Evaluation of Mesquite Mouse (Peromyscus merriami) Status in Pima County Arizona Game & Fish Department Heritage Grant Project #IO5001 May 2006 PIMA COUNTY BOARD OF SUPERVISORS Richard Elías, Chairman, District 5; Ann Day, District 1 Ramón Valadez, District 2; Sharon Bronson, District 3 Raymond J. Carroll, District 4 COUNTY ADMINISTRATOR C.H. Huckelberry mesqmousecvr506.ai•db•20157 EVALUATION OF MESQUITE MOUSE (Peromyscus merriami) STATUS IN PIMA COUNTY, ARIZONA Prepared for PIMA COUNTY REGIONAL FLOOD CONTROL DISTRICT Water Resources Division Public Works Center 201 North Stone Avenue, 4th Floor Tucson, Arizona 85701-1215 Prepared by Kenneth J. Kingsley, Ph.D. SWCA® ENVIRONMENTAL CONSULTANTS 343 West Franklin Street Tucson, Arizona 85701 (520) 325-9194 www.swca.com May 2006 DISCLAIMER The findings, opinions, and recommendations in this report are those of the investigators who have received partial or full funding from the Arizona Game and Fish Department Heritage Fund. The findings, opinions, and recommendations do not necessarily reflect those of the Arizona Game and Fish Commission or the Department, or necessarily represent official Department policy or management practice. For further information, please contact the Arizona Game and Fish Department. 2 TABLE OF CONTENTS ABSTRACT ..................................................................................................................... iii INTRODUCTION............................................................................................................. 1 Purpose And Objectives ........................................................................................... 1 BACKGROUND .............................................................................................................. 1 Species Status.......................................................................................................... 1 Taxonomy ................................................................................................................. 1 Distribution................................................................................................................ 2 Identification.............................................................................................................. 2 Ecology ..................................................................................................................... 5 Threats to the species............................................................................................... 5 METHODS ...................................................................................................................... 6 Historical Review ...................................................................................................... 6 HABITAT ASSESSMENT................................................................................................ 6 SELECTIVE TRAPPING EFFORT .................................................................................. 7 Site Selection............................................................................................................ 7 Trapping Methods ..................................................................................................... 7 Identification............................................................................................................ 10 RESULTS AND DISCUSSION...................................................................................... 12 Historical Review .................................................................................................... 12 HABITAT ASSESSMENT.............................................................................................. 13 SELECTIVE TRAPPING EFFORT ................................................................................ 15 CONCLUSIONS ............................................................................................................ 17 ACKNOWLEDGEMENTS ............................................................................................. 19 LITERATURE CITED .................................................................................................... 20 APPENDICES IN SEPARATE DOCUMENT APPENDIX A POTENTIAL TRAPPING LOCATIONS ...................................................AAPPENDIX B SITES TRAPPED ...................................................................................BAPPENDIX C FIELD KEY TO RODENTS LIKELY TO BE TRAPPED ......................... C- ii ABSTRACT The mesquite mouse, also known as Merriam’s mouse (Peromyscus merriami Mearns), is a cryptic rodent that closely resembles a common species, the cactus mouse, P. eremicus. Mesquite mice are extremely difficult to distinguish from cactus mice. Traditionally, morphological characteristics such as size of the animal and several of its body parts and skeletal characteristics, particularly of the skull and baculum (penis bone), have been used to distinguish the species. More recently, techniques involving analysis of DNA have been applied to distinguish the species. Both species are found in southern Arizona and northern Mexico. Most of the known specimens of the mesquite mouse were collected from mesquite (Prosopis velutina) bosque (woodland) areas in Pima County. Concern has been expressed by various authors and organizations that mesquite bosques are disappearing or diminishing in quality, and that the survival of the mesquite mouse is consequently threatened. This study, funded by a grant from the Arizona Heritage Fund to the Pima County Regional Flood Control District, investigated the current distribution of the mesquite mouse by reviewing historical records from museum specimens, examining existing potential habitat at historic locations of the species and elsewhere in the county, and conducting a live-trapping study at 19 sites representing a range of ecological communities with mesquite as a major component. Mesquite mice, as determined by DNA analysis and morphological characteristics, were found at 10 out of 19 sites trapped. They were found at or near six out of eight historic locations. Mesquite mice were found in a variety of mesquite-dominated areas, representing the following vegetation classifications: • 143.10, Semidesert grassland, xeroriparian scrub • 154.100, Sonoran desertscrub xeroriparian scrub • 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series All have mesquite as the dominant tree at the locations trapped, but do not appear to be exclusively remnants of old bosques. Mesquite mice were not found in: • isolated patches of mesquite surrounded by urban development; • narrow, rocky washes with few mesquites; or • mesquite-invaded grassland or upland vegetation. They were found in sites subject to a variety of human caused impacts, but the effects of these impacts cannot be determined without further specific study. iii INTRODUCTION PURPOSE AND OBJECTIVES The purpose of this project was to evaluate the current status of the mesquite mouse (Peromyscus merriami) in Pima County to identify areas where this species may still exist and can be protected. The objectives of this project were: 1) 2) 3) 4) Examine historic records and museum specimens for best available information and interview scientists who have trapped the mouse historically; to determine locations and conditions where this species has been found in the past and gain information about where to look for it today; Evaluate conditions at accessible historic locations and conduct a live-trapping survey if conditions are appropriate; Identify the best remaining mesquite forests and woodlands on public lands in Pima County and conduct live trapping surveys on lands for which permission can be obtained; Prepare a report summarizing findings, including maps and photographs of areas surveyed, narrative descriptions of habitat conditions, and management recommendations. BACKGROUND SPECIES STATUS P. merriami has no status under the Federal Endangered Species Act, and no special status under Arizona law or regulations. It is one of the 56 species considered by Pima County to be a priority vulnerable species under the Sonoran Desert Conservation Plan. The Arizona Game and Fish Department lists the mesquite mouse as a “sensitive element” and identifies specific study objectives for grant proposals: 1. Evaluate historic and current occurrence through field surveys and searches of pertinent literature and museum records; 2. Assess habitat use, population status, and population trends; 3. Identify management needs. TAXONOMY Peromyscus merriami is in the class Mammalia, Order Rodentia, Family Muridae, Genus Peromyscus, subgenus Haplomylomys. The species was named in 1896 by Mearns from specimens collected at Sonoyta, Sonora, Mexico during the international boundary survey. Osgood (1909) revised the genus, and placed merriami as a synonym of P. eremicus, noting that the specimens were apparently just larger individuals of the common species. The species was resurrected by Hall and Kelson (1952). Subsequent authors have accepted the specific identity of P. merriami based on these earlier works (Hooper 1968, Lawlor 1971, Hoffmeister 1986, Baker et al. 2003). More recent evidence, in the form of genetic and DNA analysis, clearly indicates that P. merriami is a distinct species with two subspecies merriami and goldmani (Avise et al. 1974; Riddle et al. 2000). 1 DISTRIBUTION P. merriami is found from south-central Arizona southward to Sinaloa, Mexico. P. merriami is known in Arizona from “just north of Florence” on the Gila River, southward to Lukeville on the west and southeast of Tucson on the east (Hoffmeister 1986). The majority of known records of the species in Arizona are from Pima County, along the Santa Cruz River and its tributaries. IDENTIFICATION Identification of this species is difficult based on morphology alone. Mearns (1907) separated P. merriami from P. eremicus on the basis of several characteristics (1) hind foot more than 22 mm., (2) skull length more than 26 mm., (3) total length more than 210 mm, and (4) chest and wrists usually colored. Arnold (1940) in a M.S. thesis that was an ecological study of the vertebrate animals of the mesquite forest stated (p. 38): “Peromyscus merriami Benson, Mesquite White-footed Mouse, was the most consistent mammalian inhabitant of the mesquite forest proper.” He stated (p. 38) “this mouse has been designated by Benson as a new species to go under the name of Peromyscus merriami rather than Peromyscus eremicus as previously supposed. According to him it is identical to the typical mesquite inhabiting Peromyscus which he is naming from northern Mexico.” Arnold did not give a detailed description of the species. Apparently Benson did not publish the species, probably because he discovered that it had previously been published. No work by Benson was fully cited by Arnold or any other author with regard to this species. Commissaris (1960) conducted a study of Peromyscus at an area described as 9 miles south of Tucson. This area includes Sahuarita Butte (currently known as Martinez Hill), a rocky lava hill which rises 325 feet above the valley floor, and a large mesquite forest along the south and west sides of the butte. (The area is on Tohono O’odham Nation land and the large mesquite forest is gone, although many smaller mesquites remain). He trapped both the rocky hill and the mesquite forest during the winter of 1956-57. Forty-five specimens, 23 from the mesquite area and 22 from Sahuarita Butte, were compared in his study. He noted slight external differences between freshly captured specimens collected in each habitat. Specimens trapped in the mesquite area were larger, and a “cinnamon” pectoral spot was often present. This pectoral spot was quite variable, however, with some specimens having large spots and others having none. Occasionally the cinnamon pectoral spot appeared in specimens from the butte. Commissaris cited Mearns (op. cit.) who pointed out that “The two have the same coloration, except that P. merriami is somewhat darker, has the light cinnamon color extending down the outer side of the forelimb to the hand, and a large pectoral patch of cinnamon that is usually wanting in P. eremicus.” With the exception of the pectoral area, the differences in coloration were slight, however, and often disappeared or were obscured when the animal was prepared as a study skin. Commissaris measured several features of the animals he captured. He found no sexual dimorphism in either form. He found the following differences between populations, which he presented in a table: 2 Table 1. A comparison of 22 specimens of Peromyscus eremicus pullus from Sahuarita Butte with 23 specimens of Peromyscus merriami from the adjacent mesquite forest; measurements in millimeters (from Commissaris 1960) Measurement Peromyscus eremicus Peromyscus