Photo by Zachary Folk
Conservation Endocrinology
The Northern spotted owl, Strix occidentalis caurina, (NSO) is the flagship threatened species of the Pacific Northwest. Federally listed under the Endangered Species Act in 1990, the NSO continues to decline at a rate of about 4% throughout its range. Despite the fact that the NSO is one of the best-studied wild vertebrate species in the world, the relative importance of the threats that it faces remain controversial.
Our Center has developed measures to help assess the relative impacts of pressures such as barred owl invasion, habitat loss and anthropogenic disturbance in NSO. We were the first to develop non-invasive fecal hormone measures of physiological stress (glucocorticoids), reproductive activity (sex steroids) and nutritional state (thyroid hormones) in NSO. Combined, these measures provide a comprehensive profile of NSO physiological health that can serve as a powerful diagnostic tool in studies of disturbance impacts. One of our largest studies employed these tools to document impacts of off-highway vehicle and road exposure on the NSO.
A volunteer rider helps conduct a simulated enduro in Mendocino NF. Photo by Steve Sabo
Project 1: SPOTTED OWL OFF-HIGHWAY VEHICLE STUDY
ABSTRACT
Off-highway vehicle (OHV) use has more than tripled in the last decade and continues to gain in popularity. Finding ways to manage recreation that better protect threatened and endangered species is a high priority for USDA Forest Service (USFS) and US Fish and Wildlife Service (USFWS). We measured the effects of OHV use on the Northern spotted owl (NSO) in the Shasta-Trinity and Mendocino National Forests of northern California, in collaboration with USFS, USFWS, Hubbs-Sea World, Blue Ribbon Coalition and other motorcycle non-profit groups.
We combined correlational and experimental approaches and measured sound level, annual reproductive success and fecal hormones including stress steroids (glucocorticoids) and metabolic hormones (thyroid hormones). Correlational analysis was used to quantify the effects of road noise on NSO hormone levels and reproductive success. Non-invasive fecal measures of glucocorticoids (GCs) and thyroid hormone (T3) were used to assess disturbance. GCs reflect psychological disturbance and nutritional status, whereas T3 reflects nutritional status only. The combination allowed partitioning of these two stress effects. We also experimentally applied an hour of motorcycle exposure to NSO territories in May and July to assess whether and how acute effects of traffic exposure on GC level varied by month, sex and breeding status. For each territory that received experimental OHV exposure we simultaneously sampled an untreated control site using identical methods. Treatment and control sites were assigned at random. In total, 105 experiments were conducted in Shasta-Trinity and Mendocino National Forest between the years of 2005-2008. Final analysis included physiological samples from 165 individual NSO.
Our results suggest that OHV exposure has negative effects on NSO, increasing levels of GCs in the short term and contributing to noise impacts that decrease NSO reproductive success over the long term. Proximity to road increased NSO reproductive success on sites with low ambient noise, but had the opposite effect on sites with high ambient noise. On both quiet and loud sites, levels of T3 increased with proximity to road, suggesting that the positive effects of proximity to roads were mediated through increased prey abundance or foraging efficiency. However, increased GCs associated with noise suggest the positive effects of roads on NSO reproductive success were reversed when noise level (i.e. traffic) was high. Combined, these results suggest that the negative effects of traffic noise on NSO may be amplified by the draw that roads provide for better feeding.
Don Amador, Western Representative of the Blue Ribbon Coalition and an integral member of the research team.
Experimental application of OHV exposure showed that males are most sensitive to disturbance from traffic in May when they are solely responsible for feeding themselves, their mates and their nestlings. In contrast, response to motorcycle exposure among females depended on their T3 levels (nutritional status) and number of young. GCs were highest relative to controls post treatment among females with two young and high T3 (good nutrition). Females with two young and low T3 (compromised nutrition) showed a strong treatment effect but with GC levels actually lower in treated females than in controls. This extreme response indicates that females may be in a state of chronic stress wherein the hypothalamic-pituitary-adrenal (HPA) axis actually shut down and ability to respond to further challenge is severely compromised.
METHODS
Correlational approach:
Jennifer Hartman collects scat from a spotted owl during one of her five seasons of NSO field work for the Center. Photo by Keegan Ramey.
Collaboration with bioaccoustical engineers at Hubbs-Seaworld enabled us to quantify noise on the roads closest to our NSO territories. This allowed us to test for the long-term effects of road nose on reproductive success. In 2006 and 2007 we placed Larson-Davis 820 Type I sound level meters along the road closest to each NSO nest or roost and left them to record continuously for 4 weekend and week days. Technicians later separated out background sounds from wind, water and sustained insect noise. The remaining sound was used as a measure of “road noise”- sound created mostly by vehicles but also by planes and other man-made sources. It represents the best measure available of “routine traffic exposure” for our NSO. Later analysis showed that, for NSO close to roads, this measure had a strong negative effect on reproductive success.
Experimental approach:
Twice during the breeding season, once in May and again in July, NSO in both high and low OHV use areas were experimentally exposed to an hour of motorcycle use between 8:00 and 12:00 in the morning. All scat from all NSO on the territory were collected throughout the day. Identical data collection was conducted on randomly assigned control pairs in high and low OHV use areas that did not receive the experimental OHV exposure.
The intensity of the motorcycle exposure simulated what occurs during enduros, popular events in which large numbers of motorcyclists ride hard for long distances on National Forest trails, sometimes within meters of NSO nests. Our simulated enduros were conducted by volunteers from the local OHV community. Volunteer interns from the Student Conservation Association put in long hours to assist with data collection.
Results:
During May male NSO are responsible for feeding themselves, their mates and their chicks, while females incubate almost continuously until chicks are ten days of age. In July females and males share the job of feeding their fledglings.
