AVIAN DIVERSITY AND RIPARIAN FOCAL SPECIES ABUNDANCE ON THE LOWER CARMEL RIVER, MONTEREY COUNTY, CA

Transcription

1 AVIAN DIVERSITY AND RIPARIAN FOCAL SPECIES ABUNDANCE ON THE LOWER CARMEL RIVER, MONTEREY COUNTY, CA Annual Avian Monitoring Report Prepared for the Monterey Peninsula Water Management District Prepared By: Karen Shihadeh 1 and Nellie Thorngate 2 18 March 2008 Ventana Wildlife Society Conserving Native Wildlife and Their Habitats Portola Dr., Ste. F-1 Salinas, CA karenshihadeh@ventanaws.org; nelliethorngate@ventanaws.org;

2 ACKNOWLEDGEMENTS We would like to thank the Monterey Peninsula Water Management District for providing the funding and permissions for this project. Thomas Christenson of MPWMD provided onthe-ground support for the implementation of the project, David Mullen of EIP Associates established the point count locations and survey methodology, Chris Tenney collected, collated, and reported on the data from 1999 through Sarah Stock, Jessica Griffiths, Eric Miller, Mike Tyner, Joshua Scullen, and Jennifer Olson conducted surveys, entered, and reported on data for VWS in the years since Kelly Sorenson provided project oversight. Thomas Christensen and Rosemary Kenner were helpful in draft reviews. 2

3 TABLE OF CONTENTS ABSTRACT... 6 INTRODUCTION... 7 PROJECT OBJECTIVES... 8 METHODS... 9 Study Area... 9 Field Methods... 9 Statistical Analyses... 9 Shannon-Diversity Index Simple Linear Regression Analysis of Variance Distance Diversity and Abundance Analysis RESULTS Spring Results Descriptive Statistics Riparian Focal Species Abundance Summer Results Descriptive Statistics Riparian Focal Species Abundance Distance Diversity and Abundance Analysis DISCUSSION RECOMMENDATIONS CONCLUSIONS LITERATURE CITED APPENDIX 1. SITE DESCRIPTIONS AND HABITAT FEATURES OF NINE AVIAN SURVEY TRANSECTS LOCATED ALONG THE LOWER CARMEL RIVER APPENDIX 2. GEOGRAPHIC COORDINATES OF 9 AVIAN TRANSECT SURVEYS IN THE LOWER CARMEL RIVER WATERSHED (NORTH AMERICAN DATUM 1983 GEOGRAPHIC COORDINATE SYSTEM) APPENDIX 3. SAMPLE POINT COUNT DATA FORM APPENDIX 4. SPRING SPECIES ABUNDANCE IN THE LOWER CARMEL RIVER, APPENDIX 5. VALUES AND SPECIES RICHNESS FOR SPRING, APPENDIX 6. SUMMER SPECIES ABUNDANCE IN THE LOWER CARMEL RIVER, APPENDIX 7. VALUES AND SPECIES RICHNESS FOR SUMMER,

4 LIST OF FIGURES Figure 1. The lower Carmel river watershed with land cover and avian transect point count locations, Monterey Co., CA Figure 2. Linear regression of spring mean for all transects on the lower Carmel River (r 2 =0.055, P= 0.399) Figure 3. Box plot representation of difference in value means (P= 0.088) between restored transects and reference transects on the lower Carmel River from during the spring Figure 4. Linear regression of spring abundance of Riparian Focal Species for all transects on the lower Carmel River (n= 15, r 2 =0.320, P= 0.028) Figure 5. Average abundance with standard errors of Riparian Focal Species by transect on the lower Carmel River in the spring from 1992 and Kruskal-Wallis analysis revealed significant difference in abundance between transects (P= 0.001), Monterey Co, CA Figure 6. Linear regression of spring abundance of Riparian Focal Species by transect for all years during the spring on the lower Carmel River revealed a significant trend toward increasing abundance as transects approached the coast (n= 9, r 2 =0.633, P= 0.010). 19 Figure 7. Mean count with standard errors of each Riparian Focal Species encountered on the lower Carmel River in the spring 1992, , Monterey Co., CA (P= 0.000). 21 Figure 8. Spring abundance of Song Sparrows increased with time on the lower Carmel River, Monterey Co., CA (n= 15, r 2 = , P= 0.008) Figure 9. Linear regression of summer mean for all transects on the lower Carmel River from 1992 through 2007 (r 2 =0.058, P= 0.370, n= 16), Monterey Co, CA Figure 10. Transects differed in their average values, included with standard errors, during the summer over the course of the study (P= 0.022, K-W=17.876, df= 8, , on the lower Carmel River, Monterey Co., CA Figure 11. Transect 4C values increased significantly from (r 2 = 0.342, P= 0.017, n= 16) during the summer on the lower Carmel River, Monterey Co. CA Figure 12. Box plot representation of difference in value means (P= 0.056) between restored transects and reference transects on the lower Carmel River from during the summer Figure 13. Linear regression of summer abundance of Riparian Focal Species for all transects on the lower Carmel River (n= 16, r 2 =0.427, P= 0.006) Figure 14. Average abundance with standard errors of Riparian Focal Species by transect on the lower Carmel River during the summer from , Monterey Co, CA (P= 0.082)