merriami T Mean with Standard error Mean with Standard error P< Total length 181.4 + 2.04 193.0 + 2.66 3.46 .01 Tail length 96.9 + 1.53 110.3 + 1.78 1.45 .20 Body length 84.5 + 1.18 92.2 + 1.49 4.05 .001 Hind foot length 19.9 + 0.16 22.2 + 0.18 9.54 .001 Height of ear from notch 19.0 + 0.29 21.6 + 0.19 7.50 .001 The other measurements were skull characters, which are not included here because specimens were not collected in the present study. He examined five adult specimens from the type series of P. merriami. These showed all the characters he found in the mesquite forest form of Peromyscus in his study area. Commissaris concluded: “it seems reasonable to conclude that Peromyscus merriami is a distinct species from Peromyscus eremicus, differing both in morphology and ecology, and that it is found inhabiting the mesquite infested lowlands of the southern part of Arizona and adjacent areas.” Hoffmeister and Lee (1963) made a detailed study of specimens from throughout the range of what they considered to be P. merriami. They stated that P. merriami is very similar to P. eremicus, “the similarities are so great that specimens of the two kinds are easily confused.” “The most diagnostic characters are size and shape of baculum, length of hind foot, length of ear and mastoid breadth. Total length of animal, length of skull and coloration are sometimes useful” (Hoffmeister and Lee 1963 p. 201). They discussed the history of classification of P. merriami, and pointed out that the previously published works were based on examination of very few specimens. They acknowledged that Commissaris (1960) was accurate in his conclusion, despite the small number of specimens he examined. They amplified their review of that paper, pointing out that there is little doubt from the data presented by Commissaris that real differences exist between the two kinds of mice. Although these differences were statistically significant, that did not necessarily mean that they were of great magnitude. In fact, the characteristics pointed out by Commissaris were not all consistently found through examination of more specimens from a wider range of localities. They determined that the differences were relatively small and nearly all characters showed considerable overlap between the two kinds. They examined 191 specimens possessing a combination of characters that caused them to be referred to as P. merriami. No single character was found to be always diagnostic. They arrived at a somewhat different evaluation of known characters than Commissaris had, and found additional differences between the two forms. In order of decreasing diagnostic value these features were grouped as follows: (1) size and shape of baculum; (2) length of hind foot, length of ear, mastoid breadth; (3) size and shape of skull, total length and length of tail, coloration of abdomen. Hoffmeister (1986) supported the conclusions of Hoffmeister and Lee (1963). Length of hind foot was suggested by Hoffmeister and Lee (1963) as a good diagnostic character among certain populations. They found that the average length of the hind foot in 30 specimens of P. merriami was 22.37 mm, SD 0.93, compared to the 22.2 mm + 0.18 found by Commissaris. Hoffmeister and Lee (1963) also found length of ear to be diagnostic among 3 some populations, but there was considerable overlap between P. merriami and certain other populations of P. eremicus. They found length of ear (as measured by the collector of the specimen) for 29 specimens of P. merriami to average 21.03 mm, SD 1.12. Commissaris’s mean was 21.6 mm. Total length and length of tail were stated to average larger in P. merriami than in P. eremicus but were of less diagnostic value. In P. merriami, Hoffmeister and Lee (1963) found total length to be usually more than 190 mm and tail length to be usually more than 100 mm in young adult animals and total length more than 195 mm with tail length more than 104 mm in older adults. Hoffmeister and Lee (1963) created a table that included assignment of individual specimens for each of five characters. Two of these were skull measurements. Dividing points of characters were assigned, with P. merriami being distinguished by the following measurements that can be taken from living specimens: Length of hind foot: >21.6 Length of ear: >20.03 Total length: >191.5 Hoffmeister and Lee (1963) found some differences in coloration between specimens of the two species. As noted by Commissaris, a well-marked pectoral spot, described by Hoffmeister and Lee as “ochraceous or tawny color” (differing from the “cinnamon” of Commissaris and Mearns) was usually present in P. merriami; in P. eremicus in Arizona, specimens usually lacked a pectoral spot. In addition, they noted that most P. merriami had the underparts washed with a creamy color. In P. eremicus, the underparts did not have this creamy tinge. They concluded that this character, although sometimes very subtle, is diagnostic in most specimens. Hoffmeister and Lee (1963) discussed the magnitude of these differences with regard to whether or not the two forms should be considered different species. They noted that the differences between P. merriami and P. eremicus were of the magnitude, if not entirely the kind, to be expected between subspecies. They stated that they were aware that some subspecies might be more distinctive than some species. If the specimens regarded as P. merriami were members of a marginal, allopatric group of populations, they probably would be regarded as subspecies of P. eremicus, and individuals that were intermediate in several characters would be considered as intergrades. However, the two forms were known to be generally sympatric but mutually exclusive in microhabitat selection at many localities. Because of their sympatric distribution as discontinuous units, and their morphological differences, Hoffmeister and Lee concluded that P. merriami should be regarded as a species different from P. eremicus. Riddle (personal communication) stated that the differences between P. merriami and P. eremicus are clear and easily detected at the molecular level, but that differentiation by morphology alone is difficult or impossible for every specimen, and there are probably many specimens of both species that have been incorrectly identified in museums. 4 ECOLOGY Mearns (1907) reported that P. merriami was fairly common at Sonoyta, Sonora, in fields and brush fences. Arnold (1940) stated that although it was never found in any great numbers it was trapped in his study area wherever mesquite was present regardless of the other types of vegetation. He noted that this mouse seemed to be quite as much at home among the branches of the mesquite trees as on the ground. Commissaris (1960) trapped P. merriami in his mesquite study area and never in the adjacent paloverde dominated vegetation. He concluded that there appeared to be specific habitat requirements for the two forms. He cited Mearns (1907) and Dice and Blossom (1937) stating that P. merriami was reputed to be a lowland dweller occurring in mesquite communities and fields while P. eremicus was abundant in rocky areas. Hoffmeister and Lee (1963) concluded that P. merriami occupied an ecological niche within its geographic range characterized by heavy stands of mature or large mesquites with a minimum of rocks and slope. In one specific attempt to obtain specimens of P. merriami, they made intensive collections in this type of habitat. Among 93 individuals collected, 87 proved to be of this species while only 6 were P. eremicus. Hoffmeister (1986 p. 344) stated that “in my estimation the name mesquite mouse for the species is far more appropriate than Merriam's mouse. Everywhere in Arizona and northern Sonora this mouse is found in mesquite forests called bosques.” He noted that the distribution of these mice corresponded rather closely with part of Nichol's "Mesquite and salt-bush bottoms." Near San Xavier Mission, Pima County, he trapped these mice along a creek bed with large numbers of immense mesquite. At Wilmot Station southeast of Tucson, they were trapped in thick stands of mesquite, cholla, prickly pear, paloverde, and grasses. THREATS TO THE SPECIES Mesquite bosques have declined in some parts of Arizona and northern Mexico. Areas of this community type are said to have historically represented one of the most abundant riparian communities in the Southwest, but are now reduced to remnants of what they once were. Threats that have been identified include: groundwater pumping and surface water diversion; land clearing and wood cutting; and land uses such as livestock grazing and recreation (Stromberg 1993). Hoffmeister (1986) expressed concern that many of the stands of mesquite known to have been inhabited by mesquite mice were being cut for firewood and to make more grazing land available. Thus, he concluded, the fate of mesquite mice in Arizona was precarious. The obvious loss of certain mesquite bosques, combined with the collection of very few specimens identified as P. merriami in recent decades, has inspired concern for the survival of the species among some mammalogists (Hoffmeister 1986; Y. Petryszyn, E.L. Cockrum, and R.R. Johnson, personal communications). 5 METHODS HISTORICAL REVIEW We compiled the available information on this species, using references in Hoffmeister (1986), Hoffmeister and Lee (1963), and Commissaris (1960) as a starting place. We also consulted other publications and unpublished reports of small mammal trapping in southeastern Arizona for records of this species. We contacted Arizona Game and Fish Department (AGFD) biologists Tim Snow, Bill Van Pelt, and Lin Piest and asked for information from their studies and from others of which they were aware. We interviewed Dr. Yar Petryszyn, Curator of the Mammal Museum of the University of Arizona, on two occasions, and he allowed examination of specimens in the museum and field notes of collectors. We interviewed Dr. Brett Riddle, University of Nevada, Las Vegas, by telephone and email. Dr. Riddle is co-principal investigator on a National Science Foundation funded study of the genetics and distribution of species of the genus Peromyscus, and has the distinction of being the only person known to have trapped and positively identified P. merriami in Pima County in this century. We utilized online resources and published museum records to develop a list of museums outside Tucson known or believed to have specimens of P. merriami. We corresponded with mammal curators at those museums and requested records of their specimens of this species, or we obtained the records from online catalogs. Museums that had specimens graciously provided information on them. All specimen records were compiled into an MS Excel spreadsheet that is believed to contain records of all specimens of P. merriami originating in Pima County, Arizona. This was then condensed to Appendix A, which lists all of the known historic locations at which specimens identified as P. merriami has been trapped, and used to prepare the GIS data layer and to plan sites to investigate. All identifications of P. merriami for museum specimens were tentatively accepted as accurate, but it must be noted that some may be inaccurate and it is not possible to determine which, if any, these may be. HABITAT ASSESSMENT Location data for sites at which specimens identified by collectors as P. merriami were trapped were then compiled into a Geographic Information System (GIS) database layer, under the direction of John Regan, Pima County Department of Transportation. The compiler used the best available base maps for the decades in which specimens were trapped to plot the locations as accurately as possible. Precise plotting was confounded in some cases by lack of precision in specimen locality data recorded by the collector and/or transcriber. Locality data were then plotted on a map of tentatively identified potential habitat for the species (Figure 1). Potential habitat was based on the description of habitat in Hoffmeister (1986). An existing GIS layer of data on vegetation types (Harris et al. 2000) provided the basis for tentative identification of Mesquite Series, Sonoran Deciduous Riparian Woodland as prime potential habitat for P. merriami. Where the location data from specimens did not coincide well with the Mesquite Series vegetation data layer, we made a visual search of aerial photographs and vegetation maps to identify the nearest wash or extensive mesquite-dominated area to the described site. We developed a working list of potential trapping locations that integrated information on historic trapping data, mapped vegetation type, land ownership, and access. 