Grazing by wild red deer: Management options for the conservation of semi‐natural open habitats

Abstract

1.Maintaining semi-natural open habitats requires biomass removal, which can be achieved by extensive grazing with livestock species. However, implementing this established conservation management strategy in large or access-restricted areas is often not possible. 2.We investigated grazing by wild and free-ranging red deer Cervus elaphus as an alternative conservation management approach on an active military training area in Germany. In grasslands and heathlands protected under the EU Habitats Directive, we quantified above-ground net primary productivity, forage quality and forage removal by red deer over three successive years. To assess synergistic effects between wildlife grazing and additional grassland management measures removing vegetation biomass (i.e. burning, mowing), we conducted our grazing experiment in burnt, mown and untreated grasslands.3. Annual forage removal by red deer amounted to 35%, 44%, 48% and 59% of the above-ground net primary productivity in burnt, mown and untreated grasslands, and untreated heathlands, respectively. Theoretically, a similar annual biomass removal could be obtained by livestock grazing with 0.54 animal units ha −1 in grasslands and 0.45 animal units ha −1 in heathlands. In grasslands, daily rates of forage removal peaked in spring and early summer, whereas in heathlands, forage removal rates were only significant in winter. 4.Forage removal rates in grasslands increased with productivity and forage quality, which were both enhanced in mown grasslands. This suggests mowing can increase grassland attractiveness to red deer. Forage removal rates in heathlands did not relate to productivity or forage quality but to standing biomass. Synthesis and applications.We demonstrate that the quantity of forage removed by wild red deer can be comparable to that of livestock in common conservation grazing systems. The seasonal patterns of forage removal by red deer in grasslands and heathlands coincided with the different grazing requirements of these habitats. Especially in large areas of conservation interest, we therefore encourage attempts to modify current wildlife management strategies to allow red deer to forage in open landscapes, which can contribute to the conservation of seminatural open habitats and also help to reduce damage in forest stands.

Target‐oriented habitat and wildlife management: estimating forage quantity and quality of semi‐natural grasslands with Sentinel‐1 and Sentinel‐2 data

“…For this study, the forage quantity and quality dataset provided by the study of Riesch et al (2019) was used, who investigated the grazing effect of red deer on plot level in GTA covering a 3-year period between 2015 and 2017. This dataset was collected on lowland hay meadows (EU Habitats Directive Annex I habitat type 6510, 'grasslands') in April, May, June, August and October in each of the studied years.…”

“…The dataset by Riesch et al (2019) was collected in five different months over 3 years at five different locations. Each location comprised three plots on different management treatment areas.…”

“…Details on the respective sampling dates of the utilized dataset are shown in Table 1. For each plot and sampling date, Riesch et al (2019) determined standing biomass and forage quality values with the following methods: standing biomass was estimated by a double-sampling strategy using CSH measured by a rising plate metre in combination with calibration cuts, that is, above ground biomass cut at ground level on a 0.18 m² area.…”

“…The relationship between calibration cut biomass (standing biomass dry weight, DM) and sward height was estimated using a linear model with an average adjusted R 2 of 0.81 (Riesch et al 2019). Vegetation samples for the analysed forage quality parameters were collected on each sampling date by a hand-pluck approach (Riesch et al 2019). The total nitrogen concentration in plant material was assessed according to the Dumas combustion method in a carbon nitrogen elemental analyser and subsequently converted to CP.…”

“…For this purpose, the random forest regression algorithm was applied on 102 potential predictor variables, including multispectral and SAR indices. The study site was located in the south-east of Germany and is extensively grazed by wild red deer (Cervus elaphus) (Riesch et al 2019). From a methodological point of view, this study was interested in the following research questions:…”