Estimating Biomass for New Zealand Pasture Using Optical Remote Sensing Techniques

Abstract One essential requirement for an increase in the efficiency of pastoral grazing management in New Zealand is the development of a quick, reliable, and objective method for assessing pasture bulk. The traditional method for estimating biomass entails harvesting and dry‐weighing sample quadrats, then extrapolating from the sample to the paddock. This is labour‐intensive and slow, so we seek a means of ‘remote weighing’ of the pasture using optical remote sensing techniques. Field trials in the Waikato region (latitude 38°S) of New Zealand have demonstrated that measuring pasture reflectance in three wavelength bands (near‐infrared, red, green) permits us to retrieve green biomass to within an average root-mean square error of 260 kg / ha, for pasture green‐bulk ranging from 70 to 4000 kg / ha, using a three‐band regression. We find that the usual remote measure of greenness, NDVI (normalized difference vegetation index), saturates at moderate pasture densities, so is not useful in the New Zealand p...