5 Figure 15. Linear regression of summer abundance of Riparian Focal Species by transect for all years on the lower Carmel River revealed a trend toward increasing abundance as transects approached the coast (n= 9, r 2 = 0.455, P= 0.046), Monterey Co., CA Figure 16. Mean count with standard errors of each Riparian Focal Species encountered on the lower Carmel River in the summer , Monterey Co., CA (P= 0.000) Figure 17. Summer abundance of Song Sparrows increased from on the lower Carmel River, Monterey Co., CA (n= 16, r 2 = , P= 0.016) Figure 18. Riparian Focal Species average number of detections within 50 meters during the spring of 2007, lower Carmel River, Monterey Co. CA (P= K-W= , df= 6) Figure 19. Riparian Focal Species average number of detections within 50 meters during the summer of 2007, lower Carmel River, Monterey Co. CA (P= 0.000, K-W = , df= 2) LIST OF TABLES Table 1. Linear regression results of spring trends by transect (n= 15) Table 2. Regression table of spring abundance of Riparian Focal Species over time by transect on the lower Carmel River, Table 3. Linear regression results of summer trends by transect (n= 16) Table 4. Regression table of summer abundance of Riparian Focal Species over time by transect on the lower Carmel River during the summer, Monterey Co., CA, Table 5. values, species evenness and richness of birds detected within 50 meters of point count locations in 9 transects on the lower Carmel River during the spring of 2007, Monterey Co., CA Table 6. values, species evenness and richness of birds detected within 50 meters of point count locations in 9 transects on the lower Carmel River during the summer of 2007, Monterey Co., CA Table 7. Average number of detections within 5 minutes and 50 meters of Riparian Focal Species on the lower Carmel River in the spring and summer of 2007, Monterey Co, CA. Means are compared to maximum breeding abundances reported in RHJV (2004) for the Central Coast

6 ABSTRACT Riparian ecosystems are considered the most valuable habitat for the conservation of California s resident and migrant landbirds, supporting more avian species than all other habitats combined. In transects containing 4 fixed point counts each spaced at least 200 meters apart were established on the lower Carmel River to monitor how avian diversity and abundance respond to projects undertaken by the Monterey Peninsula Water Management District. Transects were visited twice in the spring (late May) and twice in the summer (middle August). Shannon Diversity Indices and Riparian Focal Species abundance were compared between years using linear regression and between transects using nonparametric analysis of variance to assess annual trends and impact of differing restoration regimes and hydrological conditions. During 15 years of avian surveys in the lower Carmel River, has not changed in either spring (P= 0.399) or summer (P= 0.370). A marginally significant difference was detected between restored and reference sites in both the spring (P= 0.088) and summer (P= 0.056), with reference sites having higher than restored. Abundance of Riparian Focal Species has increased significantly over the past 15 years in the spring (P= 0.010) and in the summer (P= 0.006). Of the 8 Riparian Focal Species detected in spring 2007, 5 were significantly low when compared to maximum breeding abundances for the Central Coast: Common Yellowthroat (P= 0.000), Swainson s Thrush (P= 0.047), Wilson s Warbler (P= 0.012), Yellow Warbler (P= 0.000) and Yellow-breasted Chat (P= 0.000). Although the study design likely biased true abundance values for these species, this study suggests that the lower Carmel River lacks certain habitat variables, such as tall emergent vegetation, large habitat patches, and canopy closure, favorable for nesting for these species. The presence of Song Sparrows, Black-headed Grosbeaks, and Warbling Vireos in numbers consistent with maximum breeding levels reported for the Central Coast demonstrates that the lower Carmel River has adequate shrub cover close to the river, edge habitat, and large overstory trees. Since 1992, the Carmel River has continued to support the same diversity of birds despite drastic changes, including flooding and increased drawdowns from development. Findings here can be used to guide land managers in adaptive restoration measures and also act as a baseline for establishing Riparian Focal Species population targets for the Carmel River and the Central Coast. 6

7 INTRODUCTION Riparian woodlands constitute less than 0.5% of habitat in California, but support a disproportionate amount of the state s wildlife: more than 225 species of birds, mammals, and herpetofauna can currently be found in California s riparian habitats (RHJV 2004). Riparian ecosystems are considered the most valuable habitat for the conservation of California s resident and migrant landbirds, supporting more avian species than all other habitats combined. Riparian areas serve not only as important stopover sites for migratory songbirds, but also as important foraging and breeding sites for resident, temperate migrant, and neotropical migrant songbirds (Knopf et al. 1988). Despite their importance for wildlife, as well as their role in maintaining water quality and streambed integrity, riparian woodlands have suffered extensive disturbance from agricultural conversion, grazing, channelization, recreation and the invasion of exotic plants (Knopf et al. 1988). In the last century, riparian woodlands have declined by 95% nationwide (Donovan et al. 2002), covering only 2% to 15% of their historic range in California (RHJV 2004), and approximately 5% in Monterey County (Roberson and Tenney 1993). Their loss is considered the most significant cause of population decline among landbirds in the west (RHJV 2004). The ecological value of riparian habitat coupled with its widespread degradation makes it the top priority for the conservation of neotropical migrants and resident birds in the west (RHJV 2004). Studies have demonstrated that birds benefit from riparian restoration, exhibiting increases in both species diversity and abundance in response to restoration efforts (Larison et al. 2001). Consequently, the restoration of riparian habitats has become a priority for conservation ecologists, as is reflected by the increasing number of riparian restoration projects that have been initiated or planned throughout the United States (Holl et al. 2003, Naiman et al. 2000). Long-term monitoring of riparian avian populations, especially after human disturbance and/or restoration, can be an effective method of ascertaining the overall health of the ecosystem and can also lead to improved management of riparian habitats (Ralph et al. 1993). Avian populations respond quickly to changes in the environment, and are costeffective and relatively easy to monitor, making them excellent resources for environmental monitoring (Ralph et al. 1993). Although avian communities are important indicators of ecological health, it can be difficult to analyze local avifauna on a species-by-species basis, so sets of species may be chosen to collectively represent features of interest in the habitat. Riparian Focal Species (RFS) designated in the RHJV (2004) represent the breadth of niches available in a riparian system and as such their presence can be used to gauge the health of riparian system for maintaining essential spatial and habitat qualities. Monitoring and management of these key species equates to whole system management. Because RFS are often closely associated with a complex set of habitat features by identifying species in limitation, habitat variables in stress can also identified and targeted for improvement. In 2006, Thorngate et al. (2006) focused 7

