Redwood

Redwood

National Park Service U.S. Department of the Interior Redwood National and State Parks California State Parks Department of Parks & Recreation Fore...

489KB Sizes 1 Downloads 7 Views

National Park Service U.S. Department of the Interior

Redwood National and State Parks

California State Parks Department of Parks & Recreation

Forest and Beach Corvid Monitoring and Management Trail and Backcountry Management Plan Implementation 2011 Annual Progress Report

July 2012

USFWS Biological Opinion Reference #: 8-14-2003-1517

Title page photographs, clockwise from top: Steller’s jay foraging at Elk Prairie campground – note leg bands, this bird was part of an intensive Steller’s jay habitat use study in the parks; Humboldt State University researcher Will Goldenburg holding recently banded Steller’s jay at Elk Prairie campground; Humboldt State University technician B. Atkinson injecting emetic chemical in fake marbled murrelet egg (colored small chicken egg) as part of a Steller’s jay conditioned taste aversion research project conducted in parks. All photos taken in Redwood National and State Parks in 2011. INTRODUCTION This report is divided into three interrelated sections concerning the monitoring and management of common ravens (Corvus corax), American crows (Corvus brachyrhynchos) and Steller’s jays (Cyanocitta stelleri) (collectively known as corvids) in Redwood National and State Parks (RNSP or parks). Section I covers forest and beach corvid monitoring results for 2007-2011. Section II describes corvid management activities that took place in RNSP in 2011. Section III describes the progress of any projects from the proposed action section of the RNSP Trail and Backcountry Management Plan biological assessment (Bensen 2006). This report also satisfies the reporting requirements stipulated under the terms and conditions of the RNSP Trail and Backcountry Management Plan biological opinion (USFWS 2007a – USFWS ref. # 8-14-20031517). A comprehensive description of the purpose, policy, scientific background, management history, objectives and methods of corvid monitoring and management in RNSP is described in the parks’ Corvid Management Strategy (RNSP 2008a). The following paragraphs provide a brief overview of corvid predation of federally threatened and state endangered marbled murrelets (Brachyramphus marmoratus) and federally threatened western snowy plovers (Charadrius nivosus nivosus) and the parks’ response: The marbled murrelet was federally listed as threatened and California state listed as endangered in 1992. The Marbled Murrelet Recovery Plan (USFWS 1997) specifically identified RNSP as key to species conservation and recovery in California. Section 1.4 of the recovery plan states that nest predation by Steller’s jays and common ravens is a threat to the species. Recovery action 3.1.2 in the recovery plan directs agencies to “decrease adult and juvenile mortality.” This recovery action is given the highest priority rating. The most recent marbled murrelet five year conservation status review (McShane et al. 2004) revealed that nest predation is now the primary cause of current and future murrelet population decline, particularly in California. High rates of murrelet nest predation by corvids in RNSP have been conclusively recorded (Hebert and Golightly 2006, RNSP unpub. data). RNSP contains 62% of all the suitable murrelet nesting habitat in California and approximately 75% of the murrelets detected during at-sea surveys in California were off the coast of RNSP (McShane et al. 2004). Murrelets have been found to forage at sea primarily right off the coast of their inland nesting grounds (Raphael et al. 2004, Hebert and Golightly 2006). The California population represents roughly a third of the federally listed population. Current murrelet fledging success (percentage of chicks leaving the nest alive) within RNSP is estimated to be 0.3% - 2% (Hebert and Golightly 2006). To just maintain the current population size, RNSP fledging rates need to be between 18% and 28% 2

USFWS Biological Opinion Reference #: 8-14-2003-1517

(McShane et al. 2004). Thus, predation of murrelets by corvids in RNSP has the potential to have a significant negative impact on the listed murrelet population. Numerous studies (e.g. Suddijan 2004, Leibzeit and George 2002, , Luginbuhl et al. 2001, George et al. 2001, Wallen et al. 1999) in and near national and state parks in Washington and California have tied increases in localized corvid densities and nest predation rates to supplemental food provided by park visitors. Many of RNSP’s high-use visitor areas (i.e. campgrounds, visitor centers, picnic areas, trailheads) are located within high quality marbled murrelet nesting habitat. Recent studies in RNSP have revealed that Steller’s jay densities in parks’ campgrounds located in murrelet nesting habitat are two to six times greater than in murrelet nesting habitat away from campgrounds (George et al. 2001., Wallen et al. 1999). Conversely, murrelets have been found to have higher chick productivity in old growth forest areas located away from campgrounds that have lower corvid densities (Marzluff and Neatherlin 2006, Luginbuhl et al. 2001, Marzluff et al. 1996). The western snowy plover was federally listed as threatened in 1993. The Western Snowy Plover Recovery Plan (USFWS 2007b) repeatedly states that American crows and common ravens are significant snowy plover nest predators throughout California, Oregon and Washington. Regionally, predation by crows and ravens has been cited as a major cause of plover nest failure in Oregon (Lauten et al. 2006). Colwell et al. (2006, 2010) found that predation by common ravens was the primary factor limiting snowy plover productivity in breeding areas just to the south of RNSP in Humboldt county. The recovery plan (USFWS 2007b) repeatedly states that reducing or eliminating corvid attracting human food waste in or near plover breeding and wintering areas is an important task for the recovery of the species. The RNSP Staff Responsibilities and Management Strategy for Western Snowy Plovers (RNSP 2008b) also recommends the proper disposal of human food waste in and near snowy plover habitat areas in order to lower corvid predation pressure on plovers. Due to the potential negative impact of visitor activities and their influence on corvid predation of marbled murrelets and western snowy plovers within RNSP, a Corvid Management Strategy (RNSP 2008a) was developed. The aim of the strategy is to decrease the density of corvids surrounding visitor use developments in the parks.

