Catchment Management Modelling Platform

Case Study 2

Effectiveness of land management policies and agri-environment interventions for reducing pollutant loads and maintaining environmental quality at the national scale

Participants

Richard Gooday1; Adrian Collins2; Peter Daldorph3.

1RSK ADAS Ltd; 2Rothamsted Research; 3Atkins Global.

Stakeholder Representatives

Kirsten Foot, Neil Murdoch; Environment Agency; Murray Hart, Stuart Kirk, Defra.

Forum issues addressed by case study

How can I know if land management will be effective?

  • What is the effect of different land management interventions on water quality?
  • Will what I’m planning on doing be cost-effective?
  • Can I transfer these measures or approaches to other catchments? Will they be less or more effective?

Where is the pollution coming from? And how do pollutants act together?

  • Are there different sources of the same pollutants?
  • Would I make a different decision if I looked at one or several pollutants?
What did we find?

The impacts of the suite of Countryside Stewardship options were reductions in national agricultural pollution of over 6% (nitrate) and 10% (phosphorus). Accounting for non-agricultural sources of pollution reduced the net overall impact of the options to 4% (nitrate) and 3% (phosphorus). There would also be a net reduction in GHG emissions from agriculture.

ca1map1.png

How did we do it? (Synopsis)

The Farmscoper tool (Gooday et al., 2014) calculates the costs and impacts of mitigation measure implementation on multiple pollutants. The tool is pre-populated with crop areas and livestock numbers for the 92 Water Management Catchments (WMCs) in England from the 2010 agricultural survey data.

Fifteen measures from the Farmscoper measure library were identified as comparable to options in the Countryside Stewardship scheme which were likely to have high uptake. The modelling calculated the costs and impacts of these measures, assuming 100% implementation of all the measures on all appropriate land across England.

The SEPARATE database (Zhang et al., 2014) contains annual average nitrate, phosphorus and sediment data at Water Framework Directive (WFD) waterbody scale for the major sectors including agriculture, bank erosion, urban, sewage and sceptic tanks. By combing this data with the changes in agricultural pollution predicted Farmscoper, the overall impact of the agricultural changes was determined.

How did we do it? (Full)

The Farmscoper tool (Gooday et al., 2014) determines the pollutant loads at farm or catchment scale, and then the costs and impacts of current and future scenarios of mitigation measure implementation. The pollutant loads are the long-term annual average values and are calculated from a series export coefficients derived from existing models used for policy support, with the losses expressed in terms of a detailed source apportionment system. Inputs to Farmscoper include crop areas, fertiliser rates, livestock numbers and manure management – the tool is pre-populated with data for the 92 Water Management Catchments (WMCs) in England using the 2010 agricultural survey data and British Survey of Fertiliser Practice data. WMC scale outputs from Farmscoper were downscaled to Water Framework Directive (WFD) waterbody scale using agricultural areas within WFD waterbodies taken from CEH Land Cover Map 2007.

Farmscoper contains a library of over 100 mitigation measures, with each measures described in terms of cost of implementation, effectiveness at reducing the pollution (expressed as a percentage reduction for losses from the different components of the source apportionment system) and an estimate of current implementation. A total of 15 measures from this library were identified as comparable to options available under the Countryside Stewardship scheme which were likely to have a high level of uptake. The modelling calculated the costs and impacts of these measures, assuming 100% implementation of all 15 measures on all appropriate land across England. The measures were: cover crops, management of over-winter tramlines, in-field grass buffer strips, riparian buffer strips, use of clover in place of fertiliser nitrogen, reduced field stocking rates when soils are wet, installation of covers to slurry stores, minimising the volume of dirty water, fencing off rivers and streams from livestock, construction of bridges for livestock crossing rivers/streams, re-siting gateways away from high-risk areas, farm track management, creation of new hedges, establishment of artificial wetlands for steading runoff and tree shelter belts around livestock housing.

In order to assess the uncertainty in the results predicted by Farmscoper, the values for both the current implementation and the effectiveness of measures were set to their highest and lowest values within the ranges from the scoring system used within Farmscoper and the modelling then repeated. The uncertainty thus represents a big gap in implementation to be closed with measures more effective and a small gap in implementation with measures less effective. As all values were assumed to be at their highest or lowest at the same time, the values predicted are the absolute bounds of the uncertainty.

The SEPARATE database (Zhang et al., 2014) contains annual average nitrate, phosphorus and sediment data at WFD waterbody scale for the following sectors: agriculture, bank erosion, urban diffuse, sewage treatment works, storm tanks, sceptic tanks, combined sewer overflows, direct deposition and groundwater. The loads were derived from modelled and monitored data as appropriate for the sector. By combing the results of changes in agricultural pollution with the loads from all sectors, the overall impact of the agricultural changes can be determined.