6 SELECTIVE TRAPPING EFFORT SITE SELECTION We selected a total of 19 sites for trapping. Criteria for selection of the sites was a combination of: • Proximity to historic locations at which mesquite mice had been trapped; • Access by virtue of location on public lands and proximity to roads; • Variation in vegetation within the range of mesquite-dominated woodland; • Geographic spread through the historically known range in eastern Pima County; • Convenience for trapping two or more sites in a night. Detailed descriptions of the sites selected and the results obtained there are included in Appendix B and sites are indicated in Figure 1. Precise sites of historic records for mesquite mice were occasionally problematic to relocate because boundaries have changed over time or were recorded imprecisely by the original collector. Many historic sites did not fit the above criteria for site selection, including those that are on private or Tohono O’odham Nation land. General locations for trapping were determined using the historic collectors’ label information. Then specific sites for setting traps were determined by land ownership, access, and current conditions in the field. Occasionally, more than one trapping site was located within a single historic general location (e.g., three trapping sites were located within a general area described as 2 mi N and 4 mi W of Arivaca, but each site had different vegetation and soil conditions). Limitations of time and budget precluded a wider or more intensive sampling. Table 2 summarizes the trapping sites used in this study and their relationship to historic locations, the results of this study, the number of trap-nights for each site, and the number of mesquite mice caught at each site. TRAPPING METHODS At each site, we placed Sherman live traps baited with oats mixed with peanut butter. We put two or more cotton balls in each trap to provide insulation for animals. We set traps singly at approximately ten meter intervals along a more-or-less straight line through the mesquitedominated habitat or along the wash, as appropriate. The number of traps placed varied with the site, and is described in Appendix B and summarized in Table 2. Generally 25 to 50 traps were set at each site. We set traps in the afternoon, and retrieved them the following morning. We took standard measurements (total length, tail, hind foot, and ear) for all Peromyscus species captured, and digital photographs of most of the animals. For most, but not all, individuals tentatively identified as possible P. merriami, we took tail and ear clippings using sharp scissors. These were placed in ethyl alcohol and submitted to the laboratory of Dr. Brett Riddle for DNA analysis. We did not take clippings from animals that appeared to be P. eremicus. We released all animals at the trap site at which they were captured following field processing. There was no accidental mortality of Peromyscus, and no animals were killed, therefore we collected no whole animal specimens. 7 Table 2. Summary of historic and current locations trapped during this study. Appendix A includes map with historic site numbers. Historic Current n P. Historic location Record Record of merriami of P. /n trap Historic Current Current vegetation descriptions P. nights Site No. Site No. merriami classification merriami 224.52 Mesquite 5.5 mi S, 7.5 mi E series, but a narrow, rocky wash 33 1 yes no 0/50 Continental 143.10 Semidesert Sawmill Canyon grassland, xeroriparian 33 2 yes yes 6/50 Wash at FR62 scrub Weisner’s Ranch. 154.100, Sonoran 11 (or 12) mi. S desertscrub Tucson xeroriparian scrub 18 3 yes yes 2/50 Weisner’s Ranch. 154.100, Sonoran 11 (or 12) mi. S desertscrub Tucson xeroriparian scrub 18 4 yes yes 2/50 224.52, Sonoran 9 mi W of junction Riparian Deciduous Ajo and Mission Forest and Woodlands, Road; 10 Mi W, 5.5 Mesquite Series Mi S of Tucson; 10 8 5 yes yes 2/50 mi SW Tucson 154.10, Sonoran 7.8 mi. E, 3.7 mi. N Desertscrub of Silver Bell Peak 1 6 yes yes 2/50 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series, 7 no yes 6/50 mesquite-invaded upland, adjacent to 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series 8 no no 0/25 224.52 Sonoran Riparian Deciduous 3.5 mi. E Junction Forest and Woodlands, Redington Rd. on Mesquite Series 50 9 yes Mt. Lemmon Road no 0/50 224.52 Sonoran Riparian Deciduous 3.5 mi. E Junction Forest and Woodlands, Redington Rd. on Mesquite Series 50 10 yes Mt. Lemmon Road no 0/100 8 Historic Site No. Current Site No. Historic location descriptions 11 Historic Record of P. merriami no 38 or 39 12 yes 2 mi N, 4 mi W Arivaca 38 or 39 13 yes 2 mi N, 4 mi W Arivaca 38 or 39 14 yes 2 mi N, 4 mi W Arivaca 41 or 42 15 yes at or “near” Mormon Lake 41 or 42 17 yes at or “near” Mormon Lake 16 no 18 no 19 no Current vegetation classification Current Record of P. merriami no n P. merriami /n trap nights 0/50 143.10 Semidesert grassland, xeroriparian scrub yes 2/50 143.10 Semidesert grassland, xeroriparian scrub 143.10 Semidesert grassland, xeroriparian scrub 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series yes 1/25 no 0/25 yes 2/50 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series yes 2/50 143.10 Semidesert grassland, mesquite invaded, 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series no 0/50 no 0/50 no 0/50 224.53 Sonoran Riparian Deciduous Forest and Woodlands, Cottonwood-Willow Series and 143.10, Semidesert grassland which is densely invaded by mesquite Number of sites trapped in this study that have historic records of P. merriami= 7 Number of historic sites with current P. merriami= 6 Number of historic locations trapped with no P. merriami trapped in this study= 2 Number of sites with no historic records but with current P. merriami = 1 *For some locations, there were several years of record and several collectors, only one of which was selected as an example in this table. 9 IDENTIFICATION We identified all mammals trapped using a field key prepared for this project (Appendix C). Baculum and skull characteristics were not useful in the present study, which relied entirely on external characteristics of living animals. All known previous measurements for the species were based on dead specimens. Measurements of body parts, especially total body length, of living animals that are attempting to escape the clutches of the investigator are undoubtedly different from, and probably shorter than, measurements of dead specimens. We made tentative identification of Peromyscus to species based on measurements and coloration. Dividing points used to tentatively distinguish P. merriami from P. eremicus were based on Hoffmeister and Lee (1963). If total length was > 191 mm, then we tentatively identified the animal as merriami. If total length was less than 191 mm, but if the hind foot was > 21 mm or ear > 20 mm and there was an evident cinnamon spot, we tentatively identified the animal as merriami. However, we labeled all specimens as “eremicus/merriami” until DNA analysis could confirm or refute our field determinations. Identifications were subsequently confirmed by DNA analysis conducted by Dr. Lois F. Alexander, Postdoctoral Scholar in the laboratory of Dr. Brett Riddle, University of Nevada, Las Vegas, utilizing the technique described in Riddle et al. (2000). Riddle and Alexander intend to publish their results, but have permitted use of their data as confirmation of identification for this project. 10 TORTOLITA MOUNTAIN PARK 6 CORONADO BINGHAM-CIENEGA NATURAL PRESERVE ORO VALLEY CATALINA Figure 1: Mesquite Mouse Trapping Locations and Suitable Habitat STATE MARANA PARK NATIONAL Historic Mouse Locations Trapped in 2004-2005 FOREST "Peromyscus merriami" SAGUARO NATIONAL "Peromyscus eremicus" PARK 9 (WEST) 10 "no Peromyscus" TUCSON TUCSON MOUNTAIN SAGUARO 19 18 PARK Administrative Boundaries NATIONAL SOUTH TUCSON PARK 5 (EAST) Major Streets Suitable Habitat PASCUA YAQUI NATION Major Washes 12 TOHONO O'ODHAM NATION SAN XAVIER DISTRICT Trapping Location Number COLOSSAL CAVE 3 4 CIENEGA CREEK NATURAL PRESERVE 8 7 COYOTE MOUNTAIN WILDERNESS AREA SAHUARITA EMPIRITA RANCH Suitable Habitat is dreived from the following BLP codes: 124.71 - Sonoran Riparian Woodland 143.10 - Semidesert grassland xeroriparian scrub 154.10 - Sonoran desertscrub xeroriparian scrub 224.52 - Sonoran Riparian and oasis forests, mesquite series 234.71 - Sonoran deciduous swamp and riparian scrub mixed scrub series CORONADO 11 NATIONAL SANTA RITA Pima County Index Map FOREST EXPERIMENTAL RANGE AND WILDLIFE AREA 1 2 LAS CIENEGAS NATIONAL CONSERVATION AREA Index Map Scale 1:2,000,000 BUENOS AIRES NATIONAL WILDLIFE REFUGE 16 1214 13 17 15 The information depicted on this display is the result of digital analyses performed on a variety of databases provided and maintained by several governmental agencies. The accuracy of the information presented is limited to the collective accuracy of these databases on the date of the analysis. The Pima County Department of Transportation Technical Services Division makes no claims regarding the accuracy of the information depicted herein. This product is subject to the Department of Transportation Technical Services Division's Use Restriction Agreement. Pima County Department of Transportation Scale 1:136,000 Pima County Technical Services 201 North Stone Avenue - 9th Floor Tucson, Arizona 85701-1207 (520)740-6670 - FAX:(520)798-3429 http://www.dot.co.pima.az.us mprobstf\markp\neva\FCD499\mxds\mouse_trappings.mxd mp 5/24/2006 RESULTS AND DISCUSSION HISTORICAL REVIEW Specimen records were obtained from: Illinois Natural History Survey and University of Illinois; National Museum of Natural History (Smithsonian Institution); University of Arizona; Museum of Vertebrate Zoology, University of California, Berkeley; and Museum of Texas Tech University. Data use restrictions imposed by the owners of the data (the museums) prevent publication or inclusion of the full set of data in this report, however the full compiled data set has been submitted to Pima County. Appendix A lists all of the locations compiled from this data set. Most of the available Arizona records are from Pima County. Records from other Arizona counties include: Santa Cruz at Tumacacori, Pinal at Picacho Reservoir, and Gila at 20 miles east of Globe. A total of 295 museum records were obtained and compiled for specimens collected in Pima County, Arizona. Many of these (n= 125) were collected on the Tohono O’odham Reservation. There has been an evident decline in numbers of P. merriami specimens deposited in museums over the past several decades. Figure 3 graphs the number of known specimens identified as P. merriami and the decade in which they were caught. Significantly, of the 295 specimens included in this record, 133 were collected by W. & L. Goodpaster from 1956 to 1974, 21 were collected by Lee W. Arnold in 1939, and 27 by L.R. Commissaris in 1956-58. In other words, 181 specimens, 61% of the known specimens, were collected by three collectors. This suggests that the number of specimens have probably depended more upon serious efforts being made by collectors who can recognize the species and who worked in appropriate areas, than any other factor. It is not possible to support or refute a hypothesis that the species has declined by examining the specimen record. NUMBER OF SPECIMENS 120 100 80 60 40 20 0 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 DECADE BEGINNING Figure 3. Number of specimens of P. merriami and decade of capture. 