Breeding Bird Survey Results 2007 Stoller-ESER-111

EXECUTIVE SUMMARY

Annual breeding bird surveys have been conducted on the Idaho National Laboratory (INL) since 1985 (no surveys were conducted in 1992 and 1993) to monitor bird populations. In 2007, surveys were conducted in late May and early June along 13 permanently established routes, five of which are part of the U.S. Geological Survey’s nationwide program and eight around facilities specific to the INL. A total of 5,412 birds were observed during the 2007 survey which is slightly above the 1985-2007 average of 4,991 birds. A total of 69 different species were documented during the 2007 BBS which is above the average of 59 species recorded since 1985.

Similar patterns of species abundance were found with horned lark (Eremophila alpestris; n = 1416) the most abundant species in 2007 and previous years. Western meadowlark (Sturnella neglecta; n = 987), Brewer’s sparrow (Spizella breweri; n = 528), sage sparrow (Amphispiza belli; n = 459), and sage thrasher (Oreoscoptes montanus; n = 410) constitute the next four most abundant species. In the 21 years of INL breeding bird surveys, these five species have been the most abundant 16 times, and are among the six most abundant species recorded in all years. Considering the declines observed in sagebrush obligate species throughout the intermountain west, this trend of abundance suggests the quality of INL sagebrush steppe habitat remains high.

The 2007 survey yielded four species never recorded during the INL survey. A single observation of a greater scaup (Aythya marila), bald eagle (Haliaeetus leucocephalus), osprey (Pandion haliaetus), and turkey vulture (Cathartes aura) was made this year. This is the first documented field observation of a bald eagle observed on INL during spring.

Species observed during the 2007 BBS that have been assigned an Idaho state ranking of critically imperiled or imperiled include greater sage-grouse (Centrocercus urophasianus; n=4), ferruginous hawk (Buteo regalis; n=8), long-billed curlew (Numenius americanus; n=7), Franklin’s gull (Larus pipixcan; n=178), Brewer’s sparrow (n=528), and the grasshopper sparrow (Ammodramus savannarum; n=102).

INTRODUCTION

The North American Breeding Bird Survey (BBS) was developed by the U.S. Fish and Wildlife Service (USFWS) in conjunction with the Canadian Wildlife Service to initiate the collection of data to document bird population trends. Pilot surveys began in 1965 and immediately expanded to cover the U.S and Canada east of the Mississippi and by 1968 included the entire North American Continent (Bystrak 1981, Robbins et al. 1986). The North American BBS program is managed by the U.S. Geological Survey (USGS) and currently consists of over 4100 routes with approximately 3000 sampled each year.

BBS survey data continue to provide some of the only long-term bird population information covering broad geographic scales. These data have been used to estimate population changes for hundreds of bird species, and remain the primary data source for regional conservation programs and modeling efforts (Sauer et al. 2003). Numerous statistical pathways for exploring and analyzing BBS data have been proposed and discussed amongst researchers (James et al. 1996, Link and Sauer 1997, McCulloch et al. 1997, Bart et al. 2004, Sauer et al. 2005). Regardless of differences in opinion concerning the most appropriate analysis techniques, the BBS undeniably provides a wealth of information about North American bird population trends that form the foundation for broad conservation assessments extending beyond local jurisdictional boundaries.

The Idaho National Laboratory (INL) has five permanent official BBS routes originally established in 1985 (hereafter referred to as remote routes) and eight additional survey routes around INL facilities (hereafter referred to as facility routes). The facility routes were developed to monitor avifauna populations in proximity to anthropogenic activities and disturbances. The annual BBS survey provides land managers with breeding bird population trend information relative to the activities conducted on the INL. This report summarizes the results from the 2007 BBS survey, and compares observed species abundance across survey routes with long-term averages.

STUDY AREA

The INL is a Department of Energy facility encompassing almost 900 mi² (2315 km²) located on the Upper Snake River Plain in southeast Idaho (Figure 1). The INL was designated as a National Environmental Research Park in 1975 to facilitate field research assessing environmental impacts from nuclear energy development technologies. The INL lies within portions of Bingham, Bonneville, Butte, Clark, and Jefferson counties.

Topography across the INL is mostly flat with an average elevation of 4985 ft (1520 m) above sea level. Other than minor topographic variation created by basalt outcrops, the only significant relief occurs around East and Middle Buttes and the southern extent of the Lemhi mountain range located on the northwest corner of the INL.

Anderson et al. (1996) provide a detailed description of the climate, geology, and vegetation communities found on the INL. In general, the INL is located in a semi-arid desert that experiences hot, dry summers and cold winters. Annually the INL receives on average eight inches (20 cm) of precipitation with a peak common in the spring. The geology is dominated by Quaternary basalt lava flows producing the outcrops and lava tubes found across the site today. Aeolian soils consisting primarily of silt loam and sandy loam are the most common soil type found throughout the INL, while alluvial soils are more commonly found along the Big Lost River flood plain. The INL is a shrub-steppe ecosystem dominated by a woody shrub overstory and perennial bunchgrass and forb understory. Wyoming big sagebrush (Artemisia tridentata wyomingensis) is the most dominant shrub on the INL, while other sagebrush species, green rabbitbrush (Chrysothamnus viscidiflorous), spiny hopsage (Grayia spinosa) shadscale (Atriplex confertifolia) and winterfat (Krascheninnikovia lanata) can be commonly found. The most common native grasses include thickspiked wheatgrass (Elymus laceolatus), bottlebrush squirreltail (Elymus elymoides), Indian ricegrass (Achnatherum hymenoides), and needle-and-thread grass (Hesperostipa comata).

