Impacts of commercial deer recovery on forest and alpine vegetation in Fiordland National Park
This measure relates to indicator 1.3.2 – Invasive species dominance. The data also help inform measures within indicator 1.5.1 – Species composition and diversity.
Background
Red deer were released in Fiordland in the early 20th century and are now found in nearly all vegetated habitats in Fiordland National Park, where they have heavily modified the vegetative structure and composition. Wild animal recovery operations (WARO) have been permitted to harvest deer from public conservation land in New Zealand since 2006. DOC is monitoring how this commercial deer recovery is affecting alpine and forest vegetation in Fiordland National Park.
WARO hunting efforts have induced significant recovery of the alpine vegetation in Fiordland National Park but have been insufficient for the recovery of forested ecosystems.
What did we measure?
In 2006, DOC established 40 monitoring sites in alpine areas throughout Fiordland National Park to quantify deer browse on three selected alpine plant species: Celmisia verbascifolia, Dolichoglottis scorzoneroides and Ranunculus lyallii. Five transects (50 m × 2 m) were established at each monitoring site, along which all individuals of each of these plant species were counted and classified as browsed or not browsed. Deer pellet groups were also tallied for each transect.
At the same time, seedling ratio index (SRI) lines were established in three representative Fiordland catchments that are subject to WARO deer control (see Figure 5 - Maps)to evaluate whether changes in deer browse, deer density (as measured by the number of pellet groups) and the ratio of seedlings to saplings of palatable species changed over time. Within each catchment, ten transects were established in low- to mid-altitude forest at random points with random transect bearings, following published SRI standards. These sites were compared with sites on Secretary Island, which underwent deer eradication between 2006 and 2014.
What did we find?
The numbers of deer harvested increased from 2005/06 to 2007/08 and then remained relatively steady between 2008/09 and 2013/14 (Figure 1). However, lower harvesting rates have been observed over the past six seasons (2014/15–2019/20) due to fluctuations in the feral venison price, highlighting the fickle nature of the industry .
Browse on all three monitored plant species and the number of deer pellet groups recorded along the transects decreased between 2006 and 2012 but have subsequently increased(Figure 2 and Table 1).
There has been no significant change in the browse rate or ratio of seedlings to saplings for species eaten by deer in the three selected catchments in Central Fiordland, indicating that WARO deer recovery has had no measurable effect on the vegetation (Figure 4). In contrast, browse rates have declined to 0 and the ratio of seedlings to saplings of palatable species has increased from strongly negative to close to zero (i.e. equal proportions of seedlings to saplings of palatable species) on Secretary Island (Figure 3), indicating strong recovery of understorey species following a release from deer browsing.
Figure 1: Annual deer harvest in Fiordland National Park from 2005/06 to 2019/20. The dashed red lines indicate years when alpine monitoring occurred using the rapid deer browse method occured and the dashed blue line indicates the hypothesised threshold at which declines in deer browse on monitored species in alpine areas are believed to be induced.
Figure 2: Estimated browse rates for all species combined in areas subject to WARO hunting efforts. Values are means ± 95% confidence intervals. Figure supplied by Manaaki Whenua – Landcare Research (N. Mason).
| Season | Number of pellet groups |
|---|---|
| 2006 | 115 |
| 2008 | 30 |
| 2012 | 4 |
| 2017 | 37 |
Figure 3: Seedling Ratio Index (SRI) values and browse scores at the Secretary Island study sites
Figure 4: Seedling Ratio Index (SRI) values and browse scores at the Central Fiordland study sites
Data quality
This measure is classified as a case study under New Zealand’s Environmental Reporting Framework. Therefore, although the general trends may apply to similar species in similar habitats elsewhere, caution should be used if attempting to extrapolate these results.
Monitoring alpine deer browse is a novel method, so these data are considered to be of medium accuracy. However, independent peer review of the method and power analysis of the data suggested that the trends are robust and the methodology is fit for purpose.
Glossary of terms
Rapid deer browse method is a rapid monitoring method to evaluate deer browse on selected palatable species and changes in deer abundance through pellet counts.
Seedling ratio index (SRI) considers the ratio of short to tall plants grouped in palatability classes to determine the impact of deer browsing on forest understory regeneration. The method also quantifies browse on plants within these palatability groups and evaluates changes in deer abundance through pellet counts.
Additional resources
Sweetapple, P.J., Nugent, G., 2004. Seedling ratios: A simple method for assessing ungulate impacts on forest understories. Wildlife Society Bulletin 32, 137–147.