William Lidberg

My research bridges cutting-edge technology with practical applications, supporting sustainable, multi-use forestry through advanced spatial analysis, machine learning, and environmental data science.

I work across hydrology, ecology, archaeology, and forestry, developing AI-driven methods to improve forest and water management decision-making. A significant focus of my research is predicting and mapping surface water dynamics in forested landscapes, which has led to successful collaborations with governmental agencies, industry partners, and international research teams. These efforts have contributed to implementing national soil moisture and wet area maps and improving land management strategies across multiple countries.

I have developed, am course responsible for, and currently teach a master’s-level course, Analysis of Environmental Data 2, which integrates machine learning, deep learning, and geospatial data (e.g., LiDAR and satellite data) to address environmental challenges.

My research spans hydrology, forestry, and ecology, developing AI-driven tools for mapping and predicting environmental and cultural features on a national scale. I have pioneered the use of machine learning and deep learning to enhance hydrological modeling, soil moisture mapping, and biodiversity assessments, resulting in widely adopted industry tools such as the SLU Soil Moisture Map and AI-detected drainage networks. My work bridges technology and policy, investigating how AI impacts decision-making and stakeholder trust in forestry. Through collaborations with governmental agencies, forestry companies, and international partners, I ensure that my research delivers practical, high-impact solutions. By integrating social science perspectives, I strive to develop ethical and transparent AI applications that balance environmental conservation with economic and societal needs.

I am an applied scientist committed to ensuring my research has a tangible impact on society. Through extensive collaborations with government agencies, forestry companies, and stakeholders, I bridge the gap between academia and practical applications. My long-standing partnership with the Swedish Forest Agency, which has funded part of my position, has led to widely adopted industry tools such as the SLU Soil Moisture Map, the SLU Ditch Map, and AI-detected Road Culverts. These innovations support sustainable forestry, hydrological planning, and wetland restoration. My work has influenced national agencies like SMHI, the Swedish Transport Administration, and the Swedish Mapping Authority, contributing to official stream networks, flood risk models, and water management strategies. I also engage with the public through lectures, media outreach, and documentary features, such as Vi måste prata om skogen. As the founder of an award-winning startup, I further translate research into real-world applications, helping organizations use AI and geospatial data for environmental planning and conservation.

I have a background in Geoecology and specialize in landscape-scale analysis, integrating remote sensing, machine learning, and geospatial technologies to map and predict environmental and cultural features. My research began with a PhD focused on digital wet area mapping, where I developed methods to integrate diverse data sources, leading to the creation of the SLU Soil Moisture Map, now a key resource for forestry and hydrology. During my postdoc, I expanded this work with the Swedish Forest Agency and collaborated with the Swedish Geological Survey to apply machine learning in soil mapping. Now, as an Assistant Professor of Soil Science, I lead an interdisciplinary lab advancing AI-driven hydrological modeling and geospatial analysis. My work includes developing industry-standard tools for mapping drainage ditches, stream networks, and wetland features, influencing forestry and environmental planning across Sweden and the Baltic region. I also lead a Wallenberg-funded project on AI’s societal impact in forestry, exploring its role in decision-making, stakeholder trust, and policy development.

Main supervisor of one PhD student and co-supervisor of four others and a postdoc, all scheduled to finish between 2026 and 2027. 

Supervision of Postdoc 

Francesco Zignol, Co-supervisor, dynamic wet area mapping

Supervision of students (PhD level)

Mariana Dos Santos Toledo Busarello, Main-supervisor, Preliminary PhD Thesis title, “Challenges and social consequences of artificial intelligence in Swedish forests”. Forest Ecology & Management, SLU, Sweden, Expected defense Spring 2026.

Joakim wising, Co-supervisor, Preliminary PhD Thesis title, “Challenges and social consequences of artificial intelligence in Swedish forests”. political science, Umeå University, Sweden, Expected defense Spring 2026.

Yiqi Lin, Co-supervisor, Preliminary PhD Thesis title, “Potential Application of Digital Soil Mapping to Accelerate Quaternary Deposit Mapping in Sweden”. Forest Ecology & Management, SLU, Sweden, Expected defense Spring 2026.

Olivia Anderson, Co-supervisor, Preliminary PhD Thesis title, “Interdisciplinary decision support for drained wetlands”. Forest Ecology & Management, SLU, Sweden, Expected defense Fall 2026.

Alejandro Gandara, Co-supervisor, Preliminary PhD Thesis title, “Trade-offs, upscaling, & connectivity of a blue-green network”. Forest Ecology & Management, SLU, Sweden, Expected defense fall 2027.

Lidberg, W. 2025. Deep learning-enhanced detection of road culverts in high-resolution digital elevation models: Improving stream network accuracy in Sweden. Journal of Hydrology: Regional Studies, 57, 2214-5818. https://doi.org/10.1016/j.ejrh.2024.102148

Virro, H., Kmoch, A., Lidberg, W., Muru, M., Tik, W., Desalew, C., Moges, M., & Uuemaa, E. 2025. Detection of Drainage Ditches from LiDAR DTM Using U-Net and Transfer Learning. Big Earth Data.

