Integration of Soil Electrical Conductivity and Indices Obtained through Satellite Imagery for Differential Management of Pasture Fertilization

João Serrano,José Calado,Shakib Shahidian,Luís Paixão,Mário De Carvalho,José Marques Da Silva

doi:10.3390/agriengineering1040041

Abstract

Dryland pastures in the Alentejo region, located in the south of Portugal, normally occupy soils that have low fertility but, simultaneously, important spatial variability. Rational application of fertilizers requires knowledge of spatial variability of soil characteristics and crop response, which reinforces the interest of technologies that facilitates the identification of homogeneous management zones (HMZ). In this work, a pasture field of about 25 ha, integrated in the Montado mixed ecosystem (agro-silvo-pastoral), was monitored. Surveys of apparent soil electrical conductivity (ECa) were carried out in November 2017 and October 2018 with a Veris 2000 XA contact sensor. A total of 24 sampling points (30 × 30 m) were established in tree-free zones to allow readings of normalized difference vegetation index (NDVI) and normalized difference water index (NDWI). Historical time series of these indices were obtained from satellite imagery (Sentinel-2) in winter and spring 2017 and 2018. Three zones with different potential productivity were defined based on the results obtained in terms of spatial variability and temporal stability of the measured parameters. These are the basis for the elaboration of differentiated prescription maps of fertilizers with variable application rate technology, taking into account the variability of soil characteristics and pasture development, contributing to the sustainability of this ecosystem.

Highlights

Dryland pastures are the principal source of animal feed in extensive animal production systems in the Alentejo region of south of Portugal
The aim of this paper is to demonstrate the interest of associating expedited soil electrical conductivity (ECa) survey with monitoring of time series of indices obtained from satellite imagery in order to define and validate homogeneous management zones (HMZ) (Figure 1)
The same happens in the pasture quality parameters (CP, with coefficient of variation (CV) of 15.6%; and neutral detergent fiber (NDF), with CV of 7.0%), which seems to indicate a significant relationship between these pasture quality parameters and the normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) indices, what is in line with other studies [10,13]. These results show that the pattern of variability detected at vegetation level by optical sensors from remote sensing does not reflect, the pattern and extent of soil nutrient variability

Summary

Introduction

Dryland pastures are the principal source of animal feed in extensive animal production systems in the Alentejo region of south of Portugal. These occupy soils that have low fertility and, simultaneously, important spatial variability in key parameters. Even productive and intensively managed grasslands comprise a wide range of vegetation types, and there is still uncertainty about the extent to which we can measure individual properties of that vegetation needed for application of automated, reliable, and time-saving management techniques. In addition to different vegetation types, the annual variations of floristic composition and vegetation dynamics introduce significant variability and uncertainty into standardized sensing techniques on permanent grassland. In the case of pastures under tree cover such as the Montado of cork oak (Quercus suber L.) or of holm oak (Quercus ilex ssp. rotundifolia Lam.) the effect of trees on pasture is significant [3] and a direct consequence of the extent to which these trees modify the microclimate and soil properties [4]

Objectives

Methods

Conclusion