SECTION I. FOREST AND BEACH CORVID MONITORING A. Introduction The RNSP Corvid Management Strategy (RNSP 2008a) is adaptive. Effectiveness monitoring is central to the success of the strategy. Monitoring how and whether jay, crow and raven populations are responding to management actions is central to determining whether the goal of reducing corvid densities near high use visitor areas is being met. As the monitoring information is collected, it is hoped that it will assist the parks in directing corvid management activities to 3

USFWS Biological Opinion Reference #: 8-14-2003-1517

the most impacted areas and to using the most effective management techniques. At the least, the monitoring program is designed to determine whether the corvid management actions implemented by RNSP are successfully decreasing the density of corvids near park visitor use areas. Again, for more detail on the monitoring system’s design, rationale and corvid population targets (for forest corvids only), please refer to the RNSP Corvid Management Strategy (RNSP 2008a).

B. Methods 1) Forest Corvid Surveys The point count survey protocol used for the RNSP forest corvid monitoring program is described in Appendix III of the RNSP Corvid Management Strategy (RNSP 2008a). Forest corvid surveys were conducted from May 2011 to September 2011. The 30 monitoring station locations are shown in Figure 1. The stations are grouped according to one of two types they sample, control areas or type of visitor use area. Four control stations are located in marbled murrelet habitat areas at least 0.25 miles away from any visitor development (stations marked “FC” in Figure 1). Six control stations are located in marbled murrelet habitat areas along trails but at least 0.25 miles from any other visitor development (stations marked “TC” in Figure 1). Five stations are located within front country campgrounds in marbled murrelet habitat (stations marked “JS” and “PC” in Figure 1). Eight stations are located in picnic or major trailhead areas in or immediately adjacent to suitable marbled murrelet habitat (stations marked “PN” in Figure 1). Finally, seven stations are located along Redwood Creek downstream of the Bond Creek junction where dispersed backcounty camping is allowed (stations marked “RC” in Figure 1). In 2007 and 2008 the data were analyzed by treating each individual visit to each monitoring station location as a discrete data point (Bensen 2007, 2008a). Recommendations made by the USFWS in a March, 2009 memo written after a review of the Bensen (2008a) report suggested that the analysis may be erroneous. The gist of the critique was that sampling location independence was not the same as temporal sampling independence. Therefore, in order to maintain consistency between survey years and address this critique, data from 2009 through 2010 were analyzed in two ways 1) using the original analysis method of treating each station location visit as discrete and 2) treating each station location as discrete thus eliminating the temporal aspect of the data; a mean of means method was used to cluster the data for the second analysis. A normal distribution was assumed for both analyses. In 2011, an outside study design assessment and statistical analysis was made of all the RNSP Steller’s jay monitoring data (2007 – 2011). The assessment resulted in the data being analyzed using a mixed model analysis of variance (ANOVA). Refer to George and Peery (2012) for a full description of analytical methods. Their results are presented in this report for Steller’s jays. George and Peery (2012) again treated each visit to each point count station as a discrete data point, therefore, the common raven data was analyzed in a similar manner in order to maintain consistency of the results.

4

USFWS Biological Opinion Reference #: 8-14-2003-1517

Figure 1. Location of point count survey stations within Redwood National and State Parks. FC = forest control station (n = 4), TC = trail control station (n = 6), JS = Jedediah Smith campground station (n = 3), PC = Elk Prairie campground station (n = 2), PN = picnic area station (n = 8), RC = Redwood Creek dispersed camping area station (n = 7). Please note that 5

USFWS Biological Opinion Reference #: 8-14-2003-1517

FC3 was moved in 2007 approximately 0.5 miles to the southwest and is now TC6 and is not shown on the map. 2) Beach Corvid Surveys The instantaneous point count survey protocol used for the RNSP beach corvid monitoring program is described in Appendix III of the RNSP Corvid Management Strategy (RNSP 2008a) and is the same as is used throughout the western snowy plover Recovery Unit Two area. The “survey year” for snowy plovers was from October 2010 to September 2011 and so the beach corvid surveys spanned the same period of time. Instantaneous point counts were made “on the twenty minute mark” for foot based surveys, or “on the ten” for ATV based surveys for all corvids within 500 meters of the surveyor while conducting western snowy plover surveys along monitored park beach reaches. Due to the survey methodology, there were no set point count stations and no control stations. The surveys provide an index of relative corvid abundance and frequency but not density. Only the north Gold Bluffs Beach, south Gold Bluffs Beach and Freshwater Spit survey reaches were analyzed because only those reaches were surveyed during all months of the year and on the same schedule. The locations of the three beach survey reaches are described in the RNSP Staff Responsibilities and Management Strategy for Western Snowy Plovers (RNSP 2008b).

C. Results 1) Forest Corvid Surveys The 30 point count survey stations scattered throughout RNSP were visited twice a month from May through September of 2011 for a total of 272 surveys or ten visits to each station except for those on Redwood Creek which were visited only six times due to high water early in the survey season. Approximately 1,000 person hours were spent in the field completing the surveys. The results for Steller’s jays are shown in Figure 2 based on George and Peery’s (2012) analysis. Only detections made within 50m of the survey station were analyzed because it is only within 50m that a high detection probability can be assumed according Luginbuhl et al. (2001) - the methodology that this monitoring program is based upon (RNSP 2008a).

