Environmental science and engineering specialists, PDP and data science specialists, Lynker Analytics have been selected by the Ministry for the Environment (the Ministry) to jointly develop a new approach to assess the survival of plants within riparian areas across Aotearoa New Zealand using Artificial Intelligence (AI).
Riparian areas are the strips of land beside drains, streams, rivers, and lakes. They include areas on-farm where the soils are wettest, such as wetlands, springs or seeps and gullies. Riparian planting has multiple environmental benefits including water filtration, erosion prevention, moderation of water flow, shading waterways, providing habitat for indigenous species, and keeping livestock out of waterways.
UAV wetland survey, Photo Marinus Boon.
Recently riparian planting has received a major boost through the Ministry’s Mahi mō te Taiao | Jobs for Nature funding programme. This programme manages funding across multiple government agencies to benefit the environment, people and the regions and will run until 2024, as part of the COVID-19 recovery package.
Under this contract, the PDP/Lynker team are assessing the suitability of a range of remote sensing methods to monitor riparian ecosystems. These include satellite, airborne and unmanned aerial systems.
For each sensor, the team will develop “deep learning” techniques to identify and map the extent of plants within riparian systems. “Deep learning is a subset of machine learning, using artificial neural networks to learn complex and intricate patterns in large data sets”, says Matt Lythe, Managing Director of Lynker Analytics.
Upper banks are often planted in taller, woody vegetation (trees, shrubs, tī kōuka/ cabbage trees, and ponga/tree ferns), while steeper lower banks are often planted in flood-tolerant herbaceous vegetation (sedges, harakeke/flaxes, and tussocks). For riparian plantings, farmers use mostly native plants with a mix of non-native plants for long-lasting growth and weather condition tolerance.
“An important goal of the project is to determine the ratio of woody vegetation, herbaceous vegetation, sedges and grasses in these environments,” says Marinus Boon, Remote Sensing and Wetland Ecologist at PDP. “We’d also like to spot any unintended plants such as invasive blackberry plants or gorse and broom,” he goes on to say.
The team will conduct a field survey of several pilot sites to test the system before providing advice to the Ministry on how to rollout the Riparian Planting monitoring system on a regional and/or national level.
Scheduled for completion in June 2022, this project will be “crucial in helping to establish the Ministry’s Mahi mō te Taiao | Jobs for Nature Environmental Impact Evaluation programme of work,” says Moana Everson from the Ministry.