Upscaling methane fluxes from peatlands across a drainage gradient in Ireland using PlanetScope imagery and machine learning tools

Abstract

Wetlands are one of the major contributors of methane (CH4) emissions to the atmosphere and the intensity of emissions is driven by local environmental variables and spatial heterogeneity. Peatlands are a major wetland class and there are numerous studies that provide estimates of methane emissions at chamber or eddy covariance scales, but these are not often aggregated to the site/ecosystem scale. This study provides a robust approach to map dominant vegetation communities and to use these areas to upscale methane fluxes from chamber to site scale using a simple weighted-area approach. The proposed methodology was tested at three peatlands in Ireland over a duration of 2 years. The annual vegetation maps showed an accuracy ranging from 83 to 99% for near-natural to degraded sites respectively. The upscaled fluxes were highest (2.25 and 3.80 gC m−2 y−1) at the near-natural site and the rehabilitation (0.17 and 0.31 gC m−2 y−1), degraded (0.15 and 0.27 gC m−2 y−1) site emissions were close to net-zero throughout the study duration. Overall, the easy to implement methodology proposed in this study can be applied across various landuse types to assess the impact of peatland rehabilitation on methane emissions by mapping ecological change.

Species-Level Classification of Peatland Vegetation Using Ultra-High-Resolution UAV Imagery

“…At present, there is a lack of published literature comparing the two approaches for species-level classifications in peatlands, or even similar (e.g., wetland, grassland) ecosystems. We note, however, that machine-learning studies do not always report accuracies higher than we achieved at Auchencorth Moss using a Maximum Likelihood classifier (e.g., [22,41,62,64,65]). While a comparison of the performance of the two approaches was outside the scope of the present study, such work would be of great value in guiding the design of future classification studies.…”

“…Advancements in sensor technology have made monitoring possible at increasing spectral and spatial resolutions. Satellites such as QuickBird, IKONOS, and PlanetScope, for example, can view peatlands at spatial scales in the order of metres or less [20][21][22]. However, the frequent presence of cloud cover in boreal and temperate regions (where many peatlands are located) makes it difficult to conduct frequent vegetation mapping based on optical satellite data (e.g., [23,24]).…”

Assessing and Improving the Quality of IoT Sensor Data in Environmental Monitoring Networks: A Focus on Peatlands

“…These methods encompass advanced sensor calibration techniques, robust outlier detection and data validation algorithms, machine learning approaches, and effective data fusion strategies tailored to the unique characteristics of peatland ecosystems. Through a long term and continuous sensor deployment on a real-world monitoring sites as part of the Smartbog project [1]- [3], LCS are collocated with state of the art monitoring facilities including Eddy-covariance flux tower (EC-flux), chambers, cameras and satellite imagery on multiple peatland sites in Ireland [4]. The Smartbog project sites provides real world testbed to evaluate and validate the tools and techniques developed to support the integration and calibration of LCS in peatland monitoring networks.…”

Detecting spatial patterns of peatland greenhouse gas sinks and sources with geospatial environmental and remote sensing data

“…As another example, Li & Sun (2021) demonstrated the potential for improved CO 2 emission prediction through several geospatial environmental data (e.g., digital elevation model (DEM), annual average precipitation, and air temperature). Efforts have been directed toward assessing and estimating the GHG distribution patterns over various spatial scales, from peatland-level investigations (Couwenberg et al, 2011;Heiskanen et al, 2021) to landscape-level studies (Ingle et al, 2023;Kou et al, 2022), and global GHG data syntheses (Qiu et al, 2020b;Zhao et al, 2023).…”