment tests in a circular moving window extensively sampled over the study area in order to estimate HE at points of a prediction grid. The function sHe(), package biotools, is an easy-to-use and flexible implementation in R language that accepts as input geographical and genotyping data. The package biotools is distribution-free under the GPL-2/3 license and currently available from the Comprehensive R Archive Network (CRAN) at . The R platform and all R dependencies are similarly available from CRAN.

Rainfall is among the climatic factors that most affect production, as in the Brazilian Cerrado. Non-destructive and automated phenotyping methods are fast and efficient for genotype selection. The objective of this work was to evaluate, under field conditions, the morphophysiological changes, yield, and grain quality of soybean (Glycine max L. Merrill) under water stress in the Brazilian Cerrado. The plots comprised six soybean cultivars and the subplots of four water regimes, corresponding to 31, 44, 64 and 100% of crop evapotranspiration replacement. The experiments were conducted from May to September 2018 and 2019. An irrigation system with a bar of sprinklers with different flow rates was used. Gas exchange, vegetation indices (measured using a hyperspectral sensor embedded in a drone), yield and grain quality were evaluated. Water stress had different effects on gas exchange, vegetation indices, grain yield and chemical composition among the cultivars. Embrapa cultivar BRS 7280 Roundup ready (RR) and Nidera cultivar NA 5909 RG (glyphosate resistant) are yield stable and have a greater tolerance to drought. BRS 7280RR showed a higher tolerance to drought and higher water use efficiency (WUE) than all other tested cultivars. Vegetation indices, such as the NDVI (Normalized Difference Vegetation Index), correlated with the morphophysiological traits, such as plant height, were the most responsive variables to water stress. The NDVI can be used to predict soybean yield as a tool in a selection program under drought.

Potassium (K) can be easily lost by the leaching process. The objective of this study was to quantify K leaching in clayey and sandy soils under increasing irrigation depths. The experiment was conducted in 2014, in a protected environment, after extraction of undisturbed soil columns, with diameter of 144 mm and height of 300 mm. The columns were arranged in a randomized complete block in a factorial design with three replicates, corresponding to eight treatments: two soil types and four water depths, equivalent to 50, 100, 150 and 200% of the amount of water necessary to bring the soil moisture to field capacity. Potassium chloride, with 58% of K₂O, was used as K source. Water replacement in the columns was performed every three days, for a period of 81 days. After conducting joint analysis of the data according to the factor Time, a second-degree response surface model was fitted and line charts were also used to study the factors Time and Water. After the evaluations, it was found that the higher the applied water depth, the greater the percolated amount of the K⁺ ion.