HuHoLa: A novel Hummock-Hollow-Lawn mire microtopography modelling approach

Peatlands play a crucial role in global carbon storage and methane emissions. Microtopographic features (hummocks, hollows, and lawns) are one of their main characteristics. These features strongly influence hydrological and biogeochemical processes, affecting vegetation patterns and greenhouse gas exchanges. Traditional methods for mapping peatland microtopography often rely on complex algorithms or require extensive field data, typically producing only a binary hummock-hollow classification. These shortcomings limit their applicability for large-scale studies. To address these challenges, we developed and validated HuHoLa (Hummock-Hollow-Lawn), an easy applicable and scalable model that classifies peatland microtopography using only a digital elevation model (DEM). HuHoLa applies a sink-filling approach to generate classifications, with a key feature being a threshold value to better capture the subtle variations in the non-flat lawn features. In addition to the microtopographic classification, the model provides a secondary output that acts as a proxy for water table depth (WTD) and soil temperature (Ts), thus offering a useful tool for understanding spatial variations in WTD and Ts across peatland landscapes. HuHoLa delivers a more nuanced and realistic depiction of peatland surface structure compared to traditional binary methods, with field-based validation demonstrating robust performance (Kappa coefficients of 0.62 and 0.81 for DEM resolutions of 30 cm and 50 cm, respectively) that outperforms traditional binary classification approaches (Kappa < 0.5 for DEM resolutions between 10 and 25 cm). The model is particularly suited for large-scale research applications. Its simplicity, requiring only a DEM, combined with its multi-purpose use, makes it an effective tool for advancing peatland studies and integrating with land surface models.