Intentionally left blank

6

USFWS Biological Opinion Reference #: 8-14-2003-1517 3.5

Steller's Jays/Pt. Count

3 2.5 Forest Control

2

Jeddediah Smith Prairie Creek

1.5

Picnic Areas

1

Trail Control

0.5

Redwood Creek

0 -0.5

2007

2008

2009

2010

2011

Year

Figure 2. Mean number of Steller’s jays detected within 50m of point count stations (±SE) in RNSP 2007-2011 (George and Peery 2012). The large home ranges and long distance daily movements of common ravens violate the assumptions of the point count sampling methodology used as part of this monitoring program. This problem was anticipated during the design of the monitoring program (J. Marzluff, J. Black, L. George pers. comm.) and was amply demonstrated by the results in Bensen (2010), Bensen (2008a) and Bensen (2007). None or virtually no ravens were detected within 50m of monitoring stations. The same lack of detections occurred during the 2011 survey year and so a presentation of raven detections within 50m of point count station results were not included in this year’s report. The relative abundance of common ravens can be roughly represented, however, by looking at the “no boundary” plot results, as shown in Table 1 and Figure 3. These results represent all detections at each station, regardless of how far away the individual ravens were from the station. Raven population numbers cannot be estimated with this method nor can a high probability of detection be established, making the results inconclusive.

Intentionally left blank

7

USFWS Biological Opinion Reference #: 8-14-2003-1517

Table 1. Mean number of common ravens detected at any distance of point count stations in RNSP during May through September of 2007 – 2011. Results indicate relative abundance only. Forest Control 40

n

mean

SD

Range

Trail Control

Campgrounds

Picnic Areas

60

50

80

Redwood Creek 421

2011

0.43

0.19

0.38

0.26

0.33

2010

0.32

0.15

0.57

0.10

0.38

2009

0.15

0.12

0.98

0.09

0.39

2008

0.18

0.15

0.90

0.28

0.38

2007

0.2

0.18

0.72

0.24

0.54

2011

0.77

0.47

0.73

0.63

0.98

2010

0.65

0.41

0.69

0.31

0.70

2009

0.36

0.33

1.28

0.29

0.87

2008

0.50

0.4

1.29

0.71

0.96

2007

0.61

0.33

0.86

0.60

1.02

2011

0-3

0-2

0-3

0-3

0-5

2010

0-3

0-2

0-2

0-1

0-2

2009

0-1

0-1

0-7

0-1

0-4

2008

0-2

0-2

0-5

0-3

0-5

2007

0-3

0-1

0-3

0-3

0-5

1

Stations along Redwood Creek could not be visited during four survey rounds because high water made surveys unsafe.

8

USFWS Biological Opinion Reference #: 8-14-2003-1517

Mean Number (±SE) of Common Ravens Detected at Point Count Stations (No Plot Boundary Method) RNSP 2007 - 2010 Represents Relative Abundance Only 1.2

1

0.8

Forest Control Trail Control

0.6

Campgrounds Picnic Areas

0.4

Redwood Creek

0.2

0

2007

2008

2009

2010

2011

Figure 3. Mean number of common ravens (±SE) detected at any distance of point count stations (infinite plot size) in RNSP during May to September, 2007 - 2011.

2) Beach Corvid Surveys The instantaneous corvid point counts conducted during western snowy plover surveys between October 2007 - September 2008, October 2008 – September 2009, October 2009 – September 2010 and October 2010 – September 2011 on three select reaches of RNSP beaches only provide an index of relative corvid abundance between the three reaches. American crows have been marginally most abundant on Freshwater Spit. Common ravens have shown no pattern difference among the three reaches over the past three years but are the most common corvid species detected on all three reaches. The only significant change within a reach is for ravens on north Gold Bluffs Beach with a measureable and possibly significant drop between 2008/9 and 9

USFWS Biological Opinion Reference #: 8-14-2003-1517

2010/11. Only a very small number of corvids were unidentifiable to species and these results are reflected in the very low overall abundance of them on the three select survey reaches (Table 2 and Figure 4). Table 2. Mean number and proportion of common ravens, American crows and unknown corvids observed within 500m per instantaneous point count on select reaches of beach in Redwood National and State Parks 2008 - 2011. Surveys were conducted between October of the previous year and September of the year listed. Results indicate relative abundance only.

Freshwater Spit

North Gold Bluffs Beach

South Gold Bluffs Beach

2008

2009

2010

2011

2008

2009

2010

2011

2008

2009

2010

2011

n

64

82

74

77

179

208

176

185

133

156

112

158

Common Ravens

0.40

0.32

0.47

0.21

0.83

0.85

0.37

0.19

0.64

0.35

0.54

0.38

American Crows

0.28

0.22

0.35

0.10

0.03

0.06

0.03

0.03

0.11

0.21

0.02

0.03

Unknown Corvids

0.06

0.02

0.03

0

0.03

0.03

0

0

0.02

0.02

0.01

0

Common Ravens

0.20

0.66

1.05

0.68

0.22

2.03

1.09

0.58

0.26

0.70

1.85

0.93

American Crows

0.18

0.67

1.07

0.38

0.03

0.32

0.24

0.02

0.08

0.67

0.19

0.02

Unknown Corvids

0.07

0.22

0.16

0

0.03

0.35

0

0

0.02

0.14

0.09

0

Common Ravens

25%

22%

22%

9%

33%

32%

19%

13%

29%

26%

21%

20%

American Crows

17%

13%

20%

8%

2%

4%

2%

2%

7%

13%

<1%

2%

Unknown Corvids

5%

1%

1%

0%

2%

1%

0%

0%

2%

1%

<1%

0%

% survey points with at least one detection

SD

mean

Year

10

USFWS Biological Opinion Reference #: 8-14-2003-1517

Mean Number of Corvids Detected Per Instantaneous Point Counts On Select RNSP Beaches 2008 - 2011 1.2 1 0.8 Common Ravens