8 on functional guilds to describe ecosystem function. A guild is generally described as a group of organisms that exploit similar resources in similar ways (Simberloff and Dayan 1991). As with focusing on RFS, functional guild analysis allows for more focused management decision making. Results of that analysis were widely varied, likely because functional guild analysis is highly sensitive to habitat variation. Thorngate et al. (2006) postulated that microscale habitat variations influencing functional guild use were vegetation structure and composition, hydrography, geography, adjacent land use, and microclimate. Thus, because of this complexity, unless paired with habitat inventory to remove confounding variation, interpretation of ecosystem function using functional guilds is problematic. The use of RFS to describe habitat availability and functionality provides a much clearer interpretation of habitat health and is employed by many researchers to gauge ecosystem health and response of wildlife to habitat changes. RFS were selected by California Partner s in Flight based on 5 criteria: (1) they breed primarily in riparian vegetation in California, (2) they have a special management status, (3) their historic breeding range has been reduced, (4) they are common enough, ie., have large enough population sizes, to be used to gauge response to habitat change, and (5) they represent the full range of successional stages within a riparian system, and therefore can be used to gauge restoration success (RHJV 2004). Within the context of management, it is the species that has the most demanding requirements that ultimately determine the minimum acceptable value for a habitat feature, eg., patch size, canopy cover, etc. Because of this, in managing for the habitat needs of RFS, the needs of many other riparian-dependant species, including non-avian wildlife, are also met (RHJV 2004). PROJECT OBJECTIVES The Carmel River Watershed, located in Monterey County along the central coast of California has been subject to significant anthropogenic disturbance. The Monterey Peninsula Water Management District (MPWMD), charged with environmental protection and management of water resources within MPWMD s boundary, has implemented a variety of restoration efforts along the lower Carmel River in order to mitigate damage done by high river flows, damming, groundwater extraction and increased residential development in the watershed. In 1992 a Carmel River avian monitoring program was initiated by EIP Associates of Berkeley, CA, in an effort to monitor how avian diversity and abundance respond to various projects undertaken on the Carmel River by MPWMD (EIP Associates, 1993). Since 1999 the ongoing monitoring effort has been administered by the Ventana Wildlife Society (VWS). The objectives of this report are to (1) describe avian communities using an index for species diversity and RFS abundance on the lower Carmel River, (2) to evaluate temporal and geographic trends in avian species diversity and RFS abundance, (3) to compare avian population measures between restored and unmanaged riparian habitats as well as between perennially watered and seasonally dry reaches and (4) to assess the overall health of the system using RFS presence and abundance as indicators. 8

9 METHODS Study Area Approximately 437 acres of riparian forest exist along the mainstem of the Carmel River from San Clemente Dam to Carmel River State Beach south of Monterey, California (Christensen 2003). Land cover in the lower portions of the watershed is primarily Valley Foothill Riparian, with substantial residential and agricultural development throughout its length from the Carmel Valley Village to the mouth of the river just south of Carmel-by-the- Sea (Figure 1). The Carmel River is an important source of water for residents of the entire Monterey Peninsula and Carmel Valley, and provides habitat for a diversity of wildlife including Steelhead Trout (Onchorynchus mykiss), the endangered California Red-legged Frog (Rana aurora draytonii), and a wide variety of bird species including many neotropical migratory birds. In 1992, nine transects ranging from 1250 to 3300 feet in length were established in order to assess baseline avian diversity in the lower CRW, and to assess the effects of a proposed new Los Padres Dam on the upper Carmel River (Appendix 1). Transects were established in riparian habitats expected to be inundated as a result of the new dam (EIP Associates 1992). To investigate the effects of increased drawdowns and other changes in river water levels on bird populations, transects in perennially watered reaches were classified as wet and those in seasonally dry reaches were designated as dry (EIP Associates 1992; Appendix 1). The vegetation surrounding the transects varied with respect to habitat structure, floristic characteristics and hydrologic characteristics. In order to explore the effects of restoration activities on avian diversity since the inception of these surveys, the sites were classified as either restored (those sites having undergone some restoration activity since 1980) or reference (those sites with no recent restoration) sites (Appendix 1). Field Methods Two 20-minute avian surveys were conducted along each transect during spring (April 1 May 31) and late summer (August 1 September 31). Surveys were comprised of four 5- minute unlimited-radius counts conducted at equidistant points spaced at least 200 meters apart along each transect (Ralph et al. 1993, Ralph et al. 1995, see EIP Associates 1993 for detailed survey methodology). All birds seen and heard were recorded and separated by distance between those within 50 meters of the point and those outside 50 meters. Birds flying over the point were recorded but not included in the analysis. See Appendix 1 for descriptions and habitat features of each transect, Appendix 2 for the geographic coordinates of each survey point, and Appendix 3 is an example of the data sheet. Statistical Analyses Data from all previous years point count surveys conducted for this study ( ) were combined with the current year s data to examine trends in avian species diversity and RFS abundance in the lower Carmel River, to assess differences in avian communities between restored and unmanaged habitats, and to determine localized differences in avian communities using between-site comparisons. Spring and summer data sets were considered separately. Spring data was not taken in the 1993 season. Unless otherwise indicated, 9

