Adding Farm Structure to Land Use Change

The purpose of this storymap is to exemplify the use of farm structure data in designing land use change scenarios. The inclusion of farm structure data provides the means to explore more complex and realistic land use change scenarios for Scotland by considering socio-economic data on size & types of existing land holdings. As an example, this storymap explores the case of the “specialist dairying” farm type, both in the original SSP1-low emission scenario and in a new alternative version, which partly protect their land-based activities.

This work builds on a SSP1-low emission land use scenario for Scotland as presented in the following  land use transformations StoryMap . In this initial scenario, target land use changes were applied across the whole of Scotland and were shaped by biophysical suitability rules (both where change could not occur and where it was most desirable). Four ecosystem services were considered: Carbon storage through tree planting, emissions reduction through deintensification of livestock production, biodiversity enhancement through tree planting, and enhancing pollination services to support food production.  There was though no spatial consideration of socio-economic factors or entities. 

Farm structure data at field, holding and business level have previously been used in projects by the  Land Systems Research team  supporting SG policy options and impacts assessments.  These data combine data from the Single Application Form (SAF), part of the agricultural payments system (IACS), linked to mapping of land parcels from the Land Parcel Identification Systems (LPIS) and to data from the June Agricultural Census. The farm structure data can be used to generate maps compatible with the land use change modelling at field (LPIS), holding (for JAC) or business level (for IACS). 

For the first use of farm structure data in land use change modelling Farm Types were used.  Farm types use Standard Output (SO) coefficients for livestock and crops to determine the value of outputs per holding or business. The holding or business is then allocated to a type according to the source of the majority of its total SO. The coefficients vary over time, so averages are taken across a five-year period.  The farm types are based on data at holding level for 2019, they are grouped into 13 classes as used by Hutton analysts for the policy options analyses (n.b. some land have multiple users).  

The overall coverage (68% of all land) and area per farm types is shown below (pie chart: percentages, bar chart : hectares).  These were used to assess the impact of the  original StoryMap scenario  in terms of the degree of change from the baseline pattern of land use. 

“Specialist dairying” farm type, despite their limited surface area (just above 2% of Scotland), were heavily impacted by the original SSP-1 scenario; their intensive improved grasslands decreased from 52% to 9% of their land.

The high impact of the SSP1-Low emission scenario on the “specialist dairying” farm type is the result of multiple cumulative factors:

1) 54% of their land were deemed appropriate for new woodlands and silvo-pastoral.

Recognising that such high levels of change might be an unrealistic burden for one sector (“specialist dairying”), the farm type data was used to create an alternative land use change scenario.

This new scenario was based on the original storymap scenario set up, but it is also excluding any land use change (including intensity reduction) on the intensively grazed improved grassland of “specialist dairying”. With this new constraint, the potential land use changes between farm types at Scotland scale is spread more evenly and reduces potential changes in "specialist dairying" farm types.

This new alternative SSP1-Low emission scenario takes out 74,704 ha of change that occurred on those lands in the SSP-1 Story Map scenario; of which 32,415 ha were going to new woodlands, 27,648 ha to silvo-pastoral and 14,641 for reduction in intensity.

As specified in the scenario set up, the 52% of improved grassland with intensive grazing is preserved in the new scenario (it had decreased to 9% in the original SSP1-low emission scenario).

For New-woodlands, 66% of the land use change displaced from specialist dairy, on intensive grassland (19 of 28k ha), was reported as moving to lower quality grasslands and on heathers with non-intensive grazing.  For Silvo-pastoral, 64% of the land use change (11k ha of 27k ha) was transferred to non-intensive improved grasslands.

In terms of farm types, the extra burden of land use change for new woodlands is taken up by “specialist sheep and goats” (20% of the extra land use change) and “graziers” (15%) with “specialist cattle -rearing and fattening” taking 28% of the displaced silvo-pastoral land use change. The displacement to the other farm types is below <1.5% of their respective areas for all, except for “specialist cattle -rearing and fattening” at 2.24%, “Mixed crops – livestock” at 1.87% and “Sheep and cattle combined” at 1.51% .

Preserving intensively grazed improved grassland of “specialist dairying” made the land use change scenario overall less effective in delivering the Ecosystem Services outcomes sought.  The scenario sought a reduction in grazing intensity for 70-80% of the grassland area and in the revised scenario the reduction achieved was only 72% compared with the 76% in the  Land Use Transformation Storymap , and quite importantly intensive improved grasslands were only reduced by 71% instead of 78%. This smaller decrease in intensive improved grazing is an outcome both of the direct preserving of the specialist dairy area and also the displacement of LUC to other land covers and farm types, within the context for the 4 Ecosystem Services and areas exclusions.

The SSP1-low emission revised scenario is less effective in decreasing Carbon emissions from the livestock sector (see bar chart below) the overall decreased is 39.2% of C eq. emissions instead of the 42.5%. The overall numbers of livestock are 1.4M livestock units (LU) rather than 1.36M LU against a baseline of 2.2M LU so the overall change is small.  The change in the mix of livestock types preserving more dairy livestock is thought clearly seen in the mix of emission sources with 602kT of CO2e emissions from dairy livestock rather than 314kT of CO2e in the standard SSP-1 scenario, against the baseline 681kT of CO2e in the baseline.  The greater numbers of dairy LU are offset by larger reductions in beef and sheep numbers.  The farm type levels restrictions are thus effectively shaping the land use change to minimise the impacts on part of the agricultural sector but at the price of increased burdens for other types.

Further analysis of the revised SSP-1 scenario is needed but given the pattern of displacement of new woodlands away from land with higher agricultural productivity it is likely the gains in Carbon storage will be reduced because of limits on tree growth potential, more losses of soil carbon in tree establishment and thus reduced net Carbon storage for the area planted.

Further improvements to the land use change model are planned to increase its capacity to handle more complex scenarios; like being able to breakdown the burden of land use change as percentage of land use change targets per farm type instead of simply blocking off some land uses of farm types.