A Combination of Plant NDVI and LiDAR Measurements Improve the Estimation of Pasture Biomass in Tall Fescue (Festuca arundinacea var. Fletcher)

Michael Schaefer,David Lamb

doi:10.3390/rs8020109

Abstract

The total biomass of a tall fescue (Festuca arundinacea var. Fletcher) pasture was assessed by using a vehicle mounted light detection and ranging (LiDAR) unit to derive canopy height and an active optical reflectance sensor to determine the spectro-optical reflectance index, normalized difference vegetation index (NDVI). In a random plot design, measurements of NDVI and pasture height were combined to estimate biomass with a root mean square error of prediction (RMSEP) equal to ±455.28 kg green dry matter (GDM)/ha, over a range of 286 kg to 3933 kg GDM/ha. The combination of NDVI and height measurements were observed to be more accurate in assessing total biomass than just the NDVI (RMSEP ± 846.51 kg/ha) and height (RMSEP ± 708.13 kg/ha). Based on the results of the study it was concluded the use of combined LiDAR and active optical reflectance sensors can help unlock the complex interrelationship between green fraction and biomass in swards containing both green and senescent material.

Highlights

The ability to monitor and map pasture biomass in extensive grazing systems provides graziers with vital information for making timely livestock management decisions such as set stocking rates or rotation intervals [1,2]
The primary aim of this paper is to evaluate the ability to estimate pasture biomass by combining a measure of pasture canopy height using light detection and ranging (LiDAR) and pasture canopy reflectance measurements using active optical sensor (AOS)
In some pasture plots where the senescent material is standing and intermingled with the green fraction, it is unlikely the LiDAR unit would be able to discern different levels of green fraction and this is probably why there remains some scatter in the plot of Figure 3b

Summary

Introduction

The ability to monitor and map pasture biomass in extensive grazing systems provides graziers with vital information for making timely livestock management decisions such as set stocking rates or rotation intervals [1,2]. Management decisions such as daily pasture allocation, conservation and supplementary feeding are an essential pathway to increasing the efficiency of pasture grazing systems [3,4,5,6]. Extrapolating point measurements to a paddock scale, or interpolating between point measures to examine sub-paddock variability is prone to errors To address this problem, a variety of other spatially-enabled, “on-the-go” pasture biomass measuring techniques have been developed. Trotter et al [9] provides a review of these techniques which include visual assessment, pasture height recording devices, weighted plate meters, combinations

Objectives

Methods

Results

Conclusion