0.6

American Crows 0.4

Unknown Corvids

0.2 0 2008 2009 2010 2011 2008 2009 2010 2011 2008 2009 2010 2011 -0.2

Freshwater Spit

North Gold Bluffs Beach

South Gold Bluffs Beach

Figure 4. Mean number of common ravens, American crows and unknown corvids recorded within 500m per instantaneous point count on select reaches of beach in Redwood National and State Parks 2008-2011. Surveys were conducted between October of the previous year and September of the year listed. Error bars represent 95% confidence interval. Results represent relative abundance only.

D. Discussion/Recommendations 1) Forest Corvid Surveys The primary purpose of this past year’s forest corvid survey effort was to continue to compare the effect of more intensive corvid management actions made during the past year to the previous baseline and to compare corvid densities between visitor use areas with differing levels of visitation within RNSP. These objectives continued to prove logistically feasible and should be sustainable in future years if current biological technician staffing levels are maintained. Similar to 2007 - 2010, a within year comparison of means between the 2011 forest survey station categories showed, as expected, that the campground areas contained a significantly higher number of Steller’s jays as compared to the two control category types. This year campgrounds averaged six to ten times the number of jays in the control areas and were similar to results recorded in 2007 – 2010 (George and Peery 2012). There is no trend to indicate that 11

USFWS Biological Opinion Reference #: 8-14-2003-1517

jay density in the campgrounds is decreasing in response to RNSP corvid management actions. The same lack of decrease is also evident at RNSP picnic sites. The dispersed camping area along Redwood Creek continued to have roughly the same jay detection rates as the two control category types in 2011. Interestingly, in 2011, as in 2010, the reach of Redwood Creek where the survey stations are located was completely closed to dispersed campers but the number of jays detected increased slightly (but not significantly) compared to the two previous years when dispersed camping was allowed. The lack of human presence does not appear to have driven down the detection rate, as would have been expected if human food availability were the primary factor for jay population density. If this trend continues in future forest corvid survey years, it may show that human effects on jays is only measurable at high visitation sites like front country campgrounds and front-country picnic sites and is not detectable at low visitation backcountry dispersed camping areas or low use picnic sites. Finally, the two control category types were again nearly identical. Stations along trails had the same detection rates as the forest control stations. A difference analysis was conducted by George and Peery (2012) and they confirmed that there was no statistically significant difference between the two control station types. RNSP discussions with the US Fish and Wildlife Service in early 2012 (B. McIver, L. Roberts, G. Falxa pers. comm.) led to agreement that it is unnecessary to continue the forest control points. A power analysis of the 2007-2011 corvid monitoring data conducted by George and Peery (2012) indicated that the same level of statistical power to detect a 50% decrease in Steller’s jays can be obtained if 4 additional trail control monitoring points are added to the study design. The four forest control points will be dropped and four additional trail control points will be added for the 2012 corvid monitoring season. The forest corvid survey results for common ravens, also as expected (L. George and J. Black pers. comm.), were again not conclusive. Raven territories and daily movement patterns are simply too large, as shown by Scarpignato’s (2011) raven home range study conducted in RNSP, to be accurately sampled using standard point count methods within a heavily forested environment. There were almost no detections within any of the 50m plot areas and so the results were not included in this year’s report. Instead, only the “no plot boundary” results were reported. The “no plot boundary” results are also difficult to analyze because no detection reliability index can be established for birds located greater than 50m from point count stations, thus violating the assumption that all individuals are being observed. The variation in detectability is especially apparent when stations located deep in forests are compared to more open country survey stations like those along Redwood Creek. The longer sight lines of the Redwood Creek stations allow for greater visual detections and may skew results considerably (L. George pers. comm. and pers. obs.). Unfortunately, at this point in time, the raven results are not easily interpreted. The trends seen in figure 3 most likely are not mirroring reality but are instead sampling error artifacts. It is entirely likely that point count methodologies are inappropriate to monitor common ravens in the RNSP forest environment. At this point, no viable alternative common raven monitoring methods have been developed or suggested.