Very little surface water is present on the INL during the spring and summer. The Big Lost River and Birch Creek drainages are both diverted upstream for agricultural purposes and consequently little if any stream water reaches the INL. During years of high flow volume the Big Lost River can reach the INL and drains into an ephemeral wetland known as the Lost River Sinks. The Lost River Sinks wetland provides the only substantial water source for waterfowl and shorebirds on the INL, however a number of man-made waste treatment ponds near facilities also provide aquatic habitat for migrating birds.

The Idaho Comprehensive Wildlife Conservation Strategy (CWCS) recognizes wildlife species that are listed by either State or Federal agencies and provides a comprehensive listing of the Idaho Species of Greatest Conservation Need (SGCN) (Appendix B: Idaho Department of Fish and Game 2005). The CWCS also identifies Idaho Important Bird Areas (Appendix H: Idaho Department of Fish and Game 2005), and the INL has received such designation.

METHODS

Data Collection

The BBS is a roadside count of all birds seen or heard along predefined routes. Thirteen BBS routes (Figure 1) were surveyed between May 22 and June 12, 2007, consisting of five official USGS BBS remote routes in addition to eight facility routes developed specifically for the INL. Each remote survey route is 24.5 miles (39.2 km) long with 50 systematically spaced sampling locations located every 0.25 miles (0.4 km). The facility routes vary in total length depending on the size of the facility, and sampling locations are spaced approximately 0.2 miles (0.4 km) apart.

The North American BBS protocols provided by the USGS Patuxent Wildlife Research Center were followed during these surveys. At each sampling location, a trained observer records every bird species visually observed within a quarter mile radius or heard (song) during a 3-minute time period. Any bird that was suspected of being counted on the previous stop was not recorded a second time at the next sequential stop. A number of additional data such as temperature, wind speed, and sky condition are recorded at the beginning and end of each survey route and each route is only surveyed when weather conditions are appropriate (e.g., no heavy rains or strong wind). BBS surveys begin a half-hour before sunrise and can continue for up to 6 hours until the route is complete. The total number of cars that pass during the 3-min sampling period are recorded on all remote routes, and if background noise becomes loud enough to interfere with audible detection it is also noted.

Bird Abundance Correlation

In previous INL BBS reports, environmental abiotic factors were investigated in an attempt to explain the variation in observed bird abundance. Belthoff et al. (1998) found a relationship between cool and wet June weather and bird abundance for 1985-1991. Belthoff and Ellsworth (1999) found that bird abundance was significantly negatively correlated with mean June temperature, where higher bird abundance corresponded to lower temperatures. A relationship between bird abundance and June precipitation was noted, and although not statistically significant, the removal of an outlier from 1995 would have resulted in a significant p-value (Belthoff and Ellsworth 1999). Belthoff and Ellsworth (1999) used the Spearman rank correlation coefficient to identify correlations with June temperature and precipitation, and the same statistic was recalculated this year to compare current relationships in the 2007 abundance data.

The Spearman rank correlation is a non-parametric test used to investigate the relationship between variables (Spearman 1904). Instead of using the raw abundance data, both variables are ranked in increasing order and the assigned ranks are used in the statistical analysis. The Spearman rank correlation coefficient (r_s) is calculated using the following equation, where (d) is the difference between the ranks and (n) is the sample size.

In cases where two data values have a tied rank, a different equation is used to account for the tied ranks (Thomas 1989). The first equation is calculated for both variables (x and y) where (t_i) is the number of tied values, and the second equation calculates the Spearman Rank Correlation coefficient corrected to rank ties (r_s)_c .

Mean June temperature and total June precipitation data collected since 1985 at the Central Facilities Area (CFA) were used to test bird abundance correlations. Statistical significance was calculated using a two-tailed hypothesis with an alpha of 0.05. The June precipitation data contained a tied ranking, and the corrected equation described above was calculated for these data.

Community Diversity Indices

Diversity describes the number of interacting organisms in an ecological system and is commonly defined by species abundance and richness. A community with low species diversity may be indicative of an unhealthy or improperly functioning community. Higher species diversity is often interpreted as a stable, functioning system and increasing diversity is the goal of many management activities.

Species diversity indices are a mathematical way to quantify community composition. There is a number of diversity indices commonly used in ecology and each has particular strengths depending on the data and the questions. The simplest estimate of community diversity is species richness, which represents the total number of unique species present. Although species richness is a useful measure of diversity it does not account for differences in abundance between communities. For example, if there are many species with a single individual observed, richness will be high but may not be comparable to another community with the same number of species and large abundances across all species. Diversity indices that consider both species richness and species abundance may provide a more useful measure of community diversity.

One of the most popular diversity indices used by biologists is the Shannon diversity index (H) (Shannon 1948). Shannon’s diversity index takes into account both species richness (S) and relative abundance of each species present in the community. Shannon’s diversity index is derived by first calculating the proportion of species i relative to the total number of species (p_i), and then multiplied by the natural logarithm of this proportion (lnp_i). The resulting product is then summed across species and multiplied by -1. Shannon’s H can range from zero to about 4.6 with higher values representing increasing diversity.

Another useful measure that can be derived is Shannon’s Equitability (E_H). Shannon’s Equitability represents a measure of evenness which is how similar species abundance is among the community. E_H ranges from 0-1 with one representing a completely even community where all species abundances are equal.

Shannon’s H and E_H were calculated for all BBS routes, and compared to standard species richness information reported in past reports. Each survey route was considered as a representation of the local bird community, and community diversity index values reflect BBS route diversity.