Flores-Landeros et al. (2026) Low-cost CO₂ sensors reveal seasonal and management-driven soil carbon fluxes in a Mediterranean agroecosystem

Identification

• Journal: Environmental Technology & Innovation
• Year: 2026
• Date: 2026-01-02
• Authors: Humberto Flores-Landeros, Ana Grace Alvarado, Anna Jurusik, Jorge Andrés Morandé, Emily Waring, Thomas C. Harmon
• DOI: 10.1016/j.eti.2025.104743

Research Groups

• Environmental Systems Graduate Program, School of Engineering, University of California, Merced
• Department of Civil and Environmental Engineering, School of Engineering, University of California, Merced
• Sierra Nevada Research Institute, University of California, Merced
• Agricultural Experimental Station, University of California, Merced

Short Summary

This study utilized low-cost CO₂ sensors to assess seasonal and management-driven soil CO₂ efflux in a 5-hectare Mediterranean agroecosystem transitioning from flood-irrigated pasture to sprinkler-irrigated cropland, finding that soil moisture is the primary driver of efflux, significantly modulated by vegetation and temperature, with irrigation practices leading to the highest emissions.

Objective

• Principal hypothesis: Efflux regimes are primarily driven by seasonal changes in soil moisture, temperature, and vegetation, and land management activities will restructure these conditions to produce distinct efflux responses.
• Specific objectives:
  1. Provide a relatively long-term, field-scale dataset characterizing the spatiotemporal variation in soil CO₂ efflux in a changing landscape.
  2. Identify drivers of the spatial heterogeneity of soil CO₂ efflux.
  3. Assess how land management practices, climate, and weather combine to affect soil CO₂ emissions, with the aim of informing greenhouse gas (GHG) emissions monitoring efforts in natural and agricultural landscapes.

Study Configuration

• Spatial Scale: 5 hectares of agricultural land in Merced, California, with a 2-hectare study area for sampling. During the sprinkler irrigation phase, the farm was reorganized into 0.4-hectare plots.
• Temporal Scale: Sampling campaigns conducted from July 2022 through October 2024 (approximately 2 years and 3 months). Continuous monitoring stations were also installed.

Methodology and Data

• Models used:
  • Welch’s One-way ANOVA with Games-Howell post-hoc test
  • Unsupervised cluster analysis
  • Ordinary Least Squares (OLS) regression with heteroscedastic-consistent standard errors (HC0)
  • Simple kriging interpolation (for visualization)
• Data sources:
  • Low-cost portable CO₂ sensors (NDIR CO₂ sensor K30, MH-Z16) for soil CO₂ efflux.
  • TEROS 11 sensors and ProCheck handheld metering devices for soil temperature and volumetric soil moisture content.
  • Trimble R1 GNSS receiver for sampling locations.
  • Landsat 8–9 L2 dataset for Normalized Difference Vegetation Index (NDVI) at 30 m pixel resolution.
  • Continuous monitoring stations for soil moisture and temperature at 5 cm, 10 cm, and 25 cm depths.
  • ATMOS 41 All-In-One weather station for air temperature and other meteorological data.
  • National Weather Service for precipitation data.

Main Results

• Soil CO₂ efflux ranged from -0.64 to 15.33 µmol m⁻² s⁻¹, with an overall average of 2.44 ± 2.77 µmol m⁻² s⁻¹.
• Water-related management practices exhibited the highest mean CO₂ efflux: Flood irrigation (5.06 ± 3.57 µmol m⁻² s⁻¹) and Sprinkler irrigation (3.46 ± 3.41 µmol m⁻² s⁻¹), which were approximately 2–4 times greater than non-irrigated practices (Pasture: 1.22 ± 1.88 µmol m⁻² s⁻¹; Disking: 1.06 ± 1.14 µmol m⁻² s⁻¹; Cultivation: 2.00 ± 1.90 µmol m⁻² s⁻¹).
• Ordinary Least Squares regression identified soil moisture as the strongest positive predictor of CO₂ efflux (β = 5.91, p < 0.001), followed by vegetation cover (NDVI) (β = 2.92, p < 0.001), and soil temperature (β = 0.04, p = 0.043).
• Unsupervised cluster analysis revealed three distinct seasonal efflux regimes: (1) Winter (primarily Cultivation), (2) Summer (primarily Flood), and (3) Fall (including Pasture, Disking, and most Sprinkler observations).
• The OLS model explained 33.6% of the variability in CO₂ efflux (R² = 0.336) but tended to underpredict higher efflux values (above 6 µmol m⁻² s⁻¹), which were predominantly observed during flood irrigation events.
• Highest efflux values (≥10 µmol m⁻² s⁻¹) during flood events were observed in areas with moderate soil moisture and elevated, but not peak, soil temperatures, suggesting complex non-linear interactions and "hot moments."

Contributions

• Provided a unique, relatively long-term (over 2 years), field-scale dataset characterizing the spatiotemporal variation in soil CO₂ efflux within a Mediterranean agroecosystem undergoing significant land-use transitions.
• Demonstrated the efficacy of low-cost portable CO₂ sensors for robust monitoring of soil carbon dynamics in managed agricultural environments.
• Quantified the relative influence of key environmental drivers (soil moisture, temperature, vegetation) and specific land management practices (flood irrigation, sprinkler irrigation, pasture grazing, disking, winter forage cultivation) on soil CO₂ emissions.
• Advanced the understanding of the complex interplay between seasonal environmental conditions and management activities in shaping carbon cycling within Mediterranean agroecosystems.
• Offered valuable insights for developing more effective monitoring and management strategies for agricultural CO₂ emissions, emphasizing the critical need to integrate seasonal environmental context into emissions assessments.

Funding

• California Department of Conservation Multibenefit Land Repurposing Program Block Grants [Subbasins Kaweah 5–022.11 and Tule 5–022.13]
• University of California Office of the President Multicampus Research Program Initiative Labor & Automation in California Agriculture (Award M21PR3417)
• Sustainable Agricultural Systems project (Award 2021–69012–35916) from the U.S. Department of Agriculture’s National Institute of Food and Agriculture

Citation

@article{FloresLanderos2026Lowcost,
  author = {Flores-Landeros, Humberto and Alvarado, Ana Grace and Jurusik, Anna and Morandé, Jorge Andrés and Waring, Emily and Harmon, Thomas C.},
  title = {Low-cost CO₂ sensors reveal seasonal and management-driven soil carbon fluxes in a Mediterranean agroecosystem},
  journal = {Environmental Technology & Innovation},
  year = {2026},
  doi = {10.1016/j.eti.2025.104743},
  url = {https://doi.org/10.1016/j.eti.2025.104743}
}