12 P. merriami was not identified from several areas that have been intensively studied in the past several decades that have mesquite bosques and/or mesquite lined washes. These include Saguaro National Park (Duncan 1990, Swan 2004), the Santa Catalina Mountains (Lange 1960), the Chiricahua Mountains (Maza 1965), the Huachuca Mountains (Hoffmeister and Goodpaster 1954), the upper San Pedro River valley (Duncan 1988), The Gila Box National Riparian Conservation Area (Snow et al. 2004), Tumacacori National Historic Site (B. Powell, University of Arizona, personal communication), and the State of New Mexico (Frey 2004). HABITAT ASSESSMENT The initial overlay of historic mesquite mouse locations and areas currently with vegetation classified as 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series did not show a consistently strong relationship. However, all historic locations mapped are within or close to mesquite-dominated communities. The locations at or near which P. merriami had historically been trapped that were examined in this study still had an abundance of mesquite that could be classified under the following different mesquite-dominated vegetation associations: 143.10, Semidesert grassland, xeroriparian scrub 154.10, Sonoran desertscrub xeroriparian scrub 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series Pima County estimates of current acreage in each of these communities is: 143.10, Semidesert grassland, xeroriparian scrub 154.10, Sonoran desertscrub xeroriparian scrub 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series 134,768 acres 126,901 acres 26,469 acres Undoubtedly, not all vegetation in these communities provides habitat for P. merriami, but data are insufficient to clearly determine which sites within these communities can or cannot support P. merriami. Some ideas that may contribute to further characterization of suitable habitat are suggested later in this report. Mesquite-dominated floodplain vegetation remains present at or near all of the historic locations, as indicated by aerial photographs, recent vegetation maps and direct field examination. Certainly qualitative and quantitative changes in the number, size, and density of mesquites have occurred between historic conditions and current conditions, but these have not been and cannot be documented because the historic data are lacking. Appendix B describes the current vegetation conditions at the sites that were trapped during this study. It is apparent that mesquite mice are not found only in large mesquite bosques, but in a wider variety of mesquitedominated communities on floodplain soils. The most famous mesquite bosque known to science was south of Tucson, on the San Xavier District of the Tohono O’odham Nation. This bosque was the site of capture of 114 of the known 295 specimens of P. merriami known from Pima County. It included the study sites of Arnold (1940), Commissaris (1960), and some of the specimens studied by Hoffmeister and Lee (1963). Historically, trees in this bosque reached heights of 60 feet and diameters of 4 feet, and the bosque was “miles in extent” (Swarth 1905, writing of conditions in 1902). Various authors 13 have discussed this site as an important birding area and its decline due to wood cutting (reviewed by Arnold 1940, Brandt 1951, Johnson and Carothers, in prep.). Other evidence strongly implicates diversion of the river, channel changes, wood cutting, and also (and perhaps most importantly) groundwater pumping (Turner 2003, Stromberg 1993). Arnold (1940) described the decline of the bosque, pointing out (p. 6): “Where once stood magnificent trees fifty and sixty feet high now stand dense thickets of second and third growth mesquite, with occasional large stumps to mark the spots where former giants once stood. It is not because of the size and height of the trees but because of the density of the remaining growth that I feel that the botanical classification ‘Mesquite Forest Association’ …still fits the area today.” Specimens of P. merriami continued to be trapped in this area, with the last known specimen from 1981. At some point in the mid-1980s, it became more difficult to obtain permission to trap in this area (Y. Petryszyn personal communication). It is not known whether P. merriami continues to survive there. Certainly the once magnificent mesquite bosque is no more. However, there are some mesquite dominated areas remaining in the general vicinity that were probably at one time contiguous with the large bosque. Some of these have trees with basal trunks greater than one meter in diameter, although today the trees are less than ten meters tall. At two of those sites, P. merriami was caught in this study (Appendix B, Sites 3 and 4). Evidence in the form of repeat photography indicates that in at least some areas, mesquite woodland now exists where it previously did not (Turner 2003). This is evident at Cienega Creek County Park, one of the areas investigated for the present study at which several P. merriami were trapped (Appendix B, Site 7). Turner presented a photograph showing an almost total absence of trees in 1880, contrasted with a photograph from 1998 showing a dense riparian tree community, including a mesquite bosque and cottonwood-willow forest. He concludes that there has been a dramatic forest expansion along at least some Southwestern rivers and streams over the past century. In large part, these are attributed to hydrologic changes resulting in the general loss of sacaton dominated grasslands and improvement of conditions for woody plants. “Thus, today we have cottonwood-willow forests flanked by mesquite bosques where a century ago the valleys supported grassy expanses of sacaton and tobosa grass. The mesquite forests . . .have much the same history [as the cottonwood-willow forests]. Present in a few places when early travelers passed through our region, they quickly expanded following the downcutting of the late 19th century. Some of these have disappeared in recent years. For example, the dense mesquite forest on the Santa Cruz near San Xavier Mission slowly died as groundwater pumping depleted the aquifer beneath it.” (Turner 2003, p. 32-33). 14 SELECTIVE TRAPPING EFFORT P. merriami were found at six out of eight areas at or near historic locations at which P. merriami were trapped, and 10 out of 19 trapping sites (Tables 2 and 3, Appendix B, and Figure 1). P. merriami was trapped in several of mesquite-dominated vegetation communities, classified as: 143.10, Semidesert grassland, xeroriparian scrub 154.100, Sonoran desertscrub xeroriparian scrub 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series They were not found in the following circumstances: (1) mesquite-invaded grassland (sites 11 and 16) or mesquite-invaded upland sites adjacent to floodplain sites (site 8); (2) narrow, rocky washes with few mesquites growing in strips only one or two trees wide (sites 1, 14, 18, 19); although site 2 might be described in these terms and did have mesquite mice, it was appreciably broader and had more mesquite than site 1; or (3) patches of mesquite isolated by urban development (sites 9, 10, 18, 19). The number of sites representing each of the above conditions is small, and there is overlap in some conditions (narrow rocky and urban encroached for sites 18 and 19), and the number of trap nights is limited, so data are not sufficient to support a conclusion that such situations never support mesquite mice. However, this may present a hypothesis for further research. In the field, based on habitat and morphology, including measurements of living mice and observations of pelage, many of the mice trapped appeared to be consistent with the descriptions of P. merriami. Originally, all specimens were labeled as P. eremicus/merriami. As it turned out, all were proved by DNA analysis to be merriami. Some individuals that were tentatively identified as eremicus, and therefore from which we did not collect specimens, may have been merriami. Table 3 summarizes the results for all individuals in the genus that were trapped during this study. Greater detail on all rodents trapped and the sites is included in Appendix B. Table 2 summarizes the results with regard to historic and current locations of P. merriami, and the relative abundance of P. merriami (as number trapped per number of trap nights). 15 Table 3. Summary of Peromyscus trapped during this study Species eremicus? merriami merriami merriami merriami merriami merriami merriami merriami merriami merriami merriami merriami merriami merriami eremicus? eremicus? eremicus? merriami merriami eremicus? merriami merriami? merriami? merriami? eremicus? eremicus? eremicus? eremicus? eremicus? eremicus? eremicus? eremicus? eremicus? merriami merriami merriami merriami merriami merriami merriami? Specimen no. KJK495 KJK496 KJK497 KJK498 KJK499 KJK500 KJK501 KJK502 KJK503 KJK504 KJK505 KJK506 KJK507 KJK508 KJK509 KJK510 KJK511 KJK512 KJK513 KJK514 KJK515 KJK516 KJK517 TL BL Tail HF Ear Sex Age 160 190 195 165 190 191 171 180 200 183 201 199 181 200 195 155 155 175 203 181 92 205 189 195 200 190 185 180 190 172 189 180 175 175 201 195 192 193 191 179 170 68 92 95 83 98 90 80 88 87 80 94 96 86 85 91 74 71 79 83 86 89 96 93 95 89 87 87 87 86 86 95 95 83 80 94 95 93 92 87 83 96 92 98 100 82 92 101 91 92 113 103 107 103 95 115 104 81 84 96 120 95 3 109 96 100 111 103 98 93 104 86 94 85 92 95 107 100 99 101 104 96 74 18 20 22 19 21 20 21 22 23 22 22 20 21 21 19 19 20 19 21 22 18 21 20 21 21 20 19 19 20 19 19 19 18 19 21 20 20 23 21 21 22 13 20 19 20 18 20 17 17 19 18 19 20 20 17 17 15 16 15 18 19 17 18 21 20 19 17 16 18 18 16 17 15 18 17 20 20 21 20 20 21 20 F F M F F F F M F M F M M F M F M F F F F F M M F F F F M F F M F F F F F F F F F AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD JUV JUV AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD AD Pectoral spot YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES Site 1 2 2 2 2 2 2 3 3 4 4 5 5 6 6 6 6 7 7 7 7 7 7 7 7 8 8 8 8 8 8 11 11 11 13 13 12 17 17 15 15 Notes: Individuals with Species lacking ? have been positively identified as merriami by DNA analysis. Measurements: TL= total length; BL= body length; Tail= length of tail from base to tip; HF- hind foot length; EAR= greatest length of ear from notch to tip. Sex: F= female, M= male; Age: AD= adult, JUV= juvenile. Site numbers are those in Appendix B. 16 CONCLUSIONS The evidence indicates that: 1. P. merriami is not uncommon in the appropriate habitat, which appears to be mesquite-dominated vegetation of several types on floodplain soils; 2. P. merriami is geographically widespread in eastern Pima County. It was found near the northern, southern, and eastern boundaries of its historically known range, where potentially suitable habitat could be sampled. The western boundary of distribution was not examined in this study. 3. P. merriami is not restricted to large mesquite bosques, but can be found in several different vegetation associations as mapped by Harris et al.(2000). These are: 143.10, Semidesert grassland, xeroriparian scrub 154.100, Sonoran desertscrub xeroriparian scrub 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series All have mesquite as the dominant tree at the locations trapped, but do not appear to be exclusively remnants of old bosques. 4. Some areas that were historically known to provide habitat for P. merriami have suffered dramatic losses of large trees and alteration of the biological community. However, it is not known whether these areas continue to have mesquite mice present because they were not included in this study. Adjacent areas are known to have mesquite mice. 5. The observed decrease in collection of specimens of P. merriami probably reflects a decreased effort of trapping in the appropriate vegetation types by knowledgeable collectors, combined with the extreme difficulty in recognizing the species on the basis of morphology. 6. It is extremely difficult or impossible to consistently identify this species based on morphological characteristics alone. P. merriami cannot be positively identified on the basis of measurements or pelage. Of 23 individuals trapped and positively identified as P. merriami by DNA analysis, 15 had a cinnamon colored pectoral spot, eight did not. The range of measurements in millimeters for these individuals is as follows: Total Length: 165-205 Tail: 82-120 Hind foot: 19-23 Ear: 17-21. Although it may be reasonable to conclude that a large mouse with a cinnamon pectoral spot trapped in a mesquite-dominated situation in Pima County is probably P. merriami, it is not possible to rule out this species if the mouse is small and lacks the spot. 17 7. No nocturnal rodents were trapped at several sites that appeared to provide ideal vegetation conditions (Sites 9, 10, 18, 19). These sites are mesquite bosques or mesquite lined washes owned by Pima County. All are adjacent to residential developments. The sites are relatively small and isolated from other natural sites by areas of residential development and/or urban infrastructure. Feral or roaming domestic cats were observed at sites 18 and 19. It is possible that either isolation or predation are factors that account for the absence of nocturnal rodents at these sites. 8. Mesquite mice were caught at sites that had a wide variety of disturbances of human origin (e.g., livestock grazing, trash, power lines, dirt roads, intensive nocturnal traffic by immigrants, migrant trails, limited wood cutting). It is not possible to determine the effects of any of these disturbances without further study. Data do not support specific management recommendations for conservation of mesquite mice, other than basic conservation of large blocks of suitable habitat with some (as yet unknown) level of connectivity or proximity to each other. Restrictions on human activity, such as limiting access, reducing or eliminating livestock grazing, or disallowing specific activities including camping, horseback or mountain bike riding, or any other activity that does not result in conversion of mesquite-dominated woodland to some other vegetation type are not supported by the data. 9. Further research, targeted toward effects of specific impacts, would be necessary to determine the effects of any human activities on mesquite mice or to support changes in regulation or management of potentially suitable habitat. 