10 statistical analyses were performed using SYSTAT Version 10 statistical software (SPSS 2000). For all statistical tests the alpha level, or the probability of rejecting the null hypothesis when it is actually true (Type I error), was P= As was done in previous reports, data from each point within a transect were summed to give a total transect count. Likewise, data were summed between the two within-season visits. In order to analyze count data from both previous transect surveys when point count duration lasted 15 minutes ( ) and the current year s point count surveys, the transect data was standardized by multiplying each count by Analyses of the effect of count duration on detection of individual birds have reported that approximately 50 to 60 percent of all individuals are detected within the first five minutes of the count (Lynch 1995, Petit et al. 1995). Since the Carmel River point counts are five minutes each, adjusting the longer transect counts to 55% of the total should yield accurate count estimates for comparisons between sites and years. Shannon-Diversity Index A species diversity index was generated for each transect in each year using the Shannon Diversity Index () calculation in the MVSP (Multivariate Statistical Package) computer program (MVSP 2006). The formula for the Shannon Diversity Index (H') is as follows: s H' = - p i logp i i =1 where s equals the total number of species detected, and P is the proportion of the ith species. The Shannon Diversity Index combines the species richness, or total number of species, with the proportion of individuals contributed to the community by each species to arrive at a measure of relative diversity (Begon et al. 1996). Higher diversity values infer higher species diversity. The same formula was used to calculate species diversity in previous reports (e.g. Griffiths and Stock 2005); however unlike analyses for previous reports, birds identified only to genus were not included in the analysis, resulting in slightly different diversity values than those reported in previous years reports. Simple Linear Regression values and riparian focal species abundance data were normalized using the square root transformation to satisfy test assumptions of the statistical test employed to test for trends. Simple linear regression was used to determine significance of trends. Linear regression tests for a functional linear relationship between one variable and another, in this case and year and and location. This relationship is explained by the equation for a straight line: Y i = α + βx i + ε i where α and β are constants describing the population, X is the dependant variable () and Y is the independent variable (eg., year) and ε is the error term, or the residuals, and explain the distance at which data points differ from the straight line. The sum of al of the ε i s is zero. The regression describes a line with the best fit (having the least deviation) through a 10

11 series of data points (Zar 1999). Temporal and geographic trends in avian species diversity and abundance were visualized using trend lines. Analysis of Variance and RFS abundances were compared between years using non-parametric analysis of variance, the Kruskal-Wallis (K-W) test. This test investigates differences in sample means between groups, where the null hypothesis describes equality among sample means, eg., H 0 : μ 1 =μ 2 =μ 3 (Zar 1999). Differences were tested between sites and between years. Sites were then categorized according to restoration history (restored vs. reference) and water availability (perennially watered vs. seasonally dry), using non-parametric Mann-Whitney U test (Zar 1999). This tests the hypothesis that the two means are different (H A : μ B μ A ) against the null hypothesis that they are the same (H 0 : μ B = μ A ). Differences in means were visualized in a box plot. Distance Diversity and Abundance Analysis Current year (2007) data were restricted to birds recorded within 50 meters so as to facilitate comparison with other studies. As described above, birds flying over and birds only identified to genus were removed from the analysis. analysis was conducted as described above. Tests for variance in means of RFS abundance data differed from described above in that within sampling period visits were not summed but considered as separate visits. Also included was null data from visits in which 0 birds of a RFS were detected, as to accurately measure the mean detection rate per species within 5 minutes. Differences in means between average number of detections within 5 minutes on the Carmel River and those reported by RHJV (2004) were analyzed with the Mann-Whitney U test. 11

12 Figure 1. The lower Carmel river watershed with land cover and avian transect point count locations, Monterey Co., CA.

13 RESULTS Spring Results Descriptive Statistics A total of 29,513 individuals of 107 species were recorded during the spring over 15 years from 1992 through 2007 (excluding 1993 in which there were no spring surveys, Appendix 4). Most individuals (4,578) were noted from transect 4C, the Carmel River mouth highway 1 bridge while the least (2,818) were recorded from transect 2A, DeDampierre Park in Carmel Valley Village. Nine Riparian Focal Species were recorded during that same period: Black-headed Grosbeak, Common Yellowthroat, Song Sparrow, Spotted Sandpiper, Swainson s Thrush, Tree Swallow, Warbling Vireo, Wilson s Warbler, and Yellow Warbler. For all years of study there was no change in the mean for all transects (r 2 =0.055, P= 0.399, Figure 2, Appendix 5). The average for all years for the spring was Average for all transects in 2007 was did not change within transects over the course of the study (Table 1). There were also no differences in values between transects for all 15 years (P= 0.173, K-W=11.545, df= 8). A difference was detected at the P= 0.1 level between restored and reference with reference sites having higher than restored sites (t = , df= 133, P= 0.088, Figure 3). No trends were detected over time for restored (r 2 = 0.000, P= 0.915) or reference (r 2 = 0.000, P= 0.964) transects. Riparian Focal Species Abundance From 1992 to present, abundance of Riparian Focal Species although significantly different between years has not increased significantly over time in the lower Carmel River (r 2 = 0.320, P= 0.028, n= 15, Figure 4). A difference was detected between transects in the average abundance of RFS (P= 0.001). RFS were most often recorded at 4C Highway 1 bridge and least recorded from 2A DeDampierre and 3B downstream of the Schulte Road bridge (Figure 5). No difference was detected in mean abundance of Riparian Focal Species between restored and reference transects (P= 0.318). Linear regression of annual abundance by transect revealed that Riparian Focal Species mean abundance increased significantly with proximity to the ocean (r 2 = 0.633, P= 0.010, n= 9, Figure 6). Analyzed individually, among the 9 transects, 5 increased significantly in abundance of Riparian Focal Species from (n= 15): 4B Riverwood Complex (r 2 = 0.531, P= 0.003), 2A DeDampierre Park (r 2 = 0.465, P= 0.005), 2C Carmel Valley Ranch Golf Club (r 2 = 0.350, P= ), 3B Schulte Bridge (r 2 = 0.312, P= 0.038), and 4C Highway 1 bridge (r 2 = 286, P= 0.040). The remaining 4 transects showed no change in abundance over time or between years (Table 2). Riparian Focal Species differed significantly in their mean abundance on the lower Carmel River in the spring (P= 0.000, K-W=53.791, df= 8). Song Sparrows were 13

