Ask any rancher or other forage producer — insuring pasture land can be a major pain in the neck.
Because of the limitations of current assessment tools, in the event of a wreck producers can be paid nothing or an amount that does not reflect actual loss. Instead, payments are made on an area-wide basis depending on satellite results or weather station information some distance away.
But an Alberta project using new high-resolution satellite imagery aims to make pasture insurance more individualized and similar to crop insurance.
- Read more: Keeping land in grass has a side benefit
“We’re trying to find a way to measure pasture production on an individual ranch in terms of pounds per acre, said Richard McConnell, consultant with DYMAC Risk Management Solutions and project lead.
“If we can do that we could then create a history of pasture production – just like wheat – and let ranchers insure their pasture production relative to that history. Agricultural Finances Services Corporation (AFSC) is quite interested in the project and offering its support.”
In the past, pasture has been “overlooked” because of difficulty of measuring production, he said.
“Even ranchers can’t often tell you how much pasture in pounds per acre they produce.”
But refinements in satellite imaging technology has changed that, and Alberta is at the forefront, he said.
“AFSC was the first to use satellites for pasture insurance anywhere in the world,” said McConnell. “However, when it first started some 10 to 15 years ago, the most affordable satellite imagery obtained was a one-square-kilometre resolution.
“And since you need several images, in this case square kilometres, that meant the smallest insurance payment area that could be offered was one township. It wasn’t small enough to measure one ranch by itself in most cases.”
The satellite being used for the project has a 250-square-metre resolution and the imagery is only going to get better.
“In the future, we expect affordable satellite information might be available at a 30-square-metre resolution,” said McConnell. “Once you get down to 30 square metres you can get a lot of pixel images in a quarter section. You can refine your crop insurance availability to ranchers on a much smaller scale.”
Good data is key
The project, funded by Growing Forward 2 and the Canadian Cattlemen’s Association, is being conducted on native pasture on seven farms and ranches across the province. It uses on-the-ground observation to test the accuracy of satellite imaging. That data can then be used to translate historical satellite data, archived over many years, into pounds per acre for individual ranches. The grass production history can be used to set insurance coverage and then, in the current year, to assess production to make a payment if there has been a loss.
“The trick is to get grass production on-the-ground measured at the same resolution,” said McConnell. “Using a hand-held device called a spectrometer, which is calibrated to the satellite, in conjunction with the physical clipping of grass samples taken in the field provides the apples-to-apples comparison.”
The spectrometer, like the satellite, provides an index called the Normalized Difference Vegetation Index (NDVI).
“The NDVI is an index measured from two sensors on the spectrometer — one sensor measures infrared light and red light coming down from the sun and the other measures those same light wavelengths reflecting up from the ground,” he said.
“The NDVI is created from comparing light absorbed by plants and light reflected by plants. However, the sensors don’t actually measure the light absorbed. But since all light is either absorbed or reflected, the difference between the light coming down from the sun and that which bounces up from the ground gives you the amount of light that’s absorbed.”
There are two distinct operations taken in the field using the spectrometer. The first verifies the spectrometer is actually calibrated to the satellite (by taking measurements in the grids used by the satellite.
“The MODIS (Moderate Resolution Imaging Spectroradiometer) satellite goes around the Earth every day and takes the same pixel image of 250 square metres. In each of those 250-square-metre areas there is a central point that can be located using a GPS instrument. That means a series of half-metre spectrometer readings can be taken within a 250-square-metre area known as a pixel and averaged to see how close they are to the one satellite reading for that same 250-square-metre area on the same day. If those measurements are close then the spectrometer is calibrated to the satellite.”
The second in-field operation defines the relationship between the spectrometer and grass production. To do this, researchers lay down a half-metre steel ring at defined areas within a satellite pixel in the field, take a spectrometer image, and clip the grass within the ring.
“The spectrometer is really measuring the green, or the chlorophyll, in the plant. Many grass samples with corresponding NDVI readings from the spectrometer are required and more is always better,” said McConnell.
But each is time consuming. The grass is dried to zero per cent moisture; sorted to separate green grass from forbs, carry-over or brown grass; and then weighed.
“The sorting process is the most time-consuming aspect of the entire project.”
The researchers are currently in the middle of the second year of the three-year project, so only have one year of completely sorted grass samples to analyze. But what they have seen so far is encouraging.
“We found the readings of the spectrometer were very close to that of the satellite,” said McConnell. “It won’t be exact because we only took 60 or so images in each pixel and satellite images can be inhibited by cloud cover. But just the same, the measurements from the spectrometer and the satellite are very close.”
Given the historical archive of imagery, McConnell said the spectrometer could be a highly cost-effective tool for assessing native grassland.
“We have historical satellite imagery so we can get a historical average for your ranch.”
That, in turn, allows comparison to the current year’s production to past production.
“If one year is a very poor production season the insured rancher would be in line for an insurance payment. The satellite can calculate both the coverage for your insurance policy as well as the loss adjustment.”