18 ACKNOWLEDGEMENTS Many individuals made important contributions to this work. The work was supported by a grant from the Arizona Heritage Fund. Julia Fonseca, Environmental Program Manager for Pima County Department of Transportation and Flood Control District (PCFCD), was instrumental in obtaining the grant, and provided encouragement, interest, and field assistance. Neva Connolly (PCFCD) served as the project manager for the County. John Regan, GIS Manager for PCFCD, directed the preparation of the GIS and maps. Dr. Yar Petryszyn, Curator of the Mammalogy Collection, University of Arizona, provided insight and inspiration as well as access to specimens and data. The following individuals participated in fieldwork: Eleanor Gladding, Jessica Lee, Katie Schoenhorst, Robin Llewellyn (all of SWCA Environmental Consultants); Amy Gaiennie; Julia Fonseca and Dale Turner. Tim Snow, Lin Piest, and Bill Van Pelt of the Arizona Game and Fish Department provided information. Mary Hunnicut, Wildlife Biologist, and Sally Gall, Assistant Manager, Buenos Aires National Wildlife Refuge, expedited issuance of a Special Use Permit to conduct trapping on the Refuge. Dr. Brett Riddle and Dr. Lois Alexander, University of Nevada, Las Vegas, provided information on their specimens and conducted the DNA analysis of specimens collected during this study. The following museum personnel provided records of their specimens of P. merriami: Dr. Joyce Hofmann, Illinois Natural History Survey and University of Illinois; Robert D. Fisher and Craig Ludwig, National Museum of Natural History (Smithsonian Institution); Dr. Yar Petryszyn, University of Arizona. Online museum records were obtained from: Museum of Vertebrate Zoology, University of California, Berkeley and Museum of Texas Tech University. In all cases, the data are the property of the respective museums. The draft manuscript was reviewed by Eleanor Gladding and Amy Gaiennie, and produced in its final form by Lucy Simpson, SWCA, Tucson. An anonymous reviewer affiliated with the Arizona Game and Fish Department reviewed the draft manuscript and made many useful comments that resulted in significant improvement of the final report. 19 LITERATURE CITED Arnold, L.W. 1940. An Ecological Study of the Vertebrate Animals of the Mesquite Forest. Unpublished M.S. Thesis, University of Arizona. Avise, J.C., M.H. Smith, R.K. Sealander, T.E. Lawlor, and P.R. Ramsey. 1974. Biochemical polymorphism and systematics in the genus Peromyscus V. Insular and mainland species of the subgenus Haplomylomys. Systematic Zoology 23: 226-238. Baker, R.J., L.C. Bradley, R.D. Bradley, J.W. Dragoo, M.D. Engstrom, R.S. Hoffman, C.A. Jones, F Reid, D.W. Rice, and C. Jones. 2003. Revised checklist of North American Mammals North of Mexico, 2003. Occasional Papers of the Museum of Texas Tech University No. 229. Brandt, H. 1951. Arizona and its bird life. The Bird Research Foundation, Cleveland, Ohio. Commissaris 1960, Morphological and ecological differentiation of Peromyscus merriami from Southern Arizona. Journal of Mammalogy 41:305-310 Dice, L.R. and P. Blossom. 1937. Studies of mammalian ecology in southwestern North America with special attention to the color of desert mammals. Carnegie Institution of Washington Publication 485. Duncan, Douglas K. 1990. Nocturnal rodent populations and associated vegetation with implications of human use at Saguaro National Monument, Arizona. Cooperative National Park Resources Studies Unit, School of Renewable Natural Resources, The University of Arizona. Duncan, D.K. 1988. Small mammal inventory of the Upper San Pedro River Valley, Cochise County, Arizona. Progress Report. San Pedro Technical Report Number 1. Frey, J. 2004. Taxonomy and distribution of the mammals of New Mexico. Occasional Papers of the Museum, Texas Tech University No. 240. P. merriami is not listed as known from New Mexico. Hall, E.R. and K.R. Kelson. 1952. Comments on the taxonomy and geographic distribution of some North American rodents. University of Kansas Publications of the Museum of Natural History 5: 343-371. Harris, L.K., J.A. Wennerlund, and R.B. Duncan. 2000. Riparian vegetation mapping and classification, Sonoran Desert Conservation Plan. Pima County Government. Tucson, AZ. Contract No. O7-30-H-127196-0100. 56 pp. Hoffmeister, D.F. 1986. Mammals of Arizona. The University of Arizona Press and Arizona Game and Fish Department. 602 pp. Hoffmeister, D.F. and W. W. Goodpaster. 1954. Mammals of the Huachuca Mountains, southeastern Arizona. Illinois Biol. Monogr. 24: 1-152. Hoffmeister, D.F. and M.R. Lee 1963. The status of the sibling species Peromyscus merriami and P. eremicus. Journal of Mammalogy 44:201-213. Hooper, E.T. 1968. Classification. pp. 27-74 In: J.A. King, ed. Biology of Peromyscus (Rodentia). Special Publications No. 2. American Society of Mammalogists. 20 Johnson, R.R. and S.W. Carothers. In prep. The great San Xavier mesquite forest and Santa Cruz River: a lesson in impacts of overutilization of natural resources. Lange, K.L. 1960. Mammals of the Santa Catalina Mountains, Arizona. The American Midland Naturalist 64: 436-458. Lawlor, T.E. 1971. Distribution and relationships of six species of Peromyscus in Baja California and Sonora, Mexico. Occ. Papers of the Museum of Zoology, Univ. of Michigan. No. 661. Maza, B. G. 1965. Mammals of the Chiricahua Mountain Region, Cochise County, Arizona. Unpublished M.S. Thesis. The University of Arizona. 189 pp. Mearns, E.A. 1907. Mammals of the Mexican Boundary of the United States. Bull. U.S. National Museum 54: 1-384. Petryszyn, Y. and S. Russ. 1996. Nocturnal rodent population densities and distribution at Organ Pipe Cactus National Monument, Arizona. Technical Report No. 53. United States Department of the Interior National Biological Service, Cooperative Park Studies Unit, The University of Arizona and National Park Service, Organ Pipe Cactus National Monument. Riddle, B.R., D.J. Hafner, and L.F. Alexander. 2000. Phylogeography and systematics of the Peromyscus eremicus species group and the historical biogeography of North American warm regional deserts. Molecular Phylogenetics and Evolution 17: 145-160. Snow, T.K., A. Averill-Murray, and C.L. Blasch. 2004. Vertebrate inventory of the Gila Box Riparian National Conservation Area. Technical Report 230, Nongame and Endangered Wildlife Program, Arizona Game and Fish Department, Phoenix. Stromberg, J.C. 1993. Riparian mesquite forests: a review of their ecology, threats, and recovery potential. Journal of the Arizona-Nevada Academy of Science 27: 111124. Swann, D. 2004. Checklist of herpetofauna and mammals of Saguaro National Park, Arizona. Western National Parks Association. SWCA Environmental Consultants. 2003. Isabella Lee Resource Survey, Tucson, Arizona. Unpublished report submitted to Pima County Department Of Transportation And Flood Control, Floodplain Management Division. Turner, R.M. 2003. Pima County’s Withdrawal from Its Past. Report available as part of the Sonoran Desert Conservation Plan series, Pima County Administrator’s Office, Tucson, AZ. 34 pp. 21 Evaluation of Mesquite Mouse (Peromyscus merriami) Status in Pima County Appendices Arizona Game & Fish Department Heritage Grant Project #IO5001 May 2006 PIMA COUNTY BOARD OF SUPERVISORS Richard Elías, Chairman, District 5; Ann Day, District 1 Ramón Valadez, District 2; Sharon Bronson, District 3 Raymond J. Carroll, District 4 COUNTY ADMINISTRATOR C.H. Huckelberry mesqmousecvr506.ai•db•20157 Evaluation of Mesquite Mouse (Peromyscus merriami) Status in Pima County Arizona Game and Fish Heritage Grant #I05001 APPENDICES APPENDIX A: POTENTIAL TRAPPING LOCATIONS……………………...……A1 APPENDIX B: SITES TRAPPED……………..……………………………….…....B1 APPENDIX C: FIELD KEY TO RODENTS LIKELY TO BE TRAPPED….…..…C1 DISCLAIMER The findings, opinions, and recommendations in this report are those of the investigators who have received partial or full funding from the Arizona Game and Fish Department Heritage Fund. The findings, opinions, and recommendations do not necessarily represent official Department policy or management practice. For further information, please contact the Arizona Game and Fish Department. APPENDIX A HISTORIC LOCATIONS OF MESQUITE MOUSE (PEROMYSCUS MERRIAMI) BASED ON MUSEUM LOCATIONS Appendix A. Historic Locations of Peromyscus merriami collection, based on museum records Map Year of Location Description Collector* Number Record* 1 7.8 mi. E, 3.7 mi. N of Silver Bell Peak 1973 Carl Hoagstrom Avra Valley, 5 mi. W Marana on Trico Rd., 1 1/2 mi. N Marana-Trico Rd. 2 1967 Ronald Wheeland 1/2 m. W, on Avra Vallley Rd., 1/2 mil. 3 1969 J. Currlin S of the road 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 drove out freeway 10, turned onto Avra Valley Rd., went approx. 1mi. turned left to river and just across rive Fort Lowell Fort Lowell 13.7 mi W Tucson 9 mi. W Jct. Ajo & Mission Roads 10 mi W, 5.5 mi S of Tucson 10 mi SW Tucson 9 mi SW Tucson 8 mi SW Tucson 8 mi SW Tucson 8 mi SW Tucson 13.75 mi SW Tucson San Xavier Mission T15S, R14E, Sec. 21, NE1/4 mesquite forest, Weisner's Ranch, 12 mi S Tucson Wilmot Station, 13 mi SE Tucson Old Sonoita Hwy., 0.3 mi. E Rt. 83 1 mi. W, 1 mi. N, Benson Highway - Old Helvitia Rd. jct. 18 mi SE Tucson 12 mi S Tucson, Santa Cruz River Continental Continental 2 Mi S of Continental at Morales 2 Mi S of Continental at Morales 2 Mi S of Continental at Morales Morales, 28 mi. S of Tucson Santa Rita Exp. Range, 7.1 mi. ESE Sahuarita 2 mi S, 7 mi E Continental 5 mi S, 6 mi E Continental 5.5 mi S, 7.5 mi E Continental 0.5 mi E Arivaca 29 mi. S of Three Points Clark Ranch, 4 mi. E Arivaca, 3850f. 1969 1893 1893 1969 1956 1956 1939 1939 1957 1939 1939 1939 1974 1983 R.D. Ullery Hoizner, F. X. Holzner, F.X; Mearns M. R. Lee W. J. S. Keith E. Justice Lee W Arnold Lee W Arnold W. & L. Goodpaster Lee W Arnold Lee W Arnold Lee W Arnold W. D. Severinghaus J. Brown 1939 L. A. 1959 D. F. Hoffmeister 1992 D.A. Parizek 1956 1977 1980 1921 1921 1931 1931 1931 1931 L.R. Commissaris M. R. Lee Petryzyn, Y Baily, V. Baily, V. Taylor, W. P. Taylor, W. P. Taylor, W. P. Taylor W. P. 1970 1963 1963 1963 1959 1965 1939 J.C. Geest W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster S.E. Doster Lee W. Arnold Trapped at or Near This Study X X X X 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 Arivaca 2 mi N, 4 mi W Arivaca 2 mi N, 4 mi W Arivaca 4 mi. E Arivaca, 3850 near Mormon Lake, 4 mi. N, 1 mi. E Sasabe, 3500ft Morman Lake, 4 mi. N, 1 mi. E Sasabe 29 mi. S of Three Points on Sasabe Rd. Sabino Canyon, Baboquivari Mts. 3 mi SE Topawa, Baboquivari Mts. 3 mi SE Topawa, Baboquivari Mts. 3 mi SE Topawa, Baboquivari Mts. 3 mi SE Topawa, Baboquivari Mts. 3 mi SE Topawa, Baboquivari Mts. 3.5 mi. E Junction Redington Rd. on Mt. Lemmon Road 2 mi. S, 1 1/2 mi. W Redington, 4000ft. 0.75 mi E Lukeville, Organ Pipe Cactus National Monument 0.75 mi E Lukeville, Organ Pipe Cactus National Monument 0.75 mi E Lukeville, Organ Pipe Cactus National Monument Organ Pipe Cactus Nat'l Mon., Dos Lomitos 4 mi. SE Lukeville 0.75 mi E Lukeville, Organ Pipe Cactus National Monument 0.75 mi E Lukeville, Organ Pipe Cactus National Monument 1 1/2 mi. E, 2 mi. N (by rd.) Oracle, 4000ft 1960 1962 1962 1939 Bernard Maza W. & L. Goodpaster W. & L. Goodpaster Lee W. Arnold 1953 1953 1965 1956 1956 1956 1956 1956 1956 Keith E. Justice L.D. Beatty G.C. Bateman W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster W. & L. Goodpaster 1953 A.G. Baker 1953 E.L. Cockrum X X X 1962 W. & L. Goodpaster 1962 W. & L. Goodpaster 1962 W. & L. Goodpaster 1988 Peryszyn Y 1962 W. & L. Goodpaster 1962 W. & L. Goodpaster 1953 Robert B. Murray * for some locations, there were several years of record and several collectors, only one of which was selected as an example in this table. NATION NATION O'ODHAM TOHONO 45 46 47 48 49 TOHONO O'ODHAM BABOQUIVARI PEAK WILDERNESS AREA IRONWOOD 44 42 41 43 40 FOREST NATIONAL CORONADO REFUGE WILDLIFE NATIONAL BUENOS AIRES COYOTE MOUNTAIN WILDERNESS AREA MONUMENT NATIONAL FOREST 1 39 38 37 35 2 36 7 4 9 CORONADO NATIONAL FOREST 34 28 29 25 23 18 26 27 24 SAHUARITA 16 ORO VALLEY TOHONO O'ODHAM NATION SAN XAVIER DISTRICT 15 12 13 14 11 10 PASCUA YAQUI NATION 8 TUCSON MOUNTAIN PARK SAGUARO NATIONAL PARK WEST 3 MARANA TORTOLITA MOUNTAIN PARK 6 30 32 33 AND WILDLIFE AREA 31 SANTA RITA 20 21 19 FOREST NATIONAL CORONADO FOREST NATIONAL CORONADO EXPERIMENTAL RANGE 17 TUCSON 5 CATALINA STATE PARK LAS CIENEGAS NATIONAL CONSERVATION AREA CORONADO NATIONAL FOREST CIENEGA CREEK 22 PRESERVE NATURAL COLOSSAL CAVE MOUNTAIN PARK SAGUARO NATIONAL PARK EAST 51 50 BINGHAMCIENEGA NATURAL PRESERVE mprobstf/markp/neva/fcd499/mxds/mouse_ownership.mxd mp Pima County Technical Services 201 North Stone Avenue - 9th Floor Tucson, Arizona 85701-1207 (520)740-6670 - FAX:(520)798-3429 http://www.dot.co.pima.az.us Pima County Department of Transportation This product is subject to