14 encountered most often while Spotted Sandpipers and Yellow Warblers were recorded the least (Figure 7). Of the nine RFS observed from 1992 through 2007, only 7 were recorded with enough regularity in the spring to analyze with regression These included Wilson s Warbler, Black-headed Grosbeak, Song Sparrow, Warbling Vireo, Tree Swallow, Spotted Sandpiper and Swainson s Thrush. Of these, three increased in abundance significantly over time: Song Sparrow (r 2 = 0.427, P= 0.008, n= 15, Figure 8). Wilson s Warblers tended to increase with time, but the regression was not significant at the P= 0.05 level (r 2 = 0.221, P= 0.077, n= 15). Swainson s Thrush, Black-headed Grosbeaks, Warbling Vireos, and Tree Swallow abundance increase or decline over the study period. There was no difference for any species, including Yellow Warbler, with regards to past history of restoration. However, 3 of the 7 species, Black-headed Grosbeak (P= 0.025), Warbling Vireo (P= 0.032), and Swainson s Thrush (P= 0.010) showed a transect-specific preferences, in that they were recorded more frequently at some transects than others. Wilson s Warbler and Song Sparrow had marginal significance at the P= 0.1 level with regards to transect (P= and 0.055, respectively). Tree Swallow and Yellow Warbler were represented more evenly throughout the lower Carmel River. Black-headed Grosbeaks were found more often at transect 4C than at any other transect. Warbling Vireos were recorded most frequently at transects 2C and 3A and least at transect 4B. Swainson s Thrush was recorded most often at 4C and least at 2C. Wilson s Warblers tended to occur most frequently at transects 4B and 4C and less at transects 2C and 3B. Song Sparrow was found most often at 4C and 2C and was represented more or less evenly at the other transects. 14

15 Figure 2. Linear regression of spring mean for all transects on the lower Carmel River (r 2 =0.055, P= 0.399) Table 1. Linear regression results of spring trends by transect (n= 15). Transect # Transect Name Regression (r 2 ) Significance (P) 2A DeDampierre Park B Garland Park C Carmel Valley Ranch Golf Club 3A Robinson Canyon Road/ Red Rock 3B Schulte Road bridge C Quail Lodge/ Valley Greens Drive bridge 4A Rancho San Carlos B Riverwood Complex C Carmel River mouth/ highway 1 bridge

16 Figure 3. Box plot representation of difference in value means (P= 0.088) between restored transects and reference transects on the lower Carmel River from during the spring. 16

17 Figure 4. Linear regression of spring abundance of Riparian Focal Species for all transects on the lower Carmel River (n= 15, r 2 =0.320, P= 0.028) 17

18 Figure 5. Average abundance with standard errors of Riparian Focal Species by transect on the lower Carmel River in the spring from 1992 and Kruskal-Wallis analysis revealed significant difference in abundance between transects (P= 0.001), Monterey Co, CA. 18

19 Figure 6. Linear regression of spring abundance of Riparian Focal Species by transect for all years during the spring on the lower Carmel River revealed a significant trend toward increasing abundance as transects approached the coast (n= 9, r 2 =0.633, P= 0.010). 19

20 Table 2. Regression table of spring abundance of Riparian Focal Species over time by transect on the lower Carmel River, Transect # Transect Name N Regression (r 2 ) Significance (P) 2A DeDampierre Park B Garland Park C Carmel Valley Ranch Golf Club 3A Robinson Canyon Road/ Red Rock 3B Schulte Road bridge 14* C Quail Lodge/ Valley Greens Drive bridge 4A Rancho San Carlos B Riverwood Complex 14* C Carmel River mouth/ highway 1 bridge *Outliers removed from analysis: 3B, the 1999 count of 1 RFS and for 4B, 47 RFS in

21 Figure 7. Mean count with standard errors of each Riparian Focal Species encountered on the lower Carmel River in the spring 1992, , Monterey Co., CA (P= 0.000). (BHGR =Black-headed Grosbeak, COYE= Common Yellowthroat, SOSP= Song Sparrow, SPSA= Spotted Sandpiper, SWTH= Swainson s Thrush, TRES= Tree Swallow, WAVI= Warbling Vireo, WIWA= Wilson s Warbler, YWAR= Yellow Warbler) 21

22 Figure 8. Spring abundance of Song Sparrows increased with time on the lower Carmel River, Monterey Co., CA (n= 15, r 2 = , P= 0.008). 22