Female GC response to motorcycles depended on their breeding status and nutritional condition. Females with young and low T3 (poor nutritional condition) actually showed decreased GCs post enduro treatment relative to controls. Decreased GCs in response to perturbation is a marker of extreme stress, shut down of the hypothalamic-pituitary adrenal axis, and compromised ability to cope with further disturbance.
In our population of NSO levels of T3 increased with proximity to road (effect size = 0.56), indicating that NSO close to roads ate better either due to more prey or easier hunting. Reproductive success also increased with proximity to quiet roads. However, when road noise (i.e. traffic) was high reproductive success decreased with proximity to road, indicating that routine traffic exposure has a very strong negative impact on NSO productivity (effect size = 9.74).
Combined these results suggest that:
1. Northern spotted owl show a physiological response to traffic exposure that varies with sex, season, breeding status and nutritional condition.
2. In our Northern Californian study population, spotted owls close to roads were in better nutritional condition. In California, woodrats comprise a majority of the NSO diet. Similar results would not be expected further north where NSO eat primarily flying squirrels (old growth specialist).
3. Reproductive success is higher close to quiet roads (likely due to better feeding). However, proximity to roads with high noise (i.e. traffic) decreases NSO reproductive success.
These results represent the first evidence to date that Off-Highway Vehicle use is having a strong negative impact on Northern spotted owl population viability. Our report to USDA Forest Service and US Fish and Wildlife Service will be used to help manage recreation so as to minimize impacts on this federally threatened species.
Project 2: IMPROVING SURVEY TECHNIQUES FOR THE SPOTTED OWL
Improving Spotted Owl Surveys
Marvin finds a Spotted Owl roost by the scent of owl pellets. Photo by Chris Zeiminski.
Without question, one of the most serious threats facing the Northern spotted owl is the recent range expansion of another closely related owl species, the Barred owl (BO), Strix varia. Because BOs may attack and kill NSO, NSO are known to vocalize less when around BOs. This poses a serious problem for the manager whose primary means of establishing NSO presence is NSO vocal response to simulated calls. When repeated vocalization surveys yield no NSO vocal response for three consecutive years, the territory is considered unoccupied and habitat protection is lifted. However, vocalization survey results may be unreliable if NSO are unlikely to vocalize due to BO presence. We are addressing this problem by developing a new survey technique that relies on detection dogs trained to locate owl pellets by scent.
In the spring of 2008 a single detection dog, Marvin, was trained to locate spotted owls by the scent of their pellets and feces. Initial training was conducted at the Center training facility in Pack Forest. Validation work occurred on sites in Shasta-Trinity that were used in our long-term OHV study. We used the best available habitat selection model (Zabel, et al 2003) to identify 18 two km2 search cells with high probability of owl occupancy in sites where we knew current NSO locations (although dog handlers remained blind). Marvin successfully located NSO roost sites on 17 of these 18 sites. On one site where he was unsuccessful, the owl was subsequently located outside of the grid created with Zabel’s model. In two cells Marvin found spotted owls that hooters had failed to locate. Most impressively, while it took hooters 63 visits to successfully locate NSO, it only took Marvin 21 visits to find the same pairs. This suggests that incorporating detection dogs into survey protocol can not only reduce bias in the presence of barred owls, but can increase surveying efficiency by three-fold.
In 2009, two different dogs, Lulu and Shrek, located owl pellets and/or whitewash on 11 of 17 sites without prior knowledge of owl presence. This confirmed that the success off 2008 was not due merely to one special dog. Back in the lab we were able to swab the surface of each pellet for cells to determine owl species through polymerase chain reaction (PCR). PCR successfully amplified DNA from samples at 12 out of 20 sites from which pellets were collected. DNA confirmed the presence of spotted owls and barred owls at seven sites each; two of these sites contained pellets from both species. Detection dog surveys indicated the presence of barred owls on sites where only spotted owls were found through vocalization survey and vice versa. Managers were surprised by the prevalence of barred owls in a forest thought to still be relatively free of this invader.
Our results suggest that trained dogs are effective in simultaneously locating spotted and barred owl roosts without vocalization, and will likely increase spotted owl detection probability and survey efficiency in the presence of barred owls. Detection dogs may be particularly useful in providing early detection of barred owls when they are less likely to exhibit territorial behaviors in response to vocalization surveys.
Now, in 2010 Shrek and a new dog, Max, are back in the Shasta-Trinity to compare the results of the detection dog surveys with surveys conducted by vocalization surveys conducted in accordance with Forest Service protocol, newly revised to complement the 2008 Northern spotted owl Recovery Plan. If the detection dogs prove the be more effective and efficient than the vocalization surveys we hope that they will be soon be implemented widely throughout the range of the NSO.
Table 1: Summary of outcome for each site surveyed with detection dogs in 2008 and 2009. Auditory and visual confirmation was conducted by either the dog handler and orienteer at the time the dog located the roost, or by an independent crew using vocalization survey methods within two weeks. DNA confirmation occurred after PCR with cells swabbed from the surface of the pellets.
Other small-scale studies by our Center on NSO in Washington, Oregon and California examine how our physiological health measures vary with proximity to road but also with other land use practices where barred owls are also present, in collaboration with researchers conducting long-term demographic studies throughout the NSO range.
NSO with territories closest to roads have higher fecal glucocorticoids than owls with territories further from roads. Morever, fecal glucocrticoids appear to increase in a dose-dependent manner as proximity to road increases. Consistent with this we also found that NSO in National Parks have lower fecal glucocorticoids than NSO in adjacent National Forests, despite comparable numbers of barred owls in adjacent areas.