Driven grouse moors - socioeconomic and biodiversity impacts: summary report

8. Part 4- Biodiversity considerations on grouse moors

Part 4 of the Phase 2 research examined the biodiversity impacts of driven grouse moors using species distribution data for selected moorland biodiversity indicator species. This work utilised estimates of moorland management intensity for driven grouse developed in Part 3 of the Phase 2 research - GIS mapping of the management and intensity of grouse moors.)

Grouse moor management comprises of a range of management practices, including predator control, muirburn, grazing management and disease management. The Phase 1 research (Brooker et al., 2018b) amongst others (Thompson et al., 2016; Mustin et al., 2018), reported that grouse moor management has been demonstrated to have positive and negative effects on the distribution and abundance of different species and biodiversity. Predator control, the legal killing of crows, foxes, stoats and weasels undertaken as part of grouse moor management to minimise predation of red grouse, has been shown to benefit other ground nesting birds and is thought to benefit mountain hares. Whilst predator control will suppress the local population of controlled species, the Phase 1 research highlighted that the wider biodiversity impacts of predator control on the controlled species are poorly understood.

Overall, the effects of grouse moor management practices vary with habitat (e.g. wet or dry heath), species and management type, and in many cases the evidence base is not conclusive on whether specific practices have positive or negative biodiversity effects (Thompson et al., 2016; Brooker et al., 2018b; Mustin et al., 2018; GMMG, 2019). While there is a clear evidence base that grouse moor management can positively and/or negatively affect different species of wading birds, raptors, and vegetation communities, the evidence remains inconclusive for many other taxonomic groups and species (Brooker et al., 2018b; Mustin et al., 2018). Indeed, the effects of grouse moor management on the distribution and abundance of the majority of species has not been investigated. This section provides a summary of the full technical report for Part 4.

Method

Within Part 4 of the Phase 2 research, the effect of grouse moor management, based on the intensity of muirburn (the estimated percentage of ground burnt), on the distribution of selected upland species was assessed. The species used in this study were chosen through consultation with the project Research Advisory Group and Scottish Government to reflect a small selection of species that are likely to be negatively or positively affected by grouse moor management, and for which there was suitable occurrence data available for analyses within the time frame and resources available. Some obvious species of interest such as mountain hare Lepus timidus, red deer Cervus elaphus, and high conservation priority species such as lapwing Vanellus, were not included in the final list of species assessed because there is already a substantial body of evidence indicating that these species benefit from and are positively associated with moorland managed for grouse shooting (Fletcher at al., 2010; Patton et al., 2010; Newey et al., 2016; Mustin et al., 2018; Littlewood et al., 2019). Rather, the aim of this work was to assess the effects of the intensity of grouse moor management on species where the association between species distribution and grouse moor management is less well understood or unknown. The species selected for review were:

• Birch
• Green hairstreak butterfly
• Curlew
• Merlin
• Lesser redpol
• Bilberry / blaeberry
• Adder
• Golden plover
• Kestrel
• Whinchat

Using outputs from Part 3 of this research (Matthews et al., 2020) estimates of management intensity were based on the extent of muirburn at the 1 km square scale. To estimate management intensity, the number of 200 m x 200 m cells within each 1 km square (25 accessed squares) that were classified as at least 50% burnt were summed and converted to a percentage. This estimate of percentage muirburn assumed that each cell classified as burnt was 100% burnt, which is not necessarily true since the percentage of burn for each cell will range from 51-100% and is therefore an overestimate of actual muirburn. However, it was considered that this likely provided a good estimate of the intensity of management and area of land under grouse moor management.

To assess the distribution of the chosen species in relation to muirburn intensity, the species distribution and muirburn intensity data were overlaid. For those biodiversity species where the distribution data was only available at the 10 x 10 km² (hectad) scale, the percentage burn at the 1 km square level was calculated and the median value of the 1 km squares within that 100 km square was used to represent the overall level of muirburn. Twelve percent of 1 km squares assessed were classed as been less than 5% burnt, and 60% of squares were classified as less than 41% burnt.

With all species data care is needed in interpreting the relationship between species occurrence and the high levels of muirburn, as the sample size of both the number of assessed squares within each burn category, and the number of species records are low for these high intensity burn categories. In addition, it must be noted that assessment was restricted to the area for which muirburn data was available and that this was largely from areas where grouse moor management was known to be an important land use. The restricted area also had the consequence of reducing the area of intersection between areas assessed for muirburn and species occurrence data.

Biodiversity considerations - key findings

Using aerial photography or satellite imagery for 3,616 1km squares classified as burnt, and the approach to estimating burning intensity outlined above, it was estimated that the proportion of the area classed in different muirburn intensities were:

Muirburn intensity

• less than 5% burnt
• 6-20% burnt
• 21-40% burnt
• 41-60% burnt
• 61-80% burnt
• 81-100% burnt

Proportion of assessed area

• 12% of the area
• 24% of the area
• 24% of the area
• 18% of the area
• 13% of the area
• 11% of the area

The key results from the Part 4 assessment of biodiversity results included:

• Birch and blaeberry were most prevalent in areas with little or intermediate levels of burning and showed a decline with increasing burning but were also present in squares with high levels of muirburn.
• Green hairstreak butterfly and adder were both most prevalent at low to moderate levels of burning and showed a general decline in prevalence with very high levels of burning. However, the pattern in change in prevalence with increasing burning is not clear. For these 'semi-cryptic' species it was not clear whether apparent greater prevalence in intensely burnt areas reflects greater detectability in these areas or greater abundances.
• Curlew and golden plover prevalence generally increased with intensity of muirburn, though golden plover occurrence peaked in the 41-60% burn category, whereas curlew increased with greater percentage muirburn. This was particularly the case for these, and the other bird species assessed at the hectad (10 x 10 km) scale where sample sizes for squares representing intense muirburn were very small.
• Merlin prevalence increased with increasing intensity of muirburn up to the 41-60% muirburn, and then declined and was absent from the squares with 81-100% burning, whereas kestrel was present at a consistent level across all muirburn categories up to 81%. Interpretation of prevalence at the 81% plus muirburn category is likely confounded by small sample size.
• Both lesser redpoll and whinchat showed consistent levels of prevalence at low to moderate levels of muirburn and showed increases in prevalence in the 61% and higher muirburn categories. Lesser redpoll prevalence peaked in the 61-80% burn category and the species was absent in the 81-100% category, while whinchat was most prevalent in the 81-100% category.

Conclusions - biodiversity

The occurrence of ten species was assessed in relation to intensity of muirburn in areas of Scotland where grouse moor management is an important land use. Overall, it proved challenging to identify clear patterns in the occurrence of these species relative to intensity of muirburn.

It is difficult to draw any firm conclusions and for all species care is needed in interpreting the relationship between species occurrence and the high levels of muirburn as the sample size of both the number of assessed squares within each burn category, and the number of species records are low for these high intensity burn categories. Species may be responding to aspects of moorland management other than muirburn and for the bird species occurrence was likely influenced by the wider landscape.

Birch was the only species assessed where prevalence appeared to decline with increasing intensity of muirburn, though blaeberry also showed evidence of lower prevalence at the highest category of muirburn. Green hairstreak butterfly, adder, and kestrel showed fairly consistent occurrence across the range of muirburn measured. Golden plover and merlin showed an increased occurrence with greater burning, occurrence for these species peaked at intermediate levels of muirburn. Curlew, whinchat and lesser redpoll appeared to increase in prevalence with increasing percentage of ground classed as burnt.