Eye in sky to check lakes
Scientists will use satellite imagery in battle to control toxic algal bloom
Scientists are set to test the health of New Zealand’s lakes — from space. With summer turning public attention to the condition of our under-pressure freshwater estate, a new study is focusing on the problem of blue-green algal blooms.
These often-toxic blooms, which float on the surface of lakes, have become common each season. Health warnings were put in place at Rotorua’s Lake Okaro and Lake Rotoehu even before summer began.
Yet just 0.5 per cent of New Zealand’s lakes are monitored for bluegreen algae, or cyanobacteria.
“We are concerned that we are missing a lot of lakes where cyanobacteria may exceed current guidelines,” said Professor Ian Hawes, a freshwater ecologist at the University of Waikato. “This is because there is good evidence that intensive agriculture, and the increased run-off of the nutrients that support cyanobacterial growth, plays an important role in promoting toxic blooms.
“Models indicate there are many more lowland lakes where catchment land use suggests vulnerability to potentially toxic blooms than are currently monitored.”
In a new million-dollar project, a team of scientists led by Hawes will develop a new way to continually watch for blooms, by using satellite imagery that picks up and measures cyanobacteria’s blue.
Scientists had already used a similar approach to assess the dominant colour of New Zealand lakes, and had begun to measure concentrations of the sediment and algae within them.
“What we are really looking for is a way to monitor the status of a large number of New Zealand lakes — we hope more than 1000 — using freely available remote sensing images.”
Hawes said recent modelling of vulnerability to blooms has involved extrapolating water quality measurements from a few sites to other lakes with similar degrees of intensification. “But [it] depends on the assumption that lakes or rivers with similar catchments and land use will have similar water quality,” he said.
“Such an approach is so much stronger if we can provide a validation data set.”
The main issue was that all models have errors, and while they offered a good overall picture, when it came to specific lakes, they can be wrong.
Having a way of looking at lakes as they change is critical to evaluating the success of interventions.
Hawes said the study wouldn’t be quite as simple as teaching a computer model how to colour-code.
New Zealand was home to a wide range of lake types and forms of cyanobacteria. Our lakes also had varying appearances — from peat-stained waters in the West Coast, to clear dune lakes in the north.
“Each of these have a natural background colour, and we need to be able to differentiate the cyanobacterial colour signal against this,” Hawes said.
First, the team would develop an optical classification, based on factors like geography, the effects of nearby vegetation and satellite measurements, enabling them to identify the type of background colour of lakes without visiting them all.
The next part would use the colour of lake water, measured with boat and drone-based instruments, to infer cyanobacteria and other algae, and finally to use those data to develop satellite-based measurements.
“It’s a slightly risky endeavour, but we are confident that we will be able to get to the point where we can use drone-based measurements to determine cyanobacteria,” Hawes said.
“This is because the instruments that we can fly on the drone capture colour with high resolution — or lots of colours — and small pixel size, of the area of lake surface measured.”
Current generations of satellites resolved colour into fewer bands, with a bigger pixel size, which would pose a challenge.
“The one thing we can be sure of is that there will be a generation of satellites that will be deployed that will have better and better wavelength and area resolution.”
He hopes they’ll end up with a tool able to provide surveillance of dangerous water quality nationwide.