Carl G. Smith², Paul B. Hamel³ & Manoelle Fuzaro Gullo⁴

¹ U.S. Forest Service, Southern Research Center, Center for Bottomlands Research, who financed this work.
²Center for Bottomland Hardwoods Research, Southern Research Station, USDA Forest Service, carlsmith@fs.fed.us
³Center for Bottomland Hardwoods Research, Southern Research Station, USDA Forest Service, phamel@fs.fed.us. Corresponding author.
⁴Universidade Estadual Paulista, manoelle_ef@hotmail.com

Recepción: Octubre 4 de 2009/Aprobación: Enero 21 de 2010

Las especies de roble constituyen un componente evidente y frecuentemente dominante de los bos ques de tierras bajas del Valle Aluvial de Mississippi. Durante los últimos dos siglos, la extensión de estos bosques se ha reducido drásticamente como resultado del corte de árboles para el uso agríco la de tierras. Los patrones de tala han reducido la distribución de los parches de bosque restantes a un subconjunto del paisaje mucho más propenso a inundaciones que nunca antes se habían conocido en la historia, reduciendo la diversidad de especies de roble que actualmente existen en el paisa je. Talas intensivas han cambiado adicionalmente la composición de los rodales restantes. Pequeños parches restantes de los bosques primarios siguen existiendo como áreas naturales de investigación en el bosque denominado Delta National Forest ubicado en el Condado de Sharkey, Mississippi. En particular, las áreas naturales de investigación denominadas Roble Overcup (Quercus lyrata ) y Liquidámbar Americano (Liquidambar styraciflua) tienen componentes importantes de los árboles con que fueron nombradas, así como Quercus nuttallii y componentes más pequeños de otras especies. Un reciente interés en forestación ha causado el resurgimiento del interés en la restauración de bos ques de roble en las tierras agrícolas abandonadas de la región. Hemos estudiado la respuesta de pe queños mamíferos a esta restauración utilizando un experimento extensivo cerca del Bosque Nacional Delta (Delta National Forest) desde 1995. También examinamos la respuesta de pequeños mamíferos a un tornado que afectó aproximadamente la mitad del Área Natural de Investigación de Roble Overcup (Overcup Oak Research Natural Area) en 2008. Utilizamos estos estudios para demostrar la manera cómo se pueden obtener estas estimaciones de población de pequeños mamíferos con base en estudios de captura-recaptura, empleando diferentes diseños y utilizando el programa denominado Program Capture para calcular la población. Las comunidades de pequeños mamíferos que habitan en estos rodales son más especiosas en la sucesión temprana que en el bosque primario. El estudio de las respuestas a los daños que fueron causados por el tornado al área natural de investigación de roble Overcupes complicado por el hecho de que este tipo de bosque en particular es muy propenso a inundaciones, lo cual crea obstáculos a su colonización por pequeños mamíferos.

El análisis de los datos de captura-recaptura con métodos robustos como se ilustró en este artícu lo permite la extracción de más información que el análisis menos detallado por el mismo esfuerzo de campo en capturar pequeños mamíferos. Nuestro trabajo puede ser útil para otros interesados en el estudio de pequeños mamíferos en sistemas de bosques de roble en Centroamérica y Suramérica.

Palabras clave: disturbio, estimación poblacional, métodos de muestreo, Peromyscus, perturbación, programa de captura, Quercus.