the Department of Transportation Technical Services Division's Use Restriction Agreement. The information depicted on this display is the result of digital analyses performed on a variety of databases provided and maintained by several governmental agencies. The accuracy of the information presented is limited to the collective accuracy of these databases on the date of the analysis. The Pima County Department of Transportation Technical Services Division makes no claims regarding the accuracy of the information depicted herein. Index Map Scale 1:1,500,000 Pima County Index Map Major Streets 4/13/2006 Scale 1:160,000 Administrative Boundaries Private Lands Tribal Lands Municipal Lands County Lands State Lands Federal Lands Historic Mouse Locations Appendix A Map: Mesquite Mouse Historic Locations and Land Ownership Evaluation APPENDIX B SITES TRAPPED B-1 Site 1. Florida Canyon Wash Aerial photograph scale: 1:4,000. B-2 Florida Canyon Wash upstream from crossing of Madera Canyon Road. T19S R14E S14 UTM (Zone 12, NAD 27 CONUS): 511507mE and 3515171mN. Map: Green Valley. Elevation 3,785 feet Ownership: State of Arizona. Historical information: This site is close to that described as 5.5 mi S, 7.5 mi E Continental at which W. and L. Goodpaster collected P. merriami in March 1963. It is also close to the site described as 5 mi S, 6.5 mi E Continental at which B.R. Riddle trapped P. merriami in March 2002. Access: Madera Canyon Road, approximately ½ mile after road turns S, a dirt road goes off to the E and loops S to parking area between fences. To the east is private land; to the west is state land. Vegetation conditions: The Harris Riparian map shows this site as 224.52, Mesquite Series. This is a mesquite-dominated floodplain and wash area. The wash is deeply incised in some places, and the substrate is a mix of cobbles, gravel, and sand. There are some fairly large velvet mesquites (Prosopis velutina) and a few netleaf hackberry (Celtis laevigata var. reticulata) trees. The area is impacted by cattle grazing. There is a ruin of an adobe house and windmill, and at least two active water tanks in the immediate vicinity. Shrubs include wolfberry (Lycium spp.) and catclaw acacia (Acacia greggii). Grasses were Lehmann lovegrass (Eragrostis lehmannii) and plains lovegrass (Eragrostis curvula). Methods: Standard operating procedure (see methods section). Two trap lines, each consisting of 25 traps, were placed in the evening of 13 December 2004 and picked up on the morning of 14 December. One line of traps was placed along the wash bottom, the other along the floodplain. Rodents trapped: Only one animal was caught. It is tentatively identified as an adult female P. eremicus. Measurements were: 160-92-18-13. It was not evidently pregnant or lactating. No specimen clipping was taken. B-3 Site 2. Sawmill Canyon Wash Aerial photograph scale: 1:4000 B-4 Sawmill Canyon Wash upstream and downstream from crossing of Forest Road 62. T19S R15E S18. UTM (Zone 12, NAD 27 CONUS): 514201mE and 3516113mN. Map: Helvetia. Elevation: 3,948 feet Ownership: State of Arizona. Historical information: This site is close to or the same as where Riddle et al. caught P. merriami that were positively identified by genetic analysis in 2002. Access: Madera Canyon Road to Forest Road 62 to wash crossing. Vegetation conditions: The Harris Riparian map shows this site as 143.10, Semidesert grassland, xeroriparian scrub. This is a mesquite-dominated wash and narrow floodplain. The substrate is sand and gravel with a few cobbles. Some of the mesquites appear to be quite old, averaging one to two feet in basal diameter, and some are much larger. Some are up to 20 feet tall. Trees are not dense, but scattered along the wash. The wash is partially incised, but the majority of its length appears to be fairly stable. Trees were velvet mesquite and netleaf hackberry. Shrubs were spiny hackberry (Celtis pallida), lotebush (Zizyphus obtusiflolia) and catclaw acacia. Grasses included Lehmann lovegrass and sacaton (Sporobolus wrightii). There are some signs of grazing, but not as heavily impacted by cattle as Florida Canyon Wash. Methods: Standard operating procedure (see methods section). Two trap lines, each consisting of 25 traps, were placed in the evening of 13 December 2004 and picked up on the morning of 14 December. Both lines were placed along the wash, one upstream the other downstream from the parking place. Rodents trapped: Genus Species TL T HF E Sex Age Comments Peromyscus eremicus? 160 92 18 13 Chaetodipus intermedius 180 110 22 5 M AD no specimen Perognathus flavus 154 90 15 5 M AD no specimen Peromyscus merriami 190 98 20 20 Chaetodipus intermedius 155 93 21 Peromyscus F AD no specimen F AD Pregnant; specimen KJK495 8 F AD no specimen leucopus 125 60 20 17 F AD Pregnant; no specimen Peromyscus merriami 195 100 22 19 M AD testes abdominal; specimen KJK496 Chaetodipus intermedius 160 92 21 Peromyscus merriami 165 82 19 20 F Peromyscus merriami 190 92 21 18 F AD cinnamon spot on thorax; specimen KJK497 AD Pregnant; specimen KJK498 Dipodomys merriami 232 148 34 13 F AD no specimen Peromyscus merriami 191 101 20 20 F AD Pregnant; specimen KJK499 Peromyscus merriami 171 91 21 17 F AD cinnamon spot on thorax; specimen KJK 500 8 M AD testes abdominal; no specimen B-5 Site 3. Arnold #1 Aerial photograph scale 1:4,000 B-6 Unnamed wash, tributary of Santa Cruz River. T16 S R14E S21. UTM (Zone 12, NAD 27 CONUS): 506947mE and 3542578mN. Map: Tucson SW. Elevation: 2,673 feet Ownership State of Arizona. Historical information: This is close to the site worked by Lee W. Arnold in 1938 and 1939, known as Weisner’s Ranch. 11 (or 12) mi. S Tucson 20 June1939 and other dates. The actual site is on private property, currently owned by ASARCO Inc. and no longer is mesquite dominated. Access: Enter by road off Nogales Hwy at approximately 3544525N; drive E through two gates to power line. Drive south approximately 1 mile to site. Vegetation conditions: The Harris Riparian map shows this site as 154.100, Sonoran desertscrub xeroriparian scrub. The area is partially carpeted with closely cropped grass. Soil is deep sandy loam, eroded in places, also wash bottom with few cobbles. Trees: velvet mesquite, blue paloverde (Parkinsonia florida), Jerusalem thorn (Parkinsonia aculeata), catclaw acacia. Some of the mesquites were more than 2 feet basal diameter and over 20 feet tall. Shrubs: wolfberry species, catclaw acacia, lotebush. Grasses and forbs: Bermuda grass (Cynodon dactylon), buffelgrass (Pennisetum ciliare), rough cocklebur (Xanthium strumarium). Cactus apple (Opuntia phaeacantha) and walkingstick cactus (Opuntia spinosior) were also present, and a few barrel cactus (Ferocactus sp). The site is heavily grazed and cattle were present. Methods: Standard operating procedure (see methods section). Two trap lines, each consisting of 25 traps, were placed in the evening of 14 December 2004 and picked up on the morning of 15 December. Both lines were placed along the wash, one upstream the other downstream from the parking place. Rodents trapped: Genus Species TL T HF E Sex Age Comments Peromyscus merriami 180 92 22 17 M AD testes abdominal; specimen KJK501 Peromyscus merriami 200 113 23 19 F Peromyscus leucopus 167 21 19 M AD pregnant; cinnamon spot on thorax; specimen KJK502 AD testes abdominal; no specimen 79 B-7 Site 4. Arnold #2 Aerial photograph scale 1:4,000 B-8 Unnamed wash, tributary of Santa Cruz River. T16 S R14E S28. UTM (Zone 12, NAD 27 CONUS): 506995mE and 3541242mN. Map: Tucson SW. Elevation: 2,677 feet Ownership State of Arizona. Historical Information: This is close to the site worked by Lee W. Arnold in 1938 and 1939, known as Weisner’s Ranch. 11 (or 12) mi. S Tucson 20 June1939 and other dates. The actual site is on private property, currently owned by ASARCO Inc. and no longer is mesquite dominated. Access: Enter by road at off Nogales Hwy at approximately 3544525N; drive E through two gates to power line. Drive south approximately 1.8 mile to site, approximately 0.8 mi S of previous site. Vegetation conditions: Harris Riparian map shows this site as 154.100, Sonoran desertscrub xeroriparian scrub. Soil is deep sandy loam, eroded in places with some very deep and narrow gullies, also wash bottom with few cobbles. Part of the wash is braced with old car bodies. Immediately downstream from the site is a concrete dam and riprapped bank. Trees: velvet mesquite, blue paloverde, Jerusalem thorn, catclaw acacia. Some of the mesquites were more than 2 feet basal diameter and over 20 feet tall. Shrubs: wolfberry species, catclaw acacia, lotebush. Grasses and forbs: Bermuda grass, buffelgrass, rough cocklebur. Cactus apple and walkingstick cactus were also present, and a few barrel cactus. Vines of Drummond’s clematis (Clematis drummondi)are abundant. There were a few saguaros (Carnegia gigantea) and creosote bushes (Larrea tridentata) at the east end of the trap lines. The site is heavily grazed and cattle were present. Methods: Standard operating procedure (see methods section). Two trap lines, each consisting of 25 traps, were placed in the evening of 14 December 2004 and picked up on the morning of 15 December. Both lines were placed along the wash, each following a separate braid, going upstream from the parking place. One trap was not relocated. Rodents trapped: Genus Species TL T HF E Sex Age Comments Peromyscus merriami 183 103 22 18 Peromyscus merriami 201 107 22 19 Peromyscus leucopus 169 80 20 18 M AD testes abdominal; cinnamon spot on thorax; specimen KJK503 F AD cinnamon spot on thorax; specimen KJK504 M AD No specimen B-9 Site 5. Snyder Hill Road Aerial photograph scale: 1:4,000. B-10 Unnamed wash, disjunct tributary of Brawley Wash. T14S R11E S34. UTM (Zone 12, NAD 27 CONUS): 480670mE 3558348mN. Map: Brown Mountain. Elevation: 2,320 feet. Ownership: Pima County. Historical information: This site roughly coincides with several historic locations at which P. merriami was trapped. They are: W.J.S. 11 October 1956. 9 mi W of junction Ajo and Mission Road K. E. Justice. 27 December 1956. 10 Mi W, 5.5 Mi S of Tucson. Lee Arnold. 21 October 1939.10 mi SW Tucson and other locations with a mile or two and several dates. Access: Sandario Road to Snyder Hill Rd. Go E on Snyder Hill to curve. Park in one of the side cuts of the road on the north side and climb over a barbed wire fence. Vegetation conditions: The Harris riparian map shows this as 224.52, Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series. This is a dense thicket of velvet mesquite on a substrate of sandy-silty soil. Sheet flow and wash flow are interrupted by a levee at the west side of this property. The only trees are velvet mesquite, generally less than one foot in basal diameter and less than 15 feet tall. Shrubs present include wolfberry species, snakeweed (Gutierrezia sp.), burroweed (Isocoma tenuisecta), and lotebush. The area is apparently not grazed at the present time. Some scattered debris is present, and there is an old road at the north end of the area we trapped. Methods: Standard operating procedure (see methods section). Two trap lines, each of 25 traps, approximately 100 yards apart, more or less parallel northward from the fence line were set on the afternoon of 15 December 2004 and picked up on the morning of 16 December 2004. Rodents trapped: Genus Species Neotoma albigula TL T HF E Sex Age Comments AD released without measuring Peromyscus merriami 199 103 20 20 M AD testes abdominal; specimen KJK505 Dipodomys merriami 209 123 35 12 M Peromyscus merriami 181 95 21 20 M AD testes abdominal; no specimen testes abdominal; cinnamon spot on thorax; AD specimen KJK 506 Onychomys torridus 145 56 19 17 F AD Pregnant; no specimen Peromyscus leucopus 172 80 20 14 M AD testes abdominal; no specimen B-11 Site 6. Ironwood Forest National Monument Aerial photograph scale 1:5,000 B-12 Unnamed wash, tributary of Los Robles Wash. T11S, R10E, S19 UTM (Zone 12, NAD 27 CONUS): 466399mE and 3591471mN Map: West of Marana. Elevation: 1,920 feet. Ownership: Federal—Ironwood Forest National Monument. Access: along Silverbell Road, a short spur goes north just past wash. Historical information: This site is approximately where a specimen identified as P. merriami was trapped by Carl Hoagstrom on 12 May 1973 at 7.8 mi. E, 3.7 mi. N of Silver Bell Peak, 1900 feet Vegetation conditions: The Harris Riparian map shows this site as 154.10, Sonoran Desertscrub. The area is obviously impacted by illegal traffic, much littered with plastic bags, cans, blankets, clothing, etc. It has been used as a dumpsite for glass and metal, which are scattered throughout. Upstream from the road, there is a single wash, downstream there are several braids and islands. This is a classic desert wash, lined with mesquite and ironwood, some blue paloverde and catclaw acacia as trees. Shrubs included creosote bush, mostly but not entirely outside the wash, wolfberry species, catclaw and whitethorn acacia (Acacia constricta). Grasses and forbs included Bowlesia incana, and red brome (Bromus rubens). The area appears to be grazed, but not very heavily. Methods: Standard operating procedure (see methods section). Two trap lines, each of 25 traps, one upstream from the road, the other downstream. Because the downstream habitat was braided and had islands, traps were limited to the xeroriparian community and not set in a moreor-less straight line. A third trap line of 25 traps was set approximately 150 yards east of the wash, in upland vegetation, to ascertain small mammal activity for comparison to the wash. Traps were set on the afternoon of 15 December 2004 and picked up on the morning of 16 December. Rodents trapped: Genus species TL T HF E Sex Age Comments Dipodomys merriami 255 155 33 12 F AD desert trap line, only animal trapped Peromyscus merriami 200 115 21 17 F AD Specimen KJK507 195 104 19 17 M released without measuring AD testes abdominal; cinnamon spot on thorax; specimen KJK 508 Peromyscus eremicus? 