23 Summer Results Descriptive Statistics A total of 26,127 individuals of 109 species were recorded over 16 years from 1992 through 2007 (Appendix 6). Most individuals (3888) were noted from transect 4C, the Carmel River mouth highway 1 bridge while the least (1558) were recorded from transect 2B, Garland Park. Eight Riparian Focal Species were recorded during that same period: Black-headed Grosbeak, Common Yellowthroat, Song Sparrow, Swainson s Thrush, Tree Swallow, Warbling Vireo, Wilson s Warbler, and Yellow Warbler. For all years of study there was no change in the mean for all transects during the summer (r 2 =0.058, P= 0.370, Figure 9, Appendix 7). The average for the entire study was The average for all transects in 2007 was At one transect, 4C at the Carmel River mouth, increased significantly from during the summer while at 2 others, transects 3B Schulte Bridge and 3C Quail Lodge, increased marginally (at the P= 0.1 level) during that same period (Table 1, Figure 10). A significant difference was also detected between transects (P= 0.022, K=W= , df= 8, Figure 11). Transect 3C, Quail Lodge had the highest average in the summer, while transect 4C, Carmel River bridge was the lowest. A difference was detected between restored and reference with reference sites having higher than restored sites (Mann Whitney U= , df= 1, P= 0.056, Figure 12). There was no difference between transects in values depending on water availability (perennially watered vs. seasonally dry) during summer surveys. Riparian Focal Species Abundance From 1992 to present, abundance of Riparian Focal Species has increased significantly over time on the lower Carmel River in the summer (r 2 = 0.427, P= 0.006, n= 15, Figure 13). There was a marginal difference between transects (at the P= 0.1 level) in their average RFS abundance (K-W= , df= 8, P= 0.082, Figure 14). Transects 4B, Riverwood, and 2C,Carmel Valley Ranch Golf Club, had the highest average number of RFS recorded while transects 3A, Red Rock, and 2A, DeDampierre, had the lowest. No difference was detected in mean abundance of Riparian Focal Species between restored and reference transects (Mann Whitney U= 2000, X 2 = 2.798, df= 1, P= 0.094) or between perennially watered and seasonally dry transects (Mann Whitney U= , X 2 = 0.813, df= 1, P= 0.367). Linear regression of annual abundance by transect revealed a significant trend of increasing abundance of Riparian Focal Species with decreasing distance to the coast (r 2 = 0.455, P= 0.046, n= 9, Figure 15). Analyzed individually, among the 9 transects, none changed in abundance of riparian Focal Species over time (Table 4). Riparian Focal Species differed significantly in their mean abundance on the lower Carmel River in the summer (P= 0.000, K-W=37.491, df= 7). Song Sparrows were encountered most often while Common Yellowthroats were recorded the least (Figure 23

24 16). Of the eight RFS observed during the summer from 1992 through 2007, only 6 were recorded with enough regularity in the spring to analyze with regression. These included Wilson s Warbler (n= 170), Black-headed Grosbeak (n= 37), Song Sparrow (n= 629), Warbling Vireo (n= 57), Yellow Warbler (n= 43) and Swainson s Thrush (n= 13). Yellow Warblers and Swainson s Thrushes were analyzed using non-parametric tests due to non-normality of the data and low sample size respectively. Of these six, only Song Sparrows significantly increased in abundance from (r 2 = 0.350, P= 0.016, n= 16, Figure 17). There was no difference for any species between restored and reference sites nor between perennially watered and seasonally dry transects. Riparian Focal Species also did not show transect preferences, in that they were not find in higher abundance in some transects as compared to others in the summer. 24

25 Figure 9. Linear regression of summer mean for all transects on the lower Carmel River from 1992 through 2007 (r 2 =0.058, P= 0.370, n= 16), Monterey Co, CA. 25

26 Figure 10. Transects differed in their average values, included with standard errors, during the summer over the course of the study (P= 0.022, K-W=17.876, df= 8, , on the lower Carmel River, Monterey Co., CA. Table 3. Linear regression results of summer trends by transect (n= 16). Transect # Transect Name Regression (r 2 ) Significance (P) 2A DeDampierre Park B Garland Park C Carmel Valley Ranch Golf Club 3A Robinson Canyon Road/ Red Rock 3B Schulte Road bridge C Quail Lodge/ Valley Greens Drive bridge 4A Rancho San Carlos B Riverwood Complex C Carmel River mouth/ highway 1 bridge

27 Figure 11. Transect 4C values increased significantly from (r 2 = 0.342, P= 0.017, n= 16) during the summer on the lower Carmel River, Monterey Co. CA. 27

28 Figure 12. Box plot representation of difference in value means (P= 0.056) between restored transects and reference transects on the lower Carmel River from during the summer. 28

29 Figure 13. Linear regression of summer abundance of Riparian Focal Species for all transects on the lower Carmel River (n= 16, r 2 =0.427, P= 0.006) 29

30 Figure 14. Average abundance with standard errors of Riparian Focal Species by transect on the lower Carmel River during the summer from , Monterey Co, CA (P= 0.082). 30

31 Figure 15. Linear regression of summer abundance of Riparian Focal Species by transect for all years on the lower Carmel River revealed a trend toward increasing abundance as transects approached the coast (n= 9, r 2 = 0.455, P= 0.046), Monterey Co., CA. 31

32 Table 4. Regression table of summer abundance of Riparian Focal Species over time by transect on the lower Carmel River during the summer, Monterey Co., CA, Transect # Transect Name N Regression (r 2 ) 2A DeDampierre Park B Garland Park C Carmel Valley Ranch Golf Club 3A Robinson Canyon Road/ Red Rock 3B Schulte Road bridge C Quail Lodge/ Valley Greens Drive bridge 4A Rancho San Carlos B Riverwood Complex C Carmel River mouth/ highway 1 bridge Significance (P) 32

33 Figure 16. Mean count with standard errors of each Riparian Focal Species encountered on the lower Carmel River in the summer , Monterey Co., CA (P= 0.000). (BHGR =Black-headed Grosbeak, COYE= Common Yellowthroat, SOSP= Song Sparrow, SWTH= Swainson s Thrush, TRES= Tree Swallow, WAVI= Warbling Vireo, WIWA= Wilson s Warbler, YWAR= Yellow Warbler) 33

34 Figure 17. Summer abundance of Song Sparrows increased from on the lower Carmel River, Monterey Co., CA (n= 16, r 2 = , P= 0.016). 34