ABSTRACT

Oak species form a conspicuous and often dominant component of bottomland forests of the Mississippi Alluvial Valley. The extent of these forests has been drastically reduced as a result of clearing for agri culture in the past two centuries. Patterns of clearing have reduced the distribution of remaining forest patches to a much more flood-prone subset of the landscape than was historically the case, reducing the diversity of oak species currently present on the landscape. Intensive harvesting has further changed the composition of the remaining stands. Small remnant patches of primary forest continue to exist as Research Natural Areas on the Delta National Forest in Sharkey County, Mississippi. In particular, the Overcup Oak (Quercus lyrata) and Redgum (Liquidambar styraciflua) Research Natural Areas pres ent substantial components of the trees for which the areas were named, as well as Quercus nuttallii and smaller components of other species. Recent interest in afforestation has produced a resurgence of interest in restoration of oak forest to abandoned farmland in the region. We have studied small mammal response to restoration on an extensive experiment near the Delta National Forest since 1995. We have also ex amined small mammal response to a tornado that disturbed approximately half of the Overcup Oak Research Natural Area in 2008. We use these studies to demonstrate how population estimates of small mammals can be obtained from capture-recapture studies, employing different designs, and utilizing Program Capture for population estimation. Small mammal communities in these stands are more species-rich in early succession than in primary forest. The study of response to tornado damage to the Overcup Oak Research Natural Area is complicated by the fact that this particular forest type is very flood-prone, creating obstacles to colonization by small mammals. Analysis of capture-recapture data with robust methods illustrated in this study permits extraction of more information from the same field effort expended in time-consuming small mammal trapping studies that have been subjected to less de tailed analysis. Our work may prove useful to others interested in study of small mammals in oak forest systems in Central and South America.

Key words: disturbance, population estimation, sampling methods, Peromyscus, disturbance, program capture, Quercus.

RESUMO

As espécies de robles constituem um componente evidente e freqüentemente dominante nos bosques de terras baixas do Vale Aluvial do Mississipi. Durante os últimos dois séculos, a extensão destes bosques se reduziu drásticamente como resultado do corte de árvores para o uso agrícola das terras. “Os padrões de corte reduziram a distribuição dos” parches” de bosque restantes a um subconjunto da paisagem muito mais propenso a inundações que nunca antes se havia visto na historia, reduzindo a diversidade de espécies de roble que atualmente existem na paisagem. Devastações intensivas sofreram modificações adicionadas a composição dos rodales existentes. Pequenos parches restantes de bosques primários seguem existindo como áreas naturais de investigação no bosque denominado Delata National Forest localizado no Condado de Sharkey, Mississipi.Em particular, as áreas naturais de investigação denominadas Roble Overcup (Quercus lyrata) e Liquidámbar Americano (Liquidambar styraciflua) têm componentes importantes das árvores com que foram nomeadas, assim como Quercus nuttallii e componentes mais pequenos de outras espécies. Um recente interesse na reflorestação causou o resurgimento do interesse na restauração dos bosques de robles nas terras agrícolas abandonadas da região.Estivemos estudando a resposta de pequenos mamíferos a esta restauração utilizando um experimento extensivo nas proximidades do Bosque Nacional Delta (Delta National Forest) desde 1995. Também estivemos examinando a resposta de pequenos mamíferos em um tornado que afetou aproximadamente a metade da Área Natural de Investigação de Roble Overcup (Overcup Oak Research Natural Area) na 2008. Utilizamos estes estudos para demonstrar a maneira como se pode obter estas estimações da população de pequenos mamíferos com base nos estudos de captura-recaptura, empregando diferentes desenhos e utilizando o programa denominado Program Capture para calcular a população. As comunidades de pequenos mamíferos que moram nestes rodales são mais especiosas na sucessão anterior do bosque primário.O estudo das respostas aos prejuízos que foram causados pelo tornado na área natural de investigação do roble Overcup é complicado pelo fato de que este tipo de bosque em particular é muito propenso a inundações, o qual cria obstáculos a colonização de pequenos mamíferos A análise de dados de captura-recaptura com métodos robustos como ilustrado neste artigo permite a extração de mais informação que a análise mais detalhada pelo mesmo esforço de campo em capturar pequenos mamíferos. Nosso trabalho pode ser útil para outros interesados no estudo de pequenos mamíferos nos sistemas de bosques de robles na América Central e América do Sul.

