Aarhus Universitets segl

No. 397: Phosphorus mapping of arable land and water areas in Denmark

Andersen, H. E. & Heckrath, G. (redaktører). 2020. Fosforkortlægning af dyrkningsjord og vandområder i Danmark. Aarhus Universitet, DCE – Nationalt Center for Miljø og Energi, 338 s. - Videnskabelig rapport nr. 397. http://dce2.au.dk/pub/SR397.pdf

Summary

The loss of phosphorus (P) to the aquatic environment has been estimated for different transport pathways based on comprehensive new data supplemented by existing data as well as the development of a number of models. A series of maps has been produced indicating risk areas for phosphorus loss to surface water across Denmark. In addition, an attempt has been made to map the phosphorus sensitivity of streams, lakes and marine areas.

Water erosion on agricultural land and sediment delivery to surface water have been mapped with a spatial resolution of 10 m using the WaTEM model. The model incorporates detailed landscape data and has been calibrated. Phosphorus erosion from fields to surface waters has been calculated by combining soil loss from farmland to water with spatially varying estimates of the phosphorus content in the eroded soil. At national scale, the annual phosphorus loss due to erosion is estimated at 56 t P year-1 with a 95% confidence interval of 53 – 58 t P year-1.

Leaching of dissolved phosphorus by matric flow to tile drains has been calculated at field scale using a modification of the Dutch PLEASE model. The model has been parameterised based on comprehensive laboratory analyses of phosphorus sorption characteristics in representative soils from Denmark. The model utilizes a newly established nationwide map of the soils’ phosphorus sorption capacity, for which a large number of archived and new soil samples have been analyzed. At national scale, the annual phosphorus loss by matric leaching is estimated at 59 t P year--1 with a 95% confidence interval of 23 – 94 t P year-1.

An indicator for the risk of phosphorus loss to tile drains by macropore transport has been mapped nationwide. First, based on measured particle mobilization in a wide range of soil samples, a pedotransfer function has been developed for estimating and mapping the potential for particle mobilization. Second, an existing model for predicting the occurrence of macropore flow in soils has been improved by including many new soil hydraulic measurements in model development and using more advanced hydrological modelling. The maps of the potential for particle mobilisation and the risk of macropore transport have then been combined for representing risk classes for phosphorus loss in connection with macropore transport. It has not been possible to quantify phosphorus loss by macropore transport at fine spatial scale. However, based on measurements of phosphorus loss in drainage water in numerous catchments and a map of the contribution of macropore flow to drain discharge, the total phosphorus loss by macropore transport in Denmark is estimated at 162 t P year-1 with a 95% confidence interval of 138 – 191 t P year-1.

Phosphorus mobilisation in organic lowland soils from 47 locations across Denmark was studied in comprehensive laboratory experiments. They showed that the empirical model of Forsmann and Kjærgaard (2014) describing the potential for phosphorus mobilisation as a function of the Fe:P ratio in soils cannot generally be applied to all organic lowland soils. Neither was it possible to explain phosphorus mobilization as a function of the degree of phosphorus saturation. Currently, a generally applicable model for predicting the potential of phosphorus mobilisation in organic lowland soils is lacking. Based on measured phosphorus loss from selected cultivated organic lowland soils and the extent of such soils in Denmark, a crude estimate of the total phosphorus loss from cultivated organic lowland soils was obtained amounting to 326 t P year-1, with an uncertainty interval of 69 – 515 t P year-1.

A model has been developed for large-scale mapping of stream bank erosion in Denmark based on two Danish data sets of measured stream bank erosion. By combining the modelled erosion rates with a large number of new measurements of the phosphorus content in stream banks, phosphorus loss is estimated at 644 t P year-1 with a 95% confidence interval of 422 – 1373 t P year-1.

Summing the estimated phosphorus losses in connection with water erosion, stream bank erosion, matric leaching and macropore transport in upland soils, leaching from organic lowland soils as well as the contributions from wind erosion, surface runoff and groundwater, the total diffuse phosphorus loss in Denmark amounts to 1327 t P year-1 with an uncertainty interval of 715 – 2261 t P year-1. The contribution from agricultural sources is estimated at 683 t P year-1 with an uncertainty interval of 292 – 888 t P year-1. In this context, stream bank erosion is regarded as part of the natural background contribution, although some of the eroded stream bank material is likely of agricultural origin. However, the proportion of which cannot be quantified at present.

A range of maps relating to the risk of phosphorus loss from land to water at fine spatial scale (field scale and finer) has been made available for local mitigation planning.

For rivers, lakes and marine areas, an attempt has been made to assess the sensitivity of these water bodies to phosphorus inputs. Here, phosphorus sensitivity should be understood as an assessment of how likely it is that the environmental quality of an aquatic area is significantly affected by changes in the input of phosphorus to this water body.

The composition of plant communities in streams reflects a range of natural conditions, but different types of anthropogenic influences play an at least equally important role. In order to be able to assess the extent to which inorganic phosphorus can be critical for meeting the objectives of the Danish Macrophyte Index (DVPI), a method is described based on plant traits that in the future will allow for a separation of the importance of inorganic phosphorus from other types of impact. In addition, an interpretation is provided of the newly developed benthic algae index, SID_TID. By including, among other things, measurements of phosphorus dynamics in streams, it is demonstrated that the identified critical concentrations of inorganic phosphorus cannot be interpreted as stringent threshold values for when SID_TID objectives are met.

Based on data from 1213 ponds and 146 lakes > 1 ha, it has been examined whether the state of the lakes (natural or ecological condition) could be related to the environmental conditions of their catchment area. Assuming a strong relationship, it may in principle be possible to estimate the state of a given lake on the basis of catchment characteristics. The analyses showed that there were several significant correlations between the natural state index and/or its applied parameters and catchment characteristics. However, in all cases, the explanatory values (< 10%) are low. This means that the use of such catchment data will be a very uncertain method for estimating the conditions of non-examined lakes. It is therefore currently not possible on the basis of these data and analyses, to develop a model that can be  used to estimate the state of lakes with unknown conditions from information on properties of the lake catchments.

The mapping of the phosphorus sensitivity of marine areas was based on six indicators. These indicators form the basis for a categorisation of the phosphorus sensitivity of marine areas as scaled from the variation in the phosphorus sensitivity of Danish marine areas. The results of this mapping can support a qualitative assessment of the expected response of the individual marine areas to changes in phosphorus inputs relative to other Danish marine areas. That a marine area is estimated to have a high level of phosphorus sensitivity based on one or more indicators does not preclude the area from being affected by other factors too. For example, a marine area with high phosphorus limitation can also be sensitive to changes in the nitrogen input. Marine areas exhibiting high levels of phosphorus sensitivity are often characterised by a mainly phosphorus-limited algae growth, and the marine area being influenced by phosphorus input from its catchment area. Similarly, a marine area estimated to have “least” or “low” phosphorus sensitivity can also change environmental state in response to changes in phosphorus inputs. However, major or long-term changes in the phosphorus input would be required to induce a shift in environmental state as either local sources have no significant effect or accumulation of phosphorus in the sediment over time has been so high that many years of low loading are required for a system change to take place.