12

USFWS Biological Opinion Reference #: 8-14-2003-1517

2) Beach Corvid Surveys Only a gross geographic analysis of the beach corvid monitoring data was made in 2011. Only between survey reach - between years were compared. The gross analysis shows that there is only a discernable pattern in raven detections on north Gold Bluffs Beach. In 2008 and 2009 it appeared that north Gold Bluffs Beach was used most by ravens compared to the other two reaches, but this pattern no longer held in 2010 and 2011. Raven use of north Gold Bluffs Beach fell significantly in 2010 and 2011 (Table 2, Figure 4). No obvious reason for the apparent significant drop in usage was apparent. Raven use is now roughly even among the three reaches. In 2008, crows were most abundant along Freshwater Spit and least abundant along the north Gold Bluffs Beach reach. In 2009, crows were again relatively most abundant along Freshwater Spit and along south Gold Bluffs Beach but least abundant along north Gold Bluffs Beach. In 2010, crows were again most abundant along Freshwater Spit and this pattern continued in 2011, but less strongly (Table 2, Figure 4). Crows are open country birds that thrive in human altered agricultural/rural landscapes (Liebezeit and George 2002). Freshwater Spit parallels not just US Highway 101 but also the Orick valley pastoral area that contains large areas of beef and dairy livestock grazing pastures – prime crow habitat. Thus, it makes sense that crows have been most abundant on that beach. The 2008 corvid report (Bensen 2008) stated that no other comparable beach corvid survey data from the region has been published and so it is not possible to put this or last year’s RNSP data into a context outside of the park; this is still technically correct. The variation in corvid survey intensity and method (some surveyors use ATV’s and some walk) results in greatly varying survey intensities across the entire western snowy plover recovery unit two area. Recovery unit two is the USFWS species recovery region that encompasses Del Norte, Humboldt and Mendocino counties, including RNSP – similar beach corvid surveys to the parks’s surveys are conducted throughout this area. Despite that caveat, however, the range of frequency of corvid detections per point count in RNSP, approximately 10% to 35% (Table 2), are roughly equivalent to slightly below the “average” frequencies seen on other regional beaches but are considerably less than the 61% detection frequency seen at Mad River Beach (Colwell et al. 2008).

SECTION II. CORVID MANAGEMENT A. Introduction A comprehensive description of the purpose, policy, scientific background, management history, objectives and methods of corvid management in Redwood National and State Parks is described in the parks’ Corvid Management Strategy (RNSP 2008a). The following summary of actions implemented in 2011 is intended to match the organization of Section V - Management Strategy, Section VI – Effectiveness Monitoring, and Section VIII – Planned Actions If Additional 13

USFWS Biological Opinion Reference #: 8-14-2003-1517

Funding Becomes Available, of the RNSP Corvid Management Strategy (RNSP 2008a), for ease of tracking.

B. Corvid Management Actions Implemented Section V. A. - Visitor Education; was implemented with the following tasks accomplished: • Three California State Park (CSP) seasonal Interpreter Specialists and one partial State Park Interpreter II dedicated-to-the-murrelet/corvid program salary equivalents were hired for the May – September high park visitor season. These positions were in addition to other seasonal and permanent National and California State Park interpretive staff that informed visitors about murrelets, corvids and food. These positions patrolled/roved (contacted visitors while moving around high use areas like campgrounds and trailheads) the park complex’s three largest front country campgrounds and various high visitation day use areas providing information on marbled murrelets, clean camping, proper trash disposal and the negative effects of intentionally or unintentionally feeding corvids and other wildlife. The roves were timed to occur from 0800 – 1000 and 1700 – 1900 to coincide with the maximum number of visitors in the campgrounds. In addition, the senior aides gave formal interpretive programs, campfire talks, and Junior ranger programs. Funding for these extra positions came from the Kure/Stuyvesant Oil Spill Restoration Trust Fund that is administered by a trustee council made up of officials from the California Department of Fish and Game – Oil Spill Prevention and Response division and the US Fish and Wildlife Service (USFWS). Non-oil spill funded seasonal and permanent interpretive staff from both the National Park Service (NPS) and CSP roved and made presentations and their numbers are included in the totals above. Interpretive staff conducted approximately 200 murrelet-corvid roves where approximately 10,000 individuals were contacted and 5,000 publications on murreletscorvids were distributed. • A corvid-marbled murrelet education article was included in the 2011 issue of the RNSP visitor guide newsletter. Other murrelet corvid specific publications were also produced and distributed. 40,000 of these educational publications were handed out across the parks. • A short corvid-murrelet educational video produced by California State Parks Santa Cruz District was included in most campfire programs from June through September using newly renovated amphitheater projection booths. • The same educational video was shown at two RNSP visitor centers on a daily basis. A large flat screen television was purchased specifically for this purpose in 2011 and installed in the Crescent City visitor center. • RNSP staffed a murrelet-corvid educational booth for the entire length of the 2011 Humboldt County Fair. Approximately 1,000 fair visitors were contacted and 1,000 publications were distributed. • In 2011, approximately 250,000 visitors were contacted during their time at RNSP and it can be assumed that the vast majority of them were exposed, at least in passing, to some 14

USFWS Biological Opinion Reference #: 8-14-2003-1517



sort of murrelet-corvid educational media (e.g. sign, pamphlet, video, staff contact). Proportions of visitors receiving the message and retaining it were analyzed in Ward et al. (2011) as described in section VI. A., below. A corvid-murrelet dedicated web page was maintained at the publicly accessible Redwood National Park website - http://www.nps.gov/redw/naturescience/marbledmurrelet.htm. Approximately 4,000 unique hits were made to this webpage in 2011.