Palavras chave: Distúrbio, estimação populacional, métodos de amostragem, Peromyscus, perturbação, programa de captura, Quercus

Oak forests are important habitats for small mammals (Dickson 2001, Saitoh et al. 2008, Tioli et al. 2009) throughout the range of Quercus and presumably other genera in the family Fagaceae as well. Small mammals in mature and old-growth stands of temperate oak forests often are associated with coarse woody debris (Harmon et al. 1986, McMinn & Crossley 1996, Loeb 1999, Bowman et al. 2000, McCay 2000, Osbourne & Anderson 2002, Os bourne et al. 2005). Studies of small mammals in upland oak forests in North America are numerous (e.g. Greenberg 2002, Fantz & Renken 2005,Tietje et al. 2008); Hammond & Anthony (2006) evaluat ed 1535 data sets from a variety of habitats. Those in Central and South America are fewer (Sánchez- Cordero 2001, Otalora Ardila 2003, Navarro Arquez 2005, Lopez-Barrera & Manson 2006, van den Bergh & Kappelle 2006, Ramirez & Perez 2007, Aragon et al. 2009, Corredor Prado & Bejarano Bonilla 2009). Like investigations of larger mammals in these systems (Walker & Cardenas 2004, Cujar-Tovar 2006, Tobler et al. 2006), these studies have depended upon indices of activity rather than population size estimations to assess habitat relationships and community dynamics. Attempts to determine population size or relative abundance often have used less precise estimation methods such as Minimum Number Left Alive (Vazquez et al. 2000), Lincoln Index (Fleming 1974), or Jolly- Seber methods (Rojas Rojas & Barboza Rodriguez 2007). Recent work using more robust population estimation methods such as available in Program Capture (Rexstad & Burnham 1991) or Program Mark (White & Burnham 1999) has proved to be very useful and informative comparing small mam mal use of different habitats (e.g. Shanker 2000, Hadley & Wilson 2004). Wiewel et al. (2009) fur ther show that the robust methods are ultimately more time efficient than less precise ones.

Oak species form a conspicuous and often dominant component of bottomland forests of the Mis sissippi Alluvial Valley in North America (Hodges 1997). Clearing for agriculture in the past two centuries (MacDonald et al. 1979) has reduced the distribution of remaining forest patches to a more flood-prone subset of the landscape than previously, lowering the diversity of oak species currently present on the landscape. Intensive harvesting has further changed the composition of the remaining stands. Small remnant patches of primary forest such as the Overcup Oak ( Quercus lyrata) and Redgum (Liquidambarb styraciflu) Re search Natural Areas (RNA) on the Delta National Forest in Sharkey County, Mississippi, present substantial components of the trees for which the areas were named, as well as Quercus nuttallii and smaller components of other species (Devall & Ramp 1992). Small mammal communities of the old-growth stands have been little studied.

Recent efforts in afforestation have produced a resurgence of interest in restoration of oak forest to abandoned farmland in the region (Gardiner et al. this symposium). Concomitant with that increased interest have been questions concerning rehabilitation of wildlife communities associated with oak forests, and the restoration of this aspect of ecosystem function in the process of restoration. Parti cularly limited in this respect has been long-term study of small mammal community development in afforestation areas (Savage et al. 1996). Long-term study of small mammals in afforestation situations can extend our knowledge of the development of these communities and their functioning. Illustration of techniques of study of populations of small mammals in these ecosystems, existing in a matrix of agricultural lands, may form a useful reference situation for those contemplating conservation and restoration of oak ecosystems in mixed landscapes in South America.

In this report we present study designs and some results from studies of small mammal communities we have conducted in oak forests in Mississippi. In each of these studies we have used a capture-mark- recapture design with robust population estimation using Program Capture. We describe two such stu dies, (a) small mammal response to restoration on an extensive, long-term experiment near the Del ta National Forest, and small mammal population response to a tornado that disturbed approximately half of the Overcup Oak RNA in 2008. This work allowed us to test the null hypotheses (a) that res ponses of small mammals to different techniques of afforestation are similar among species, and (b) that tornado damage to an old-growth stand of oak forest growing in a floodplain situation has no effect on small mammal populations in the area. In addition to providing useful data on small mam mal communities and populations for these North American habitats, we suggest the techniques are applicable to similar questions in oak forests in Central and South America as well.