155 81 19 15 F JUV creosote bush dominant; no specimen Peromyscus eremicus? 155 84 20 16 M JUV no specimen Neotoma albigula Peromyscus merriami B-13 Site 7. Cienega Creek County Park #1 Aerial photograph scale 1:4000 B-14 Mesquite woodland along Cienega Creek floodplain. T16S R17E S33. UTM (Zone 12, NAD 27 CONUS): 537102mE and 3540785mN. Map: Rincon Peak. Elevation: 3,460 feet. Ownership: Pima County. Cienega Creek County Park. Access: From Pantano Road (aka Marsh Station Road), bridge over Cienega Creek, go 3.16 miles to gate at 538002mE and 3541704mN. Go through two gates, to parking area immediately E of railroad track. Walk SW to the mesquite area. Historical information: There are no records of Mesquite Mouse from this area. It is unknown whether anyone has trapped this area. Turner (2003) shows that the area had few, if any, mesquite or other trees in 1880, but had become dominated by mesquite and cottonwood-willow vegetation in 1998. Vegetation conditions: The Harris Riparian map shows this area as 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series, and 234.712 Sonoran Deciduous Riparian Scrub. There are two terraces of sandy-loamy soil above Cienega Creek. The first terrace has cottonwood (Populus fremontii), Goodding’s willow (Salix gooddingii), and mesquite trees, some quite large. The second terrace has dense, smaller mesquites, with dense grass and vines. Shrubs include wolfberry species, lotebush, with creosote bush in the drier areas of the upper terrace. There is dense native bunch grass, mostly big sacaton (Sporobolus wrightii), and some young grass, apparently red brome and other annuals beginning to grow. The area has not been grazed for about ten years. Methods: Two trap lines of 25 traps were set on 16 December 2004 and picked up on 17 December 2004. One was entirely on the upper terrace, mostly under the mesquites at the edge of the open grassland. The second was along the edge of the upper terrace, then down to the lower terrace where ten traps were placed. Rodents trapped: Genus Species TL Chaetodipus penicillatus 180 95 23 8 F AD upper terrace; no specimen Onychomys torridus 145 46 19 15 F AD pregnant, upper terrace; no specimen Peromyscus eremicus? 175 96 19 15 F AD pregnant, upper terrace; no specimen Peromyscus merriami 203 120 21 18 F AD upper terrace; specimen KJK509 Chaetodipus baileyi 190 115 23 9 F AD upper terrace; no specimen Chaetodipus intermedius 160 89 20 9 F AD upper terrace; no specimen Peromyscus merriami 181 95 22 19 F Peromyscus eremicus? 92 18 17 F AD upper terrace, cinnamon spot; specimen KJK510 AD pregnant, no specimen Peromyscus merriami 205 109 21 18 F Peromyscus merriami? 189 96 20 21 M AD pregnant, cinnamon spot: specimen KJK511 AD cinnamon spot, no specimen Peromyscus merriami? 195 100 21 20 M AD lower terrace, no specimen Reithrodontomys fulvescens 138 79 17 12 F AD lower terrace, no specimen Peromyscus 200 111 21 19 F AD pregnant, upper terrace, cinnamon spot, no specimen merriami? T HF 3 E Sex Age Comments Site 8: Cienega Creek County Park #2 B-15 Aerial photograph scale 1:4,000 Floodplain of Cienega Creek, downstream from railroad bridge. T16S R17E S30. UTM (Zone 12, NAD 27 CONUS): 533196mE and 3542375mN. Map: Vail. Elevation 3,291. Ownership: Pima County. Cienega Creek County Park. Access: Park on W side of Marsh Station Road, walk along road that parallels railroad tracks to drainage running down into creek. B-16 Historical information: There are no records of Mesquite Mouse from this area. It is unknown whether anyone has trapped at this area. Vegetation conditions: The Harris Riparian map shows this area as 224.52 Sonoran Riparian Deciduous Forest and Woodlands, Mesquite Series. However, the actual trap line went along the edge of the mesquite bosque, along the base of the hills which are Arizona Upland with saguaros, various Opuntia species and barrel cactus, with scattered mesquites. Shrubs included creosote bush, triangle bur ragweed (Ambrosia deltoidea), snakeweed, and spiny hackberry. Mesquites were generally small, less than one foot basal diameter. Closer to the stream, not on the trap line, there are larger mesquites. The area is very heavily impacted by illegal traffic. There is much trash present and many trails. Methods: One line of 25 traps was placed along the edge of the bosque on 16 December and picked up on 17 December 2004. Rodents trapped: T HF E Sex Age Comments Genus Species TL Peromyscus eremicus? 190 103 20 17 F AD pregnant, no specimen Peromyscus eremicus? 185 98 19 16 F AD pregnant, no specimen Peromyscus eremicus? 180 93 19 18 F AD pregnant, no specimen Peromyscus eremicus? 190 104 20 18 M AD testes abdominal, no specimen Peromyscus eremicus? 172 86 19 16 F AD pregnant, no specimen Peromyscus eremicus? 189 94 19 17 F AD not pregnant, no specimen B-17 Site 9. King Street Aerial photograph scale 1:4,000 B-18 Tanque Verde Wash floodplain. T14S R15E S1 UTM (Zone 12, NAD 27 CONUS): 522314mE 3566894mN. 1480 N King Street, on E side of end of King Street. Map: Tucson East. Elevation: Ownership: Pima County. Access: Take Tanque Verde Road E to Tanque Verde Loop Rd. go S to Linden, W to King, S to end. Historic information: This may be close to the historic location: 3.5 mi. E Junction Redington Rd. on Mt. Lemmon Road 5 January 1953 A.G. Baker Following these directions exactly puts one on private land that does not appear to be suitable habitat, and the Mt. Lemmon Road goes NE. If the directions should have been “3.5 mi E of Junction of Redington Rd. and Mt. Lemmon Road” this would be very near this location. Vegetation conditions: This is a grove of tall, large diameter mesquite trees that seems to be in virgin condition. It apparently was someone’s home and acreage. Now the home is gone, there is a small amount of scattered junk. The only trees present appear to be velvet mesquite. There are few shrubs. The area is surrounded by low-density residential development. Methods: A total of 50 traps were placed in 3 parallel lines, from the fence on the West side of the property, extending eastward. Traps were placed at about 10 m intervals, and lines were approximately 10 m apart. Traps were placed on the afternoon of 21 December and retrieved on the morning of 22 December 2004. Rodents trapped: None. B-19 Site 10. Isabella Lee County Preserve Aerial photograph scale: 1: 3,000. B-20 Floodplain of Tanque Verde and Agua Caliente Wash. T14S R15E S2 UTM (Zone 12, NAD 27 CONUS): 520415mE 3567046mN. Map: Tucson East. Elevation: 2,550 ft. Ownership: Pima County Access: Tanque Verde Road E to Bonanza, then S to end of road. Historic information: This site includes the confluence of two major washes, Tanque Verde Creek and Agua Caliente Wash. This may be close to the historic location: 3.5 mi. E Junction Redington Rd. on Mt. Lemmon Road 5 January 1953 A.G. Baker However, at that time, this was private land that might not have been accessible to trapping. Vegetation conditions: Vegetation is a mosaic of several groves of tall, large diameter mesquite trees that seem to be in virgin condition, clusters of smaller mesquites, large mesquites, walnuts (Juglans major), a few cottonwoods and netleaf hackberries along the washes, and open areas dominated by grasses and annual plants. Prior uses of the site were for grazing. A sanitary sewer line crosses the site. Currently the site is heavily used by recreational equestrians, who have created multiple trails, and occasionally by birders, hikers, and mountain bicyclists. The site is surrounded by low density residential development. Methods: A total of 50 traps were placed in 2 parallel lines, beginning at the trail on the east side of the property and extending westward through the densest mesquites available. Traps were placed at about 10 m intervals, and lines were approximately 10 m apart. Variations from this spacing were made to avoid open areas with no trees. Traps were placed on the afternoon of 22 December and retrieved on the morning of 23 December 2004. Rodents trapped: None. A second survey, placing the traps along approximately the same lines, was conducted on 1314 January, with moon at new plus 3 days, to rule out potential adverse effects of moonlight on trap success at this site. Once again, no rodents were trapped. It appears that there are few nocturnal rodents on this site. The area has abundant round-tailed ground squirrels (Citellus tereticaudus) and nests of white-throated woodrats (Neotoma albigula), but none of these rodents were caught in traps. B-21 Site 11. Cienega Creek, Empire-Cienega Ranch Aerial photograph scale: 1:3,000 B-22 Floodplain of Cienega Creek, within Empire-Cienega Resource Conservation Area. T18S R17E S26. UTM (Zone 12, NAD 27 CONUS): 539445mE and 3522507mN. Map: Spring Water Canyon. Elevation: 4,280 feet Ownership: U.S. Government (BLM) Historical information: This area was formerly private land. merriami having been trapped here. There are no records of P. Access: Take entrance road to Empire-Cienega Conservation Area from Sonoita Highway. This is road EC901. Stay on it until it comes to unmarked side road leading down to an old agricultural field. Continue on the road around the field to NW corner, where there is a gate. Vegetation conditions: This area is shown on the Harris Riparian map as 224.53 Sonoran Riparian Deciduous Forest and Woodlands, Cottonwood-Willow Series and 143.10, Semidesert grassland. This is an area of some very large mesquites, basal diameter up to 6 feet, but most of them are smaller and of short stature. The understory is dense dead prickly Russian thistle (Salsola tragus) in places, and down wood in some places and, and much of it consist of dense, vigorously growing big sacaton. Trees present include mesquite, Goodding’s willow, Fremont cottonwood, and netleaf hackberry. Shrubs include wolfberry and lotebush. Methods: Fifty traps in two sets of 25 were set on January 1, 2005 and picked up on January 2, 2005. One set went upstream, the other downstream from the gate. Each set consisted of two parallel lines placed so as to remain within the mesquite-dominated area. Rodents trapped: Genus Peromyscus Peromyscus Peromyscus Species eremicus? eremicus? eremicus? TL T 180 85 175 92 175 95 HF E Sex 19 15 M 18 18 F 19 17 F B-23 Age Comments AD AD AD testes abdominal; no specimen Pregnant; no specimen Pregnant; no specimen Site 12. Arivaca Creek Access Site, Buenos Aires National Wildlife Refuge W A RI V A C A R Aerial photograph scale: 1:3,000. B-24 Floodplain of Arivaca Creek, at creek access point parking lot. Map: Arivaca. T21S R10E Section 19, NW ¼. UTM (Zone 12, NAD 27 CONUS): 465648mE 3495155mN. Elevation: 3,540 feet Ownership: United States of America (Buenos Aires National Wildlife Refuge). Access: Parking area for trail access, south side of Arivaca-Sasabe Road, approximately milepost 9. Historical Information: This site is close to that described as 2 mi N, 4 mi W Arivaca at which W. and L. Goodpaster trapped several P. merriami in March 1962. Vegetation conditions: The Harris Riparian Vegetation map indicates that this site is 143.10, Semidesert grassland, xeroriparian scrub. To the west of the parking area is a grove of mesquites that are approximately evenly aged and spaced, about 1 to 2 feet in basal diameter and 15 to 20 feet tall. Mid- and understory are not well developed in this area, except an understory of annual grasses and forbs. To the east of the parking lot is a narrow strip of big, old mesquites with a dense midstory of wolfberry, spiny hackberry, and lotebush and dense understory of mixed annual forbs. Methods: Fifty traps in two groups were placed at this location on the afternoon of January 15, 2005 and picked up on the morning of January 16, 2005. One group was a 5 by 5 trap grid with 10 