Detecting soil degradation and restoration through a novel coupled sensor and machine learning framework

Lead Research Organisation: Lancaster University
Department Name: Lancaster Environment Centre

Abstract

Overview
In this proposal we outline an ambitious cross-disciplinary project focused on detecting soil degradation and restoration through a novel multi-functional soil sensing platform that combines conventional and newly created sensors and a machine learning framework. Our proposed work directly addresses the Signals in the Soil call to 'advance our understanding of dynamic soil processes that operate at different temporal/spatial scales.' Through the creation of an innovative new approach to capturing and analysing high frequency data from in-situ sensors, this project will predict the rate and direction of soil system functions for sites undergoing degradation or restoration. To do this, we will build and train a new mechanistically-informed machine learning system to turn high frequency data on multiple soil functions, such as water infiltration, CO2 production, and surface soil movement, into predictions of longer term changes in soil health including the status of microbial processes, soil organic matter (SOM) content, and other properties and processes. Such an approach could be transformative: a system that will allow short-term sensor data to be used to evaluate longer term soil transformations in key ecosystem functions. We will start our work with a suite of off-the-shelf sensors observing multiple soil functions that can be installed quickly. These data will allow us to rapidly initiate development and training of a novel mechanistically informed machine learning framework. In parallel we will develop two new soil health sensors focused on in-situ real time measurement of decomposition rates and transformation of soil colour that reflects the accumulation or loss of SOM. We will then link these new sensors with a suite of conventional sensors in a novel data collection and networking system coupled to the Swarm satellite network to create a low cost sensor array that can be deployed in remote areas and used to support studies of soil degradation or progress toward restoration worldwide.

Planned Impact

The work proposed here focuses on an issue of critical global need. The global decline in soil health and the extensive degradation of soils in managed ecosystems has been identified as a major issue by the UN Food and Agriculture Organisation, the Intergovernmental Platform on Biodiversity and Ecosystem Services, the UN Global Resource Panel and others. The work described here explores fundamental research and engineering challenges but does so with an eye toward the eventual use of our approach in early detection of both degradation and recovery of both managed and natural ecosystems around the world. Our proposed research, and sensor development, are designed with this long-term goal in mind, including our focus on transitioning from high cost, high power conventional sensors to low power (or passive) sensors linked to inexpensive networked field stations that have been used by PI Thomas (U of Colorado) to monitor groundwater pumps serving over 1.5 million people in East Africa. Our proposed use of the Swarm Satellite network for data exchange is also developed to support low cost, remote deployment of the sensor packages developed in this project. As our work progresses, we will explore different avenues to scale our technology through market or non-market mechanisms and our team overall has extensive experience in the transfer of technologies into private or non-profit sector application. In the UK, The Government has, through its 25 Year Environment plan, committed itself to improving soil health, and in particular addressing the need to ensure healthier soils by addressing soil degradation and the factors responsible for it. These factors include soil erosion, soil compaction and the decline in accelerated loss of organic matter, all of which are central to this proposal. A key pillar of the UK government's approach is to develop better information on soil health and the work proposed here plays strongly to the UK government's stated desire for research that provides a clearer picture of soil health. In addition, the plan sets out a desire to develop 'cost-effective and innovative ways to monitor soil at farm level' and to utilise technological developments that have the potential to 'revolutionise how we monitor the environment', both areas are clearly aligned to the sensing developments we propose here. Developing new approaches to machine learning that can be applied in multiple sectors has clear links to the UK's Industrial Strategy and particularly the desire to putting the UK at the forefront of the artificial intelligence and data revolution. The US research site is located on lands managed by the City of Boulder Open Space Department to restore the landscape to one that can support agricultural production. The City of Boulder is strongly supportive of this work and would like to use the data generated in this study to communicate the nature and results of the site restoration efforts to the public. To this end, we will develop a public-facing website (including a UK section, see below) that shows the live data flow from the site along with an explanation of what the data illustrates about changing soil function at the site. This site will be developed with the input of city staff and will include broader information on the causes and consequences of soil degradation at the site and the remediation work underway to address those issues. We will build on this experience and our networks to facilitate knowledge exchange between the project team, end users and policy makers. The project website will provide a valuable tool to both demonstrate the advances in soil health monitoring developed in this project and as a platform for communicating issues pertaining to the sustainable soil management, with a focus on grasslands in the UK context and degraded agricultural lands in the US.

Publications

10 25 50