Using a digital elevation model to place overland flow fields and uncleaned ditch sections for water protection in peatland forest management

Abstract

Peatland forest management is an important source of human-induced additional nitrogen, phosphorus, and organic carbon loads to surface waters in Finland. Ditch network maintenance (DNM) is typically used to ensure sufficient soil drainage for profitable timber production on peatland sites, but nutrient mobilization, particularly of phosphorus, and suspended sediment loads from ditches can cause harmful ecological effects on the quality of water courses. Various water protection structures have been developed to capture the loads released after treatments and to decrease the management impacts in forestry. Overland flow fields are efficient mechanisms for water purification, but they are rarely used in practice due to difficulties in finding suitable locations. In addition, leaving breaks between cleaned ditch sections is recommended wherever possible, as uncleaned sections decrease flow velocity and enhance sediment deposition from runoff water. To plan the most adequate protection methods in DNM, new GIS-based tools are needed to propose e.g. appropriate locations for water protection structures. We developed a spatial analysis to suggest locations for overland flow fields and uncleaned ditch sections within the catchment of lake Kovesjärvi (Parkano, Western Finland). Airborne lidar data consisting of 5 pulses m−2 enabled detailed terrain analysis and finding strictly specified terrain properties that satisfied the conditions of the water protection structures. Half of the field-reviewed overland flow fields were evaluated as good suggestions by forest management professionals and researchers. The model for uncleaned ditch sections worked well on the relatively flat study area, as 77% of the field-reviewed suggestions were rated good by expert opinions. Overall, the approach was evaluated to be useful in peatland forest management. Model parameters can be modified for various topographies, but more studies are needed prior to wider use of the model.