MATERIALS AND METHODS

STUDY AREAS

Delta National Forest (DNF) is located east of Rolling Fork, in the southern portion of Sharkey County, Mississippi, 32º 46’ N, 90º 47’ W, elevation 25-29 m

above mean sea level. The DNF is the sole unit composed entirely of bottomland hardwood forests in the US National Forest System. Managed by the U.S. Forest Service to produce multiple-use outputs of wood, water, forage, wildlife, and recreation, the DNF comprises 63000 acres in a single contiguous block, long the largest such tract of bottomland forest in the Yazoo River tributary basin of the Mississippi Alluvial Valley. Soils are predominantly heavy clay in composition, reflecting the low, flood-prone po sition of this landscape. Forest composition reflects the low position of the forest in the floodplain, with stands dominated by species characteristic of the more flood-prone sites. Within the DNF, three RNAs were set aside from commercial harvest activity, the Redgum, Overcup Oak, and Green Ash RNAs (Devall & Ramp 1992). We have worked in two of these RNAs. Redgum RNA (32º 54’ N, 90º 42’ W, elevation 28 m) lies on a relatively higher site representati ve of stands dominated by sweetgum (Liquidambar styraciflua; large old, individuals with distinctive red heartwood are called “Redgum” trees) and Nuttall Oak (Quercus nuttallii). Reflective of the higher site position, understories of this RNA are continuous and dominated by Sabal minor. The Overcup Oak RNA(32º 54’ N, 90º 44’ W, 27-28 m elevation) is situated on a lower site position, and dominated by a stand of the flood tolerant Overcup Oak ( Quercus lyrata) and Water Hickory (Carya aquatica). Understory com munities here are depauperate, presumably because the frequent and prolonged flooding in the site is too severe for many understory species. Each of these RNAs is approximately 16 ha in extent. On March 3, 2008, a tornado damaged some forest stands in the DFN extensively, including approximately half of the Overcup Oak RNA (National Weather Service 2008).

The Sharkey Large-Scale Afforestation Experiment is an approximately 800 ha area of abandoned farmland immediately to the north of the DNF (32º 58 N, 90º 54 W, elevation 28-30 m above mean sea level). Extensive growing-season flooding of parts of the site made it unprofitable for farming. The land was acquired in the 1990s by the U.S. Fish and Wildlife Service, and is managed as forestland for wildlife populations it might support. In 1995, staff of the US Forest Service Center for Bottomland Hardwoods Research and representatives of a number of other agencies and academic institutions established an experiment on the area, in order to investigate alternative methods of restoring margi nally productive farmland to forest in the Mississippi Alluvial Valley. Four afforestation practices reflective of thinking at that time were chosen. The experiment was established as a randomized, com plete block design of three blocks each consisting of four 8-ha plots to which the treatments were randomly assigned. The treatments, in order of intensity of labor effort and of initial cost of establis hment, were Natural Regeneration (NAT) involving no direct intervention; Direct Seeding ( SOW) of acorns of the site-appropriate Nuttall Oak, planted mechanically using specially modified agricultural planting equipment; Hand Planted (PLN) seedlings of Nuttall Oak; and a two-stage interplanting ( NUR) of a) a cottonwood (Populus deltoides) plantation designed to simulate succession and for early har vest and b) followed by hand planted Nuttall Oak seedlings two years later. Details of the treatments are available in Schweitzer et al. (1997). A ten-year rotation of the cottonwood plantation was antici pated at initiation of the experiment, during which additional treatments would be applied to smaller subportions of the treatment plots. These additional treatments were installed during harvest opera tions of the cottonwood stands in March 2008, and include four intensities of harvest of the cottonwo od trees, a complete harvest, a harvest followed by coppice regrowth of the cut cottonwood stems, a thinning removal of half of the cottonwood stems, and an uncut control.