35 Distance Diversity and Abundance Analysis To compare this study to other similar studies, for the 2007 data, we restricted the point count distance analyzed to birds within 50 meters. In the spring of 2007, the average for all transects was In the summer, it was In 2007, there was no difference in values between transects for either spring (P= 0.433, K-W= 8.00, df= 8, Table 5) or summer (P= 0.433, K-W= 8.00, df= 8, Table 6). In the spring of 2007, 7 Riparian Focal Species were recorded within 50m of the point count locations: Black-headed Grosbeak, Song Sparrow, Swainson s Thrush, Tree Swallow, Warbling Vireo, Wilson s Warbler, and Yellow Warbler. During the summer of that year, 3 RFS were detected: Song Sparrow, Swainson s Thrush, and Wilson s Warbler. Song Sparrow on average was detected most often within the 5 minute count window in both seasons (P= K-W= , df= 6 [spring], P= 0.000, K-W = , df= 2 [summer], Table 7, Figures 18,19). In the spring, mean abundance values for Black-headed Grosbeak, Song Sparrow, and Warbling Vireo did not differ significantly from those reported in RHJV (2004, Table 7). However, five Riparian Focal Species were detected at significantly lower abundances in the spring than expected as compared to RHJV (2004) targets: Common Yellowthroat (P= 0.000), Swainson s Thrush (P= 0.000), Wilson s Warbler (P= 0.012), Yellow Warbler (P= 0.000) and Yellow-breasted Chat (P= 0.000). In the summer all Riparian Focal Species, except Song Sparrow, were detected in significantly lower numbers than expected (Table 7). There was no difference in number of detections depending on time of day in the spring (P= 0.106), but the difference was significant in the summer (P= 0.013) with more Riparian Focal Species detected in the morning as compared to the evening. 35

36 Table 5. values, species evenness and richness of birds detected within 50 meters of point count locations in 9 transects on the lower Carmel River during the spring of 2007, Monterey Co., CA. Transect Evenness Richness 2A B C A B C A B C Mean Table 6. values, species evenness and richness of birds detected within 50 meters of point count locations in 9 transects on the lower Carmel River during the summer of 2007, Monterey Co., CA. Transect Evenness Richness 2A B C A B C A B C Mean

37 Table 7. Average number of detections within 5 minutes and 50 meters of Riparian Focal Species on the lower Carmel River in the spring and summer of 2007, Monterey Co, CA. Means are compared to maximum breeding abundances reported in RHJV (2004) for the Central Coast. Average number of detections in 50m within 5 minutes RHJV (2003) Spring Summer Mean Mean SE Δ P Mean SE Δ P Black-headed Grosbeak Blue Grosbeak a 0.07 Common Yellowthroat Song Sparrow Swainson s Thrush Tree Swallow Warbling Vireo Wilson s Warbler Yellow Warbler Yellowbreasted Chat a Blue Grosbeaks have only been reported from the upper Salinas Valley and have never been recorded on the Carmel River (Roberson and Tenney 1993) 37

38 Figure 18. Riparian Focal Species average number of detections within 50 meters during the spring of 2007, lower Carmel River, Monterey Co. CA (P= K-W= , df= 6). (BHGR =Black-headed Grosbeak, COYE= Common Yellowthroat, SOSP= Song Sparrow, SWTH= Swainson s Thrush, TRES= Tree Swallow, WAVI= Warbling Vireo, WIWA= Wilson s Warbler, YWAR= Yellow Warbler) 38

39 Figure 19. Riparian Focal Species average number of detections within 50 meters during the summer of 2007, lower Carmel River, Monterey Co. CA (P= 0.000, K-W = , df= 2). (SOSP= Song Sparrow, SWTH= Swainson s Thrush, WIWA= Wilson s Warbler) 39

40 DISCUSSION The Shannon Diversity Index accounts for and combines species richness and evenness. The maximum theoretical achievable value for this index is 10.0 and would represent a system in which a high number of species are sustained in exactly equal numbers. values generally increase with increasing proximity to the tropics and increasing number of habitat niches. For the lower Carmel River, an of 4.5 is considered maximum (Thorngate and Scullen in prep). This value represents a theoretical state in which all riparian associated species found on the Central Coast are represented evenly and equally. Bird richness and evenness on the lower Carmel averaged 3.12 in the spring and 2.72 in the summer. These averages represent 69% and 60% respectively of the maximum theoretical value possible on the lower Carmel River. Although it is not expected that the Carmel River (or any other Central Coast river) reach the theoretical maximum, results from the lower Carmel River demonstrate a stable riparian system, but one at a capacity that may be lower than its potential. However, because of marginally significant differences between restored and reference sites, results here suggest this capacity may change as riparian restoration projects mature. In both spring and summer, restored sites were slightly lower than reference sites in values (and both were significant at the P= 0.01 level). As these restored sites mature over the next years values on the lower Carmel River should increase. The increasing summer trends noted from the Carmel River mouth and Schulte bridge, both restored since 1995, provide good evidence for this expectation. This project was initiated with the objective of looking for differences in avian abundance and diversity associated with seasonality of water flows on the lower Carmel River. This question was approached via timing of data collection (in late May and middle to late August). Results from the past 15 years have demonstrated that there are no seasonal differences in avian diversity and abundance between perennially watered and seasonally dry reaches. Water availability seems to not influence. Perennially watered and seasonally dry reaches were not compared during the spring, but analysis in the summer, when transects closest to the ocean are typically dry, showed no difference in values between transects with water and those without. In 2007, there was no water in any transect during the summer surveys; thus, comparison in avian diversity based on water availability should be based on the 2006 study which also reported no difference between perennially watered and seasonally dry transects (Thorngate et al. 2006). To accurately interpret RFS abundance data, results from summer should not be considered. The summer data collection period, in late August, is after the departure date for most of the Neotropical migratory species. Black-headed Grosbeaks begin the southward migration as early as the beginning of July (BSOL unpublished data, Hill 1995); Warbling Vireos typically leave in early August (Gardali and Ballard 2000) but have been reported from Big Sur Ornithology Lab leaving in some years as early as July 17 (BSOL unpublished data). Wilson Warblers are also reported to begin migration in 40

