Kelly Gang - Pasture estimation using AO

Summary

Being able to measure green pasture dry matter yield (DMY) is important for both research and developing profitable livestock feeding programs. The method commonly used to do this involves cutting the pasture to ground level and washing, drying and weighing it.  While the results are accurate the method is very slow and very tedious. This limits its use.
Active Optical Sensor technology, in the form of a GreenSeeker or Crop Circle, is a hand held tool which has been used to successfully and quickly measure a range of things in a wide range of diverse situations.  It has not been used to measure green pastures. The aim of this project was to assess their capacity to quickly and accurately measure green pasture biomass.
The project involved one pasture on each of three properties (Mountjup, Woorndoo and Coojar) in SW Victoria. All the pastures were at least 25 years old and were within 65 km of Hamilton, Victoria.  All the pastures were grazed from time to time.  Phalaris was the perennial grass in two of the paddocks and perennial ryegrass was the main grass in the other.
The green pasture was measured on these three properties on three occasions (autumn, winter and spring) in each of three years (2014-16). On each occasion 12 or 15 points were measured and sampled. The aim was to select points with a similar botanical composition but the DMY range was to be as wide as possible.
The procedure followed at each sampling point was: Visual estimation of DMY, clover percent and percent of standing dead pasture. Falling plate meter to measure pasture height (cm).GreenSeeker and Crop Circle measurements. Photograph the pasture.Cut pasture in the quadrat to ground level using a shearing hand piece.   Wash, sort into botanical components, and dry.   
It soon became clear that the GreenSeeker only partly told the story; pasture height appeared to be far more important.  Because of this it was included in future measurements. 
Both the GreenSeeker and Crop Circle were used in the early stages of this project but continuous problems with the operation of the Crop Circle led to its use being discontinued. 
Only GreenSeeker data is reported here. 
The results have been analysed on two levels, one (reported here) involving an analysis of local results and a second (reported in B.GSM.0010) more complex analysis combining data from all the sites involved in the broader project (see bibliography) including the botanical composition. 
The local data was analysed using a simple regression
Y = a + bx with y = DMY, x = GreenSeeker reading or pasture height, and a and b being constants. 
The residual deviation (R2) was the outcome of most interest. The higher the R2 the better the association between DMY and the GreenSeeker reading, or pasture height.
In comparing the R2 for pasture height and DMY, and GreenSeeker and DMY, at each campaign pasture height, while not perfect was the highest in 19 out of 25 cases, well above the GreenSeeker figures of 6 out of 25.  Any future plans to develop tools to measure green pasture biomass must include pasture height.
Producers involved in the project were enthusiastic about the project goals and took a keen interest in the projects progress. 
Producers, reflecting on what IT has done for their businesses in recent years, were excited at the thought of being able to measure the amount of green biomass in a paddock as they drove from gate to gate and on reaching the second gate tapping an iPad to get the details about the pasture, and the implications of them for the stock grazing the paddock. A whole farm assessment within half a day was looking a reality and the tools required to do it would find a ready market. Cost will not be a barrier to adoption.
The initial enthusiasm for using the GreenSeeker for measuring pasture mass was proven to be unfounded. Height was a better indicator of pasture biomass than NDVI and that measuring NDVI using the GreenSeeker didn't add much to improve the calibration than just measuring height in Western Victoria.