Section V. B. - Temporary Partial Dispersed Camping Prohibition and Removal of Select Picnic Tables; was implemented this year. No dispersed camping was allowed along the lower section of Redwood Creek from the Bond Creek junction to the downstream park boundary, or approximately 7 miles of previously available dispersed camping area was made unavailable. All backcountry campers must obtain backcountry camping permits at park visitor centers where they received the notice of areas available for camping. Information signs were also installed at Redwood Creek access trails informing backcountry campers of the areas available for dispersed camping. The partial dispersed camping closure on Redwood Creek was feasible and will continue next year. Picnic tables were removed this year previous the beginning of the high use visitor season from five sites per the Corvid Management Strategy: Tall Trees Trail trailhead, Redwood Creek Overlook, Lady Bird Johnson Trail trailhead, Orick Horse Three Hour Loop Trail picnic area and Mill Creek Loop Horse Trail picnic area. Section V. C. - Law Enforcement; was implemented as part of standard law enforcement practices within RNSP. No specific actions were reported to the Corvid Program Manager. Section V. D. - Facility Management; was implemented as part of the standard maintenance procedures of RNSP. Specific actions include the replacement of 12 bear proof food storage lockers and six bear proof trash bins. Section V. E. – Inventory of Potential Human Created Corvid Food Sources Outside of Park. A preliminary inventory was completed in 2008 and the results are included in Bensen (2008a). Section V. F. – Program Coordination and Reporting; a CSP and a NPS staff member continued as the co-Corvid Program Managers to coordinate corvid management activities in RNSP. This report partially satisfies the data analysis and reporting component of this task. Section VI. A. - Visitor Education Evaluation; field data was gathered in 2010, a final report was completed (Ward et al. 2011) and a wrap up meeting between the researchers and NPS, CSP and oil spill trustee council representatives was held in December 2011. The study was conducted by Dr. Carolyn Ward and Dr. Steven Martin. Recommendations are now being incorporated into the RNSP corvid/murrelet visitor education program. Approximately $98,500 in funding was provided by the Kure/Stuyvesant Oil Spill Restoration Trust Fund in 2009 for the study. Section VI. B. - Corvid Monitoring and Reporting; was completed. The survey effort and data analysis described in this report documents this task for 2011. 15

USFWS Biological Opinion Reference #: 8-14-2003-1517

Section VII. A. – Outside-the-parks Corvid Management; was not implemented in 2011 due to lack of additional funding and staff. Section VII. B. – Research; An unprecedented number of corvid management related research studies conducted within RNSP were completed or neared completion in 2011, including: 1) The corvid – marbled murrelet education program effectiveness study headed by Drs. Carolyn Ward and Steven Martin completed and was described above in Section VI. A. Recommendations from the study indicate a need for more direct language to instruct visitors on proper food management behaviors and an increased use of universally recognizable signage in addition to personal contacts would increase the efficacy of the education program. 2) A study on common raven home range use along Prairie Creek was initiated in 2009 and field work continued in 2010, write up was completed in 2011. The parks are in the process of obtaining a copy of the master’s thesis. RNSP staff attended presentations on the study in 2011. The study was conducted by Humboldt State University Wildlife Management graduate student, Amy Scarpignato, under the direction of Dr. Luke George. Results indicated that ravens primarily target roadways preferentially. More detailed analyses and management recommendations are available in the thesis (Scarpignato 2011). 3) A Steller’s jay home range use study in and around RNSP front country campgrounds began in 2010 and continued in 2011. Field work/data gathering was completed. A completed master thesis is expected in 2012. The study is being conducted by Humboldt State University Wildlife Management graduate student Will Goldenburg under the direction of Dr. Luke George. Approximately $1,500 of Kure/Stuyvesant oil spill restoration trust funds administered by Redwood National and State Parks and $1,500 of Redwood National Park base funds were given in support of this study. All of those funds have been spent. Preliminary results indicated that Steller’s jays utilized campgrounds as a feeding and fledged young rearing ground with nests arrayed in a spoke like pattern extending up to one kilometer around the campgrounds. More detailed analyses and management implications will be forthcoming in 2012. 4) Lab experiment for a conditioned taste aversion study targeting Steller’s jays was completed in 2010. Field trials in Redwood National Park were completed in 2011. A final report (Gabriel and Golightly 2011) was received in by RNSP in December 2011. The study was led by Dr. Richard Golightly and Dr. Pia Gabriel from Humboldt State University. Approximately $65,300 of Command oil spill restoration trust funds administered by Redwood National and State Parks were given in support of this study. Conditioned taste aversion of simulated marbled murrelet eggs by Steller’s jays appeared to be effective in wild field conditions. Management implications are provided in the report. Section VII. C. – Additional Visitor Education was implemented in 2011 – primarily through additional visitor contacts by the additional seasonal interpretive ranger staff described above for Section V.A. Section VIII. A. – Adaptive Management Process, was implemented in 2011 with increased educational program intensity, closure of the lower Redwood Creek gravel bars to dispersed camping and removal of picnic tables from five locations throughout Redwood National Park. 16

USFWS Biological Opinion Reference #: 8-14-2003-1517

The four recently completed corvid management research studies will greatly inform park managers in 2012 and a number of changes are expected over the next two years given the additional findings. Section VII. B. - Future Corvid Management Options, will be implemented, if necessary, in one year with a full review of current actions occurring in early 2013 in consultation with the USFWS, California Department of Fish and Game and regional corvid and marbled murrelet experts.

SECTION IV. TRAIL AND BACKCOUNTRY MANAGEMENT PLAN ACTIONS AND AVOIDANCE AND MINIMIZATION MEASURES A. Introduction This section of the report describes all visitor development construction minimization measures implemented by the parks in 2011 as stipulated in the terms and conditions of the RNSP Trail and Backcountry biological opinion (USFWS 2007a).