41 mid-august (Ammon and Gilbert 1999), and have been recorded in fall migration from BSOL at the earliest on August 19 (BSOL unpublished data). Swainson s Thrushes are known to begin their southward migration in early August (Mack and Yong 2000) and have been captured migrating in Big Sur as early as August 18. Finally, Yellow Warblers begin migration in middle July (Lowther et al. 1999, BSOL unpublished data). The spring point counts on this study were typically conducted in middle to late May. This is generally considered the pre-nesting period from migrating species, so not ideal from estimating population maximum, because birds are still arriving from wintering grounds and establishing territories. However, most individuals should be on territory and singing by late May. Results of Riparian Focal Species abundance on the lower Carmel were promising and also, revealed areas for habitat improvement. Abundance of RFS has increased on the lower Carmel in both spring and summer since 1992 suggesting that the availability of riparian habitat niches has expanded. However, this trend may just reflect the increase in Song Sparrows. Song Sparrows showed increasing trends over the past 15 years in both spring and summer; while all other RFS had no trends. Some species were underrepresented or only sporadically recorded since 1992 including Common Yellowthroat and Yellow-breasted Chat. Song Sparrows are year-round residents on the Central Coast and riparian obligates requiring a dense shrub layer close to fresh water for nesting (Arcese et al. 2002). They nest readily in a variety of disturbance regimes including early succession, clear-cut, urban, agricultural, non-native (eg. salt-cedar, Tamarix ramosissima) and mature habitats. Likewise, they require little in habitat structural complexity, needing only a shrub layer to thrive. They rarely move farther than 10 meters outside of this protective shrub layer to forage (Arcese et al. 2002). Clearly, the lower Carmel meets the relaxed habitat requirements for Song Sparrows. Common Yellowthroat, like Song Sparrows in that they require little structural complexity save thick vegetation along rivers, differ from Song Sparrows in that they require grasses and sedges, or other emergent marshy vegetation such as cattails for nesting. Nests are usually within 10 cm of the ground but above flood level and usually not directly above water (Guzy et al. 1999). There are few places on the Carmel River that provide this type of habitat. In spring 2001, 2 Common Yellowthroats were recorded from 2 transects: 4A Rancho San Carlos and 4C Carmel River bridge. Because May is early in their nesting season, these individuals may have been transients; nonetheless, banding data from a station approximately 300 meters downstream of the 4C transect confirms that Common Yellowthroats nest at the Carmel River mouth (Thorngate and Scullen 2007). Common Yellowthroats were not captured at any of the other 3 mist-net stations operated on the Carmel River in 2007 (Thorngate and Scullen in prep). Yellow-breasted Chats have never been recorded in this study in either season. It is possible that due to the constraints of the point count period employed on this study, breeding chats were missed. Chats are known to be late arriving migrants with the bulk of spring movement occurring in May (Eckerle and Thompson 2001). Even so, chats 41

42 have been recorded on territory in Monterey County as early as 19 April (Roberson and Tenney 1993). Yellow-breasted Chats were also absent in another study from the Carmel River conducted throughout the breeding season in 2007 (Thorngate and Scullen in prep). Chats were confirmed breeding on the Carmel River as recently as 1993, but even then noted to be in decline (Roberson and Tenney 1993). In addition to requiring riparian habitat, chats are an open-canopy obligate, preferring early succession and forest edges and openings for nesting. Thriving populations are associated with dense shrub (>4.5 meters tall) with little to no overstory (Eckerle and Thompson 2001). Although they have a habitat preference superficially similar to the Song Sparrow s tolerance for disturbance, chats require large patches of interlocking shrubs (>5.4 meters wide) and tall shrubs for successful nesting. Further, throughout their range, chats suffer a high depredation rate (Eckerle and Thompson 2001) Distance analysis using only species heard within 50 meters in 2007 allowed for comparison of the results from the Carmel River to other Central Coast studies. values for 2007 inclusive of all distances although not significantly different were lower than the average for all 15 years. A lower indicates that fewer species were recorded overall or that some species were recorded in disproportionately lower or higher numbers compared to others. Lower overall values for 2007 coincide with other avian studies from Monterey County and were likely associated with the 2007 draught (Thorngate 2007). It is within this context that the single year distance analysis should be interpreted. Again, due to the timing of migration, data from the summer should be disregarded. Comparison of 2007 results from the Carmel River against RHJV maximum breeding abundances for the Central Coast revealed that the Carmel River supports healthy numbers of Song Sparrows, Black-headed Grosbeaks and Warbling Vireos. Riparian Focal Species that were significantly low on the lower Carmel were Common Yellowthroat, Swainson s Thrush, Wilson s Warbler, Yellow Warbler and Yellowbreasted Chat. Tree Swallows were not included in this analysis due to lack of RHJV data. Spotted Sandpipers were not included because they have only been recorded breeding on the Salinas River. Individuals seen on the Carmel in May are generally considered migrants (Roberson and Tenney 1993). In addition to shrubs lining the river that support Song Sparrows, the Carmel also has adequate edge habitat to meet the habitat requirements of Black-headed Grosbeaks (Hill 1995). Although considered to occupy a diversity of habitats, Black-headed Grosbeaks typically build their nest in trees close to a river, about 3 meters above the ground (Hill 1995). The presence of good numbers of Warbling Vireos in this study suggests that the lower Carmel possesses some tall, large deciduous trees, their preferred nest sites. Warbling Vireos have little else in habitat requirements and are found in open and semiopen canopies and in a variety of shrub types from grass to dense shrub (Gardali et al. 2000). Investigation of habitat requirements for RFS with lower than expected abundances on the lower Carmel can elucidate focal areas for habitat enhancement. As mentioned 42