Myrstener, M., Greenberg, L.A., Lidberg, W., & Kuglerova, L. 2025. Riparian buffers can mitigate downstream effects of clearcutting on instream metabolic rates. Journal of Environmental Management. Accepted January 2025.

Busarello, M.D.S.T., Ågren, A., Westphal, F., & Lidberg, W. 2024. Automatic Detection of Ditches and Natural Streams from Digital Elevation Models Using Deep Learning. Computers and Geosciences. Accepted December 2024.

Lin, Y., Lidberg, W., Karlsson, C., Sohlenius, G., Westphal, F., Larson, J., & Ågren, A.M. 2024. Mapping soil parent materials in a previously glaciated landscape: Potential for a machine learning approach for detailed nationwide mapping. Geoderma Regional, 40, 2352-0094. https://doi.org/10.1016/j.geodrs.2024.e00905

Peacock, M., Futter, M.N., Lidberg, W., Lundblad, M., & Geranmayeh, P. 2024. An upscaling of methane emissions from Swedish flooded land. Environmental Science & Policy, 162, 103945. https://doi.org/10.1016/j.envsci.2024.103945

Wising, J., Sandström, C., & Lidberg, W. 2024. Forest owners’ perceptions of machine learning: Insights from Swedish forestry. Environmental Science & Policy, 162, 103945. https://doi.org/10.1016/j.envsci.2024.103945

Lidberg, W., Westphal, F., Brax, C., Sandström, C., & Östlund, L. 2024. Detection of Hunting Pits using Airborne Laser Scanning and Deep Learning. Journal of Field Archaeology, 1–11. https://doi.org/10.1080/00934690.2024.2364428

Bakx, T., Akselsson, C., Droste, N., Lidberg, W., & Trubins, R. 2024. Riparian buffer zones in production forests create unequal costs among forest owners. European Journal of Forest Research, 143, 1035–1046. https://doi.org/10.1007/s10342-024-01657-1

Ågren, A., Andersson, O., Lidberg, W., Öquist, M., & Hasselquist, E.M. 2024. Ditches show systematic impacts on soil and vegetation properties across the Swedish forest landscape. Forest Ecology and Management, 555. https://doi.org/10.1016/j.foreco.2024.121707

Peng, H., Nijp, J.J., Ratcliffe, J.L., Li, C., Hong, B., Lidberg, W., Zeng, M., Mauquoy, D., Bishop, K., & Nilsson, M.B. 2024. Climatic controls on the dynamic lateral expansion of northern peatlands and its potential implication for the ‘anomalous’ atmospheric CH4 rise since the mid-Holocene. Science of the Total Environment. https://doi.org/10.1016/j.scitotenv.2023.168450

Ehnvall, B., Ågren, A.M., Nilsson, M.B., Ratcliffe, J.L., Noumonvi, K.D., Peichl, M., Lidberg, W., Giesler, R., Mörth, C., & Öquist, M. 2023. Catchment characteristics control boreal mire nutrient regime and vegetation patterns over ~5000 years of landscape development. Science of the Total Environment, 895, 165132. https://doi.org/10.1016/j.scitotenv.2023.165132

Lupon, A., Gómez‐Gener, L., Fork, M.L., Laudon, H., Martí, E., Lidberg, W., & Sponseller, R.A. 2023. Groundwater‐stream connections shape the spatial pattern and rates of aquatic metabolism. Limnology and Oceanography Letters, 8, 350-358. https://doi.org/10.1002/lol2.10305

Larson, J., Lidberg, W., Ågren, A.M., & Laudon, H. 2022. Predicting soil moisture conditions across a heterogeneous boreal catchment using terrain indices. Hydrology and Earth System Sciences, 26, 4837-4851. https://doi.org/10.5194/hess-26-4837-2022

Laudon, H., Lidberg, W., Sponseller, R.A., Hasselquist, E.M., Westphal, F., Östlund, L., Sandström, C., Järveoja, J., Peichl, M., & Ågren, A.M. 2022. Emerging technology can guide ecosystem restoration for future water security. Hydrological Processes, 36. https://doi.org/10.1002/hyp.14729

Ågren, A.M., Larson, J., Paul, S.S., Laudon, H., & Lidberg, W. 2021. Use of multiple LIDAR-derived digital terrain indices and machine learning for high-resolution national-scale soil moisture mapping of the Swedish forest landscape. Geoderma, 404, 115280.

https://www.slu.se/en/departments/forest-ecology-management/forskning/main-research-areas/forest-landscape-biogeochemistry2/researchers/geographical-intelligence/

https://www.slu.se/institutioner/skogens-ekologi-skotsel/forskning2/dikeskartor/

https://www.slu.se/institutioner/skogens-ekologi-skotsel/forskning2/markfuktighetskartor/