Soil and Weed Survey for Spatially Variable Herbicide Application on Railways

Abstract

Weed control on railways is frequently done by herbicide application, using large amounts of chemicals and labour. Only a few hectares of surface area on a railway can be many kilometres long so high levels of spatial variability of soils and weeds are expected. Railway sprayers use similar technology to crop sprayers so it is possible to adopt concepts from precision agriculture for weed control. The objective of this study was to evaluate spatial variability of soils and weeds on railways in order to define weed control strategies and basic characteristics of a railway sprayer. This work is part of a research project aiming at the development of a system for spatially variable herbicide application on railways. Soil and weed data were collected from five different railway companies in Brazil. These data were used to simulate treatment maps based on weed control strategies and to estimate technical requirements for spatially variable herbicide application. The results showed that soil and weed survey gave important information for defining weed control strategies and spraying equipment. The sprayer must treat the three swaths (left, central and right parts of the railway surface) separately applying at least two different herbicides at the same time. On average, a variable dose rate application based on the treatment maps generated in this work could save up to 59% of herbicides.