What did this case study show? (Synopsis)

Calculated reductions in national annual average agricultural pollutant loads due to the selected Countryside Stewardship were 5%, 10% and 16% for nitrate, phosphorus and sediment respectively, with significant spatial variation. Greatest reductions in nitrate were in arable areas, whilst for phosphorus and sediment they were in areas where losses are mostly in surface runoff and so disruption of the surface pathway is effective. National agricultural emissions of nitrous oxide and carbon dioxide from energy use were calculated to be reduced by 2% and 10% respectively, whilst ammonia and methane emissions were calculated to increase by up to 1%.

Non-agricultural sources of pollution reduce the overall national reduction to 4% for nitrate and 3% for phosphorus. Although the impact on phosphorus in some waterbodies remains above 20%, there are many where, despite a large impact on the agricultural load, the net effect is small due to sizeable non-agricultural contributions.

 

ca1map2.jpg

 

ca1map3.jpg

 

ca1map4.jpg

 

ca1map5.jpg

What did this case study show? (Full)

The calculated reductions in national annual average agricultural pollutant loads due to 100% implementation of the suite of 15 mitigation measures chosen to represent some of the most common Countryside Stewardship options were 5%, 10% and 16% for nitrate, phosphorus and sediment respectively. There is significant spatial variation in the impacts, with reductions in some waterbodies over twice the national average and reductions under 1% in others. Highest percentage reductions in nitrate were found in arable areas where cover crops, buffer strips and tramline management would be effective. Highest percentage reductions in phosphorus and sediment were in areas where losses are mostly in surface runoff (as opposed to drain flow) and so disruption of the surface pathway will have a high impact on the overall loss. The impacts of the measures on national agricultural emissions of nitrous oxide and carbon dioxide from energy use were reductions of 2% and 10% respectively. National emissions of ammonia and methane are calculated to increase by up to 1%. The total cost of measure implementation is almost £300m per annum, which equates to just over £30 per hectare of agricultural land. The bounds placed upon these national estimates due to uncertainty in both measure effectiveness and current implementation are reductions in nitrate between 2% and 13% (compared with an average of 5%) between 4% and 17% for phosphorus and 8% and 25% for sediment (see Table). The upper bound for ammonia is a reduction of 5% in emissions compared with a 1% increase for the average situation.

Accounting for non-agricultural sources of pollution (using the data from SEPARATE) reduces the net overall impact of the mitigation measures on national loads to 4% for nitrate and 3% for phosphorus. The larger reduction in net impact for phosphorus is because agriculture only accounts for 30% of the national phosphorus load, compared with 70% for nitrate. Although the impact of the mitigation measures on phosphorus loads in some waterbodies remains above 20%, there are many waterbodies where despite the impact on the agricultural load being large, the measures are not effective overall due to sizeable non-agricultural contributions, particularly around urban areas.

The results show that there is significant spatial variability in the effectiveness of the mitigation measures at reducing pollution, and thus that targeting of Countryside Stewardship options would be an effective strategy to ensure value for money for the scheme.

The maps below show impacts of mitigation measure implementation on annual average agricultural loads and total loads from all sectors for nitrate, phosphorus and sediment, summarised by WFD waterbody.

ca1map2.jpg

 

ca1map3.jpg

 

ca1map4.jpg

 

ca1map5.jpg

 

ca1map6.jpg

 

ca1map7.jpg

 

The table below shows impacts of mitigation measure implementation on annual average agricultural pollutant loads for England, plus annual costs of implementation. Results are shown for the average scenario and the upper and lower uncertainty bounds.

ca1table1.png

What were the benefits of using more than one model?

Without accounting for the contribution to pollutant loads from all sectors using data from SEPARATE, the impacts of changes in farm practice associated with countryside stewardship on pollution delivered to watercourses could not be fully quantified.

What were the lessons learned about how to apply the models?

Identifying key outputs was important to make these relevant to the operators and regulators.

What pre- and post-processing was done on the input and output data?

The WMC scale outputs from Farmscoper were converted into loss coefficients per hectare for the different climates and soils within a catchment. To generate WFD waterbody losses to combine with SEPARATE, the appropriate WMC scale coefficients could then be multiplied by the areas within each WFD waterbody.

W​hat datasets were used in the case study?

June Agricultural Census 2010

CEH Land cover map 2007

Videos
CaMMP_Case_Study_2_-_Project_view-0.jpg
Project view
Stakeholder view
Location

Whole of England

Issues
  • Management outcome
  • Cost effectiveness
  • Transferability
  • Pollution source
Pollutants
  • Nitrate
  • Phosphorus
  • Sediment
  • Ammonia
  • Methane
  • Nitrous Oxide
  • Carbon dioxide from energy use
Scale
  • Catchment, national
 Models
  • Farmscoper
  • SEPARATE
CaMMP

Improved access to and integration among data and models to address key questions in catchment management for water quality and wider ecosystem services, providing a more holistic view to inform scientific understanding and policy development.

LEARN MORE

Featured video
CaMMP_Community_Forum-0.jpg