We conducted small mammal sampling using Sherman live traps and capture-recapture methodology. We set a single trap on the ground at each predetermined location. We baited the traps with whole oats moistened with imitation vanilla flavoring, and mixed in an approximate ratio of 65ml vanilla flavoring per kg of oats (1 oz. of vanilla flavoring to 1 lb of oats). A small amount of polyester fiber fill was added to the trap as thermal insulation for animals captured on cool nights (< 10 ºC).

Traps were arrayed in the field in square grids, with trap spacing held constant for each study. At the Sharkey Large-Scale Afforestation Experiment we conducted a preliminary study of two plots with a grid of 100 traps spaced in a 10 x 10 array with individual traps 20 m apart, sampling an area of 3.2 ha. High density of captures and low numbers of recaptures in this study caused us to reduce the distance between traps to 10 x 10 m and to reduce the number of traps to 64, sampling an area of 0.49 ha. We have used the same trapping array in all subsequent work on that experiment. This effort in volves 448 traps in seven grids of 64 traps in each experimental block. The sampling unit is the trap ping grid.

On the Overcup Oak RNA, we arranged the traps in two scales. At the larger scale of the entire 16-ha RNA, we sampled at each of the 20 vegetation sampling plots were established on a grid consisting of 4 rows of plots 90 m apart with five plots spaced 70 m apart in each row, covering an area of 12.6 ha. Ten of these sampling plots were located in the portion of the RNA damaged by the tornado and ten in the undamaged portion. At each of the vegeta tion plots, the smaller scale of sampling, we arranged six traps at 20 m intervals around the periphery of the 0.08-ha vegetation sampling plots. A total of 120 traps was used in this study, 60 each in the un damaged and tornado damaged portions of the RNA.

Our trapping sessions extended for one week per location. In the Sharkey Large-scale Afforestation Experiment we randomly assigned our sampling blocks to weeks in the trapping season to account for potential confounding effects during the analysis. We set the traps on the first day of the session, and checked them once per day for five days. We did not disinfect our traps, a practice suggested to be followed in areas where hantavirus is endemic (Mills et al. 1995). Traps were checked once per day, during the morning hours. When traps contained animals, we shook the animal into a plastic bag for handling. We then cleaned, rebaited, and reset the trap at the original location.

Individual captures were identified to species, sex, and reproductive condition when possible, and their body mass, body length, tail length, hind foot length, and ear length measured. The specific location of the trap in which the capture occurred was noted, and each animal was given a numbered metal ear tag. Recaptured animals were weighed, their tag number, species, sex, and reproductive condition noted. Because of the difficulty of identifying Peromyscus species, we remeasured all indi vidual Peromyscus on each capture. Animals were released at the capture site.

Capture data resulting from each trapping session were summarized for each animal as a vector of day-location combinations for each capture grid. All capture data for each species from each grid in each session were analyzed in a single run of Program Capture. Analysis of the resulting popu lation estimates has been conducted using SAS. We accepted statistical significance at P = 0.05 in the work on the Sharkey Large-scale Afforestation Experiment, and at P = 0.10 in the Overcup Oak RNA study with its more limited sample sizes.

RESULTS

We have conducted 18 trapping sessions on the Sharkey Large-Scale Afforestation Experiment, 1995-2009 (264 days; 106,136 trap-nights; 13,623 captures); and one on the Overcup Oak RNA, 2009 (5 days, 600 trap-nights, 32 captures). Small mammal communities in these oak ecosystems in bottomland hardwood forests of the Mississippi Alluvial Valley are more species-rich in early suc cession than in primary forest. We have captured nine species in the Sharkey Large-Scale Affores tation Experiment (Table 1); and three on DNF, Neotoma floridana, and Peromyscus leucopus in Overcup Oak RNA during this work We also captu red Peromyscus gossypinus in the Redgum RNA in unrelated work in both 2008 and 2009 (P. B. Hamel & C. G. Smith, III, personal observations).