B. Trail Plan Actions and Avoidance and Minimization Measures Implemented The Lady Bird Johnson – Berry Glen Connector Trail construction was completed and the trail was opened to the public in October, 2010. The 2011 high visitor use period (May through September) was the first high visitation season for this trail. All avoidance and minimization measures were implemented, including primarily that all above ambient noise producing work was conducted outside of the marbled murrelet noise restriction period (24 March – 15 September). Spotted owl (Strix occidentalis caurina) presence surveys were conducted in preparation of the construction of the new route in 2007 and 2008. No spotted owls were detected during the surveys. As with last year, human food availability surveys were again not conducted in 2011 after it was determined after repeated preliminary observation and trials that a meaningful, repeatable monitoring method was impossible. Preliminary trials in 2007 and 2008 showed that the vast majority of campsites and trashcan areas are clean. No food waste was detectable to the observers. Virtually no food waste available to corvids was detected. Subsequent day long observations of individual corvids feeding within the campsites showed that food scraps were so small as to be unnoticed by observers. In addition, successful feeding bouts were extremely short lived, on the order of seconds. Therefore, it was determined that to develop a statistically meaningful, repeatable monitoring program would require a near continuous observation effort. Even with such effort, it is highly debatable whether such information would result in actionable management decisions. More intensive research based observations, such as the results of the Steller’s jay and common raven habitat use studies within RNSP campgrounds, may result in actionable management decisions. 17

USFWS Biological Opinion Reference #: 8-14-2003-1517

All campsites within the Mill Creek Campground have had wildlife proof food storage lockers and all trashcans are wildlife proof. Funding was provided by outside private donations and private non-profit wildlife conservation groups. This project was completed three years ago. Wildlife proof food storage lockers and trashcans were maintained throughout the RNSP complex using internal funding sources as well as oil spill restoration funds.

REPORT PREPARER Keith Bensen, Co-Lead Corvid Management Program, Fish and Wildlife Biologist – Redwood National Park.

ACKNOWLEDGMENTS Amber Transou, Co-Lead Corvid Management Program, Environmental Scientist – North Coast Redwoods District, California State Parks. Marnin Robbins, Interpreter II – North Coast Redwoods District, California State Parks. Jeff Denny, Supervisory District Ranger – Redwood National Park. Debbie Savage, Supervisory District Ranger – Redwood National Park. Terry Hines, Supervisory Biological Science Technician – Redwood National Park. Heather Brown, Biological Science Technician – Redwood National Park. Kyle Max, Biological Technician – Redwood National Park. Numerous seasonal National and California State Parks interpretive staff and biological technicians.

LITERATURE CITED Bensen, K.J. 2009a. Beached Marine Mammal and Seabird Carcass Monitoring – Annual Progress Report 2008. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 10 pp. Bensen, K.J. 2009b. Beached Marine Mammal and Seabird Carcass Monitoring – Annual Progress Report 2009. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 10 pp. 18

USFWS Biological Opinion Reference #: 8-14-2003-1517

Bensen, K.J. 2008a. Forest and Beach Corvid Monitoring and Management, Redwood Creek Corvid and Backcounty Campsite Mapping, Trail and Backcountry Management Plan Implementation – 2008 Progress Report. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 23 pp. Bensen, K.J. 2008b. Western Snowy Plover Winter (Oct. 2007 – Feb. 2008) and Breeding Season (March – Sept. 2008) Surveys, Annual Progress Report. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 10 pp. Bensen, K.J. 2007. Corvid Monitoring, Corvid Management, Trail and Backcountry Management Plan Implementation – 2007 Progress Report. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 13 pp. Bensen, K.J. 2006. A Biological Assessment of Impacts to Terrestrial and Non-Anadromous Aquatic Threatened, Endangered and Candidate Species from the Trail and Backcountry Management Plan in Redwood National and State Parks. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 93 pp. Colwell, M.A., N.S. Burrell, M.A. Hardy, S.E. McAllister, W.J. Pearson, S.A. Peterson, K.G. Ross, and K.A. Sesser. 2010. Final Report: 2010 Snowy Plover Breeding in Coastal Northern California, Recovery Unit 2. Unpub. report to the USFWS. Department of Wildlife, Humboldt State University, Mad River Biologists. 15 pp. Colwell, M.A., N.S. Burrell, K.M. Brindock, M.A. Hardy, and J.J. Muir. 2008. A summary of corvid activity at western snowy plover breeding locations in Humboldt county, CA. Unpublished Report. Wildlife Department, Humboldt State University, Arcata, CA 95521. Colwell, M.A., S.M. Mullin, Z.J. Nelson, C.A. Wilson, J.M. Muir, W.P. Goldenberg, S.E. McAllister, and K.G. Ross. 2006. Final Report: 2006 Snowy Plover breeding in coastal northern California, Recovery Unit 2. Unpublished report, Mad River Biologists, Inc., and Humboldt State University Wildlife Department, Arcata, California. Gabriel, P.O. and R.T. Golightly. 2011. Experimental Assessment of Taste Aversion Conditioning on Steller’s Jays to Provide Potential Short Term Improvement of Nest Survival of Marbled Murrelets in Northern California – Report to National Park Service. Humboldt State University, Arcata. California. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, California. 34 pp. George, T.L. and Z. Peery. 2012. Power Analysis of Steller’s Jay Point Count Surveys in Redwood National and State Parks 2007 – 2011. Humboldt State University, Arcata, California. Unpub. Report on file at South Operations Center, Redwood National and State Parks, Orick, California. 11 pp. 19

