Abstract

ERAZO-MESA, Edwin; ECHEVERRI-SANCHEZ, Andrés  and  URRUTIA COBO, Norberto. New algorithm to compute soil spatial sampling patterns in agricultural experiments. Cienc. Tecnol. Agropecuaria [online]. 2022, vol.23, n.1, e2306.  Epub Dec 31, 2021. ISSN 0122-8706.  https://doi.org/10.21930/rcta.vol23_num1_art:2306.

Soil spatial variability is an essential factor in understanding the change of the dependent variables in agricultural experiments. Soil sampling is carried out based on a spatial pattern, which can be random or systematic. The work aimed to validate a new algorithm to generate spatial patterns for sampling soils in the context of agricultural experiments. In this sense, three functions were designed using the software R and compared with five applications (three software and two R packages). The validation was performed by replicating three spatial patterns of agricultural experiments reported in previous studies and comparing hand localization of sampling points in a sugarcane harvesting experiment with localization of points generated by the algorithm. Results show that the new algorithm can exclusively compute the localization of sampling points by experimental unit (represented by a polygon) and center these in the corresponding polygon. Other characteristics, such as computing the most common spatial point patterns and generating points along crop lines, are also founded in the other applications. Regarding in-field validation, the average distance between points generated by the algorithm and those localized manually in the field is 2.58 m. The average distance between hand located points and the closest crop row line is 0.46 m. In conclusion, the new algorithm allows locating sampling points in specific sites of the plots as the highest or lowest part of crop rows.

Keywords : furrow; precision agriculture; R software; Real Time Kinematic (RTK); soil sampling; spatial variability.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )