Generation of Lime Recommendation Maps Using Traditional Thematic Maps and Management Zones

Abstract

ABSTRACT The uniform management used in conventional agriculture is not the most efficient strategy for managing agricultural areas. In this way, precision agriculture (PA), based on assessments of the variability of chemical, physical, and biological attributes, seeks to optimize the activities management by site-specific management of inputs, to increase profitability in the production system, and to minimize environmental impacts. In this context, this work aimed to compare two techniques to make traditional thematic maps (TTMs) and management zones (MZs) for site-specific lime application, using data from three commercial agricultural areas (Western Paraná State, Brazil) and three methods to calculate the lime requirement (LR): (1) neutralization of exchangeable aluminum (Al3+) (NEM), (2) increase of calcium (Ca2+) and magnesium (Mg2+) (KM), and (3) base saturation method (BSM). The LR was calculated on AgDataBox-Map (ADB-Map) web platform, using AgDataBox-Map-Aplic module (ADB-Map-Aplic). ADB-Map is an application of the web platform, which aims at integrating data, software, procedures, and methodologies for Digital Agriculture. The Variance Reduction (VR%), Fuzziness Performance Index (FPI), Modified Partition Entropy (MPE), Improved Cluster Validation Index (ICVI), Smoothness Index (SI%) and Fragmentation index (FI%) indices were used to assess MZs quality, and the Kappa and global accuracy indices compared the agreement between TTMs and MZs. The results showed that the BSM method presented the best results in LR and distribution. According to the indices, MZs delineated with two classes generally (45 of the 47) showed the best results. MZs with application rates in each class calculated by the median of the lime quantities provided the lowest total LR per area. Thus, the MZs delineation allows applying lime more efficiently, reducing production costs, and allowing more sustainable use of inputs.