]]>

Colonizing communities of small mammals in the Sharkey Large-scale Afforestation Experiment are characterized primarily by Sigmodon hispidus and Oryzomys palustris, with smaller numbers of Peromyscus leucopus, Reithrodontomys humulis, and Mus musculus. Sampling conducted after ten years of succession on the site captured primarily the same species, at different frequencies of oc currence. O. palustris and S. hispidus are still the dominant species. P. leucopus has increased significantly in frequency, while S. hispidus has decli ned, and R. humulis has virtually disappeared. Mus musculus is significantly less frequent than in the initial stage of colonization as well. Neotoma floridana has appeared in the fauna, as have Glaucomys volans and Cryptotis parva. Sylvilagus aquaticus, which we have captured only incidentally, became frequent and abundant enough to hunt early in the development of the stands; these rabbits are ge nerally distributed on the site. The species occurs within the Delta National Forest as well. The tree squirrels, Sciurus carolinensis and S. niger occur in the forest but are not yet present in the affores tation site as no hard mast is yet produced. Like the rabbits, these animals are difficult to capture with the trapping scheme we have employed. We have captured only one Peromyscus nuttallii in this work; this animal characteristic of bottomlands is one we have expected to capture regularly. The scarcity of this animal is noteworthy (Barrett & Feldhamer 2008).

We encountered two small rodents in the trapping on the Overcup Oak RNA in 2009 (Table 2). Two Neotoma floridana, one of them a juvenile, indicate that this species is present on the RNA in num bers too low to estimate populations separately for the tornado-damaged and undamaged portions of the RNA. We were able to estimate populations of P. leucopus for the entire area and for the tornado-damaged and undamaged portions of the area se parately. Two growing seasons after the tornado, populations of this species are uniformly distribu ted across the RNA (Table 2).

Populations of Peromyscus leucopus on the Sharkey Site in November 2008 (Table 3) reflect the difficulty of estimating populations of small mammals in capture-recapture studies without analysis using robust methods as exemplified in Program Capture. Three different models were necessary to estimate populations of these animals on the site, that involving heterogeneity among animals in cap ture probability [M(h)], heterogeneity among days and individuals in capture probability [M(th)], as well as the null model of uniform capture probability among days and individuals [M(o)]. Analysis of variance of population estimates using block and treatment as main effects yields a significant result (F_8.12 = 4.05, P < 0.02, R² = 0.73), in which block is a significant effect (F = 9.2, P < 0.01), but treatment is not (F = 2.3, P = 0.1). The inference from this result is that the species is generally distributed among the treatments in this experiment after 13 growing seasons, but that significant variations in population distribution exist, external to those factors controlled in the experimental design.

DISCUSSION

Small mammal capture-recapture data are time-consuming to obtain, and subject to a variety of con founding variables that make interpretation of the results problematical. Each of these factors affects the likelihood of capture or recapture of individual animals, and affects each species in a particular way (Otis & Burnham 1978, White et al. 1982). Hence determination of relationships between spe cies and the sampled oak ecosystems is complicated. Simply expressing the number of captures obtained during a session as a function of the effort expended, while an interesting reflection of the be havior of the animals during the individual trapping session, is insufficient for determining population responses or for comparing species abundances or responses to treatments. Several existing analytical programs are available which can estimate populations, select appropriate mathematical formulations of analysis that account for particular combinations of confounding effects, and provide standard errors of population estimates as well. Two of these are Program Capture (Rexstad & Burnham 1991) and Program Mark (White & Burnham 1999). We began using Program Capture in the initial stages of the Sharkey Large-Scale Afforestation Experiment and continue to do so here. Program Mark is a more recent development which we have not used. (Note: The use of trade or firm names in this publi cation is for reader information and does not imply endorsement by the United States Department of Agriculture of any product or service.) The robust methods utilized in these analyses are applied to field data similar to those gathered by workers in oak forests in Central and South America (Fleming 1974, Van den Bergh & Kappelle 1998, Vazquez et al. 2000, Sánchez-Cordero 2001, and Rojas Rojas & Barboza Rodriguez 2007). We can extract more information from