USFWS Biological Opinion Reference #: 8-14-2003-1517

George, T. L., R. LeValley, K. Nelson. 2001. Corvid Surveys in Campgrounds and Control Areas in Northern California. Report to the Oil Spill Prevention and Response Office, California Department of Fish and Game. Humboldt State University, Arcata, California. 16 pp. Hebert, P.N. and R.T. Golightly. 2006. Movements, nesting, and response to anthropogenic disturbance of Marbled Murrelets (Brachyramphus marmoratus) in Redwood National and State Parks, California. Unpublished report, Department of Wildlife, Humboldt State University, Arcata, California and California Department of Fish and Game Report 20062, Sacramento, California. 321 pp. Lauten, D.J., K.A. Castelein, E. Seckinger, and E.P Gaines. 2006. The distribution and reproductive success of the western snowy plover along the Oregon coast – 2005. The Oregon Natural Heritage Information Center Institute for Natural Resources, Portland, Oregon. Liebezeit, J.R. and George, T.L. 2002. A Summary of Predation By Corvids on Threatened and Endangered Species in California and Management Recommendations to Reduce Corvid Predation. California Department of Fish and Game Report. Sacramento, California. 144 pp. Luginbuhl, J.M., J.M. Marzluff, J.E. Bradley, M.G. Raphael and D.E. Varland. 2001. Corvid survey techniques and relationship between corvid relative abundance and nest predation. Journal of Field Ornithology 72(4):556-572. Marzluff, J.M., and E. Neatherlin. 2006. Corvid response to human settlements and campgrounds: Causes, consequences, and challenges for conservation. Biological conservation 130:301-314. Marzluff, J.M., J.M. Luginbuhl, M.G. Raphael, D.M. Evans, D.E. Varland, L.S. Young, S.P.Horton, , S.P. Courtney. 1996. The influence of stand structure, proximity to human activity, and forest fragmentation on the risk of predation to nests of marbled murrelets on the Olympic Pennisula. Unpublished Report submitted to the USFWS. January 31, 1996. 63 pp. McShane, C., T. Hamer, H. Carter, G. Swartzman, V. Friesen, D. Ainley, R. Tressler, K. Nelson, A. Burger, L. Spear, T. Mohagen, R. Martin, L. Henkel, K. Prindle, C. Strong and J. Keany. 2004. Evaluation report for the 5-year status review of the marbled murrelet in Washington, Oregon, and Calfornia. Unpublished report. EADW, Inc. Seattle, Washington. Prepared for the USFWS, Region 1. Portland, Oregon. 16 pp. Raphael, M.G., D.E. Mack, B.A. Cooper. Landscape scale relationship between abundance of marbled murrelets and distribution of nesting habitat. Condor. 104(2): 331-342. 20

USFWS Biological Opinion Reference #: 8-14-2003-1517

RNSP. 2010. Forest and Beach Corvid Monitoring and Management, Trail and Backcountry Management Plan Implementation – 2010 Annual Progress Report. Unpublished report on file at Redwood National and State Parks, Orick, California. 21 pp. RNSP. 2008a. Corvid Management Strategy – Redwood National and State Parks. Unpub. report on file at South Operations Center, Redwood National and State Parks, Orick, CA. 45 pp. RNSP. 2008b. Redwood National and State Parks Staff Responsibilities and Management Strategy for Western Snowy Plovers. Unpublished report on file at Redwood National and State Parks, Orick, California. Humboldt State University, Arcata, California. 19 pp. Scarpignato, A. 2011. Home Range and Habitat Use of Breeding Common Ravens in Redwood National and State Parks. Master’s Thesis. Humboldt State University, Arcata, CA. 51 pp. Suddijian, D. 2004. Summary of Monitoring Surveys for the Santa Cruz Mountains. Unpublished report prepared for the Command Oil Spill Trustee Council, OSPR-CDFG. On file at South Operations Center, Redwood National and State Parks, Orick, CA. 30 pp. USFWS. 1997. Recovery Plan for the Marbled Murrelet (Brachyramphus marmoratus) in Washington, Oregon and California. Portland, Oregon. 203 pp. USFWS. 2007a. Formal Consultation on the Trail and Backcountry Management Plan, Redwood National and State Parks, Humboldt and Del Norte Counties, California (#814-2003-1517). 108 pp. USFWS. 2007b. Recovery Plan for the Pacific Coast Population of the Western Snowy Plover (Charadrius alexandrinus nivosus). In 2 volumes. Sacramento, California. 751 pp. Wallen, R., C. Arguello, J. Friedman and D. Baldwin. 1999. The corvids of Redwood National and State Parks: What is their relative abundance and how does abundance vary relative to human activities in the forest? Unpublished report on file at Redwood National and State Parks, Orick, California. 12 pp. Ward, C., J. Taylor and S. Martin. 2011. Evaluation of Communication Strategies to Mitigate Visitor Use Impacts on Marbled Murrelets – Final Report to Redwood National and State Parks. Humboldt State University, Arcata, California. Unpublished report on file at South Operations Center, Redwood National and State Parks, Orick, California. 129 pp.

21

USFWS Biological Opinion Reference #: 8-14-2003-1517

Personal Communications Baldwin, Jim. Statistician, Pacific Southwest Research Station, US Forest Service, Albany, CA. Black, Jeff. Professor, Department of Wildlife, Humboldt State University, Arcata, CA. Falxa, Gary. Fish and Wildlife Biologist, US Fish and Wildlife Service, Arcata, CA. George, Luke. Emeritus Professor, Department of Wildlife, Humboldt State University, Arcata, CA. Gordon, Joel. Computer Specialist (former Resource Management Support Staff member), Redwood National and State Parks, Orick, CA. McIver, Bill. Fish and Wildlife Biologist, US Fish and Wildlife Service, Arcata, CA. Marzluff, John. Associate Professor, Division of Ecosystem Sciences, College of Forest Resources, University of Washington, Seattle, WA. Peery, Zach. Statistician And Researcher, University of California, Berkeley/Moss Landing Marine Laboratories, Moss Landing,California. Roberts, Lynn. Fish and Wildlife Biologist, US Fish and Wildlife Service, Arcata, CA.

22