meter spacing set in the mesquite grove on the west side of the parking area. The second group was a line through the dense vegetation on the east side of the parking area, with a right angle bend to parallel Arivaca Creek eastward. Measurements were taken only on animals in the genus Peromyscus. Rodents trapped: No small mammals were captured on the west side. At least ten of the traps had been disturbed, possibly by coyotes, foxes, or ravens, and one trap was missing and could not be relocated. On the east side, the following animals were trapped and released: Genus Species TL T HF E Sex Age Comments Peromyscus merriami 201 107 21 20 F Peromyscus merriami Neotoma albigula 195 100 20 20 F B-25 pregnant, cinnamon spot, specimen AD KJK512 pregnant, cinnamon spot, specimen AD KJK513 Site 13. McCafferty Canyon at confluence with Alamito Canyon Aerial photograph scale: 1:3,000. B-26 Rocky wash in McCafferty Canyon at confluence with Alamito Canyon. Map: Arivaca. T21S R10E S18. UTM (Zone 12, NAD 27 CONUS): 465463mE and 3496037mN Elevation 3,585 feet Ownership: State of Arizona. Historical Information: This site is close to that described as 2 mi N, 4 mi W Arivaca at which W. and L. Goodpaster trapped several P. merriami in March 1962. Vegetation conditions: The Harris Riparian Vegetation map indicates that this site is 143.10. The site is a narrow (<20 feet wide) rocky sandy wash lined with small mesquites and catclaw acacia. Desert broom, lotebush, and spiny hackberry were also present. Methods: Twenty-five traps were placed along the wash upstream from the road on the afternoon of January 15, 2005 and picked up on the morning of January 16, 2005. Rodents trapped: Genus Species TL T Peromyscus merriami 192 99 HF E Sex Age Comments pregnant, cinnamon spot, specimen 20 21 F AD KJK514 B-27 Site 14. Unnamed Rocky Wash. Buenos Aires National Wildlife Refuge No aerial photograph is available. An unnamed rocky wash located in T21S R9E S13. Map: Arivaca. UTM (Zone 12, NAD 27 CONUS): 464681mE 3496284mN. This wash is a tributary of Arivaca Creek. Elevation 3,557 feet Ownership: United States of America (Buenos Aires National Wildlife Refuge). Historical Information: This site is close to that described as 2 mi N, 4 mi W Arivaca at which W. and L. Goodpaster trapped several P. merriami in March 1962. Vegetation conditions: The site is not indicated as Class A or B mesquite mouse habitat, as originally defined. The Harris Riparian Vegetation map does not distinguish this site. The site consists of an incised wash and terrace of mesquites, most of the trees are <15 feettall and <1 feetin basal diameter, but some are much larger. Methods: Set 25 traps, 12 on S side of road, 13 on N side of road on the afternoon of January 15, 2005 and picked up on the morning of January 16, 2005. Rodents trapped: Only one animal, an Onychomys torridus, was trapped. No measurements were taken. B-28 Site 15. Mormon Lake, West Side, Buenos Aires National Wildlife Refuge Mormon Lake, on the west side of the lake. T22 S R8E S4. UTM (Zone 12, NAD 27 CONUS): 450450mE and 3489932m N. Map: Presumido Peak. Elevation: 3,500 feet Historical information: P. merriami was trapped at or “near” Mormon Lake on 19 December 1953 by K. E. Justice and L.D. Beatty. Access: The site is bordered on the east and west sides by dirt roads. Vegetation conditions: The site is a dense patch of fairly small (to 5 meters tall) mesquites. Tree and brush clearing has occurred on the west side of the patch, along the dirt road. There is little mid or under story vegetation. Methods: Standard operating procedure (see methods section). Two parallel lines of 25 traps each were placed on 16 January and retrieved on 17 January 2005. Measurements were taken only on animals suspected P. merriami. B-29 Rodents trapped: Genus Species Onychomys torridus TL T Peromyscus merriami 179 96 HF E Sex Age Comments 21 21 F AD emaciated; pregnant, cinnamon spot, specimen KJK517 22 20 F AD tail amputated and healed; pregnant, cinnamon spot, cream wash Dipodomys merriami Peromyscus merriami 170 74 Dipodomys Peromyscus Peromyscus Peromyscus Onychomys Dipodomys Onychomys merriami leucopus leucopus leucopus torridus merriami torridus B-30 Site 16. Upland Site, Buenos Aires National Wildlife Refuge No aerial photograph is available. To contrast with the small mammals trapped in the riparian areas, we also trapped the mesquiteinvaded grassland area between Arivaca Wash and Puertocito Wash, T20S R9E Section 32. UTM (Zone 12, NAD 27 CONUS): 458328mE and 3500771mN. Map: Las Guijas. Elevation: 3,430 feet Ownership: United States of America (Buenos Aires National Wildlife Refuge). Historical information: No historical information is available for this site. Access: The area is around a campsite along a dirt road leading northward from the Arivaca Road. Vegetation conditions: This is a mesquite-invaded grassland, with mesquites generally small (<10 feet tall and <1 feet basal diameter), with mixed native and non-native grasses. Methods: Standard operating procedure (see methods section). One line consisting of 25 traps was set on 15 January and checked on 16 January. Animals were minimally handled, not measured, weighed, or sexed. The same traps were reopened and rebaited in the evening of 16 January and picked up in the morning of 17 January, 2005. B-31 Rodents trapped: Genus Jan 16 Dipodomys Dipodomys Dipodomys Dipodomys Dipodomys Dipodomys Onychomys Chaetodipus Jan 17 Dipodomys Dipodomys Dipodomys Dipodomys Dipodomys Dipodomys Onychomys Chaetodipus Onychomys Species merriami merriami merriami merriami merriami merriami torridus baileyi merriami merriami merriami merriami merriami merriami torridus intermedius torridus B-32 Site 17. Mormon Lake, North Side, Buenos Aires National Wildlife Refuge No aerial photograph is available. Mormon Lake, an artificial impoundment on the Buenos Aires National Wildlife Refuge, along a dirt bank at the north end of the lake. T22S R8E S3. UTM (Zone 12, NAD 27 CONUS): 450831mE and 3490211mN. Map: Presumido Peak. Elevation: 3,500 feet Ownership: United States of America (Buenos Aires National Wildlife Refuge). Historical Information: P. merriami was trapped at or “near” Mormon Lake on 19 December 1953 by K. E. Justice and L.D. Beatty. Access: The site is easily accessed by means of a dirt road on the north side of the lake. Vegetation conditions: Vegetation consists of primarily mesquites growing along the earthen dam that creates the “lake” and north of the dam, where soil moisture was higher than surrounding areas. Mid- and under-story vegetation was very limited, and consisted of sparse wolfberry, snakeweed, and burroweed. The area is very heavily impacted by immigrant traffic and there was much litter on the ground. Methods: Standard operating procedure (see methods section). One line of 50 traps was set on 16 January and picked up on 17 January 2005. Measurements were taken only on animals in the genus Peromyscus. B-33 Rodents trapped: Genus Species Peromyscus leucopus T 63 HF 17 E 15 Peromyscus merriami 193 101 23 20 62 19 14 Peromyscus merriami 191 104 21 20 Peromyscus leucopus Onychomys torridus TL 134 142 Sex Age Comments F AD Pregnant Pregnant, cinnamon spot, cream F AD wash, specimen KJK515 M AD testes abdominal F Peromyscus leucopus B-34 AD Pregnant, cinnamon spot, cream wash, specimen KJK515 Site 18. West Branch of the Santa Cruz River, South A narrow wash with steep sides, at this location only about 3 or 4 feet deep. T14S R13E S27. UTM (Zone 12, NAD 27 CONUS): 500245mE and 3560720 mN. Map: Tucson. Elevation: 2,390 feet Ownership: Pima County. Historical information: The West Branch has been the focus of biological studies recently and has been found to include diverse plants (Maus 2003) and animals (Rosen 2003) that are considered representative of the Santa Cruz River valley in Tucson before the ecological devastation that affected most of the area. On both sides of the wash, within a few hundred meters, there are houses and house trailers, horse properties, and other domestic livestock. Feral cats were seen in the vicinity. As far as is known, this site has not been trapped for rodents. Access: Along the power line access road on the north side of the mobile home park at Freedom Lane and Ajo Road. Vegetation conditions: This is a mesquite-dominated wash, with some trees up to 2 feetin basal diameter, most smaller. The riparian vegetation at this site is generally only one or two trees wide, and is bordered by inactive agricultural land and residences. Many of the trees show signs of having been cut at some time in past decades. Methods: Standard operating procedure (see methods section). Two parallel trap lines consisting of 25 traps each were placed on each side of the wash on the afternoon of 21 March 2005 and retrieved in the morning of 22 March. Rodents trapped: None. B-35 Site 19. West Branch of the Santa Cruz River, North A narrow wash with steep sides, at this location only about 3 or 6 feet deep. T14S R13E S26. UTM (Zone 12, NAD 27 CONUS): 500565mE and 3561499 mN. Map: Tucson. Elevation: 2,380 feet Ownership: Pima County. Historical information: The West Branch has been the focus of biological studies recently and has been found to include diverse plants (Maus 2003) and animals (Rosen 2003) that are considered representative of the Santa Cruz River valley in Tucson before the ecological devastation that affected most of the area. On both sides of the wash, within a few hundred meters, there are houses and house trailers, horse properties, and other domestic livestock. Feral cats were seen in the vicinity. As far as is known, this site has not been trapped for rodents. Access: Park at the gate at the end of the County easement to the South of Church Wash and walk to the site. Vegetation conditions: This is a mesquite-dominated wash, with some trees up to 2 feetin basal diameter, most smaller. The riparian vegetation at this site is generally only one or two trees wide, and is bordered by inactive agricultural land and residences. Many of the trees show signs of having been cut at some time in past decades. B-36 Methods: Standard operating procedure (see methods section). Two parallel trap lines consisting of 25 traps each were placed on each side of the wash on the afternoon of 21 March 2005 and retrieved in the morning of 22 March. Rodents trapped: None. B-37 B-38 APPENDIX C FIELD KEY TO RODENTS LIKELY TO BE TRAPPED C-1 FIELD KEY TO RODENTS LIKELY TO BE TRAPPED Squirrels Not Squirrels Fur-lined cheek pouches........................................................................ HETEROMYIDAE No cheek pouches ............................................................................................ MURIDAE HETEROMYIDAE Soles of hind feet densely haired ................................................................................. Dipodomys 5 toes on hind foot........................................................................................................ ordii 4 toes on hind foot Size large, total length >300 mm................................................... spectabilis Size small, total length <260 mm ..................................................... merriami Soles of hind feet naked ................................................................ Perognathus and Chaetodipus TL < 150 mm, HF <20mm (tiny, soft and silky) .................................... Perognathus flavus TL>150, HF >20 mm (bigger, rougher looking) HF <22 mm, H and B <80 ................................................Chaetodipus intermedius HF >22 mm, HB >80 Tail much longer than HB, HB usually > 95 mm .......................................baileyi Tail slightly longer than HB, HB usually < 95 mm, pencil tip tail .... penicillatus MURIDAE Larger animal, > 200 mm Small ears ............................................................................................ Sigmodon arizonae Large ears ............................................................................................... Neotoma albigula Smaller animal, < 200 mm Upper incisors grooved on anterior face ..................................................Reithrodontomys Dorsum golden brownish, belly grayish white washed with buff, tail >77 mmfulvescens Dorsum grayish or brownish, venter grayish, tail < 77mm .......................megalotis Upper incisors not grooved Short tail, < 60% of HB ............................................................ Onychomys torridus Longer tail, > 60% HB ......................................................................... Peromyscus TAKE ALL MEASUREMENTS!!!!!!!!!!!!!!!!!!!!!!!!!! Tail shorter than body ....................................................................leucopus Tail longer than body TL < 190, Tail < 100, Body <85, HF <22, Ear < 20....................... eremicus TL > 190, Tail > 100, Body > 85, HF > 22, Ear > 20 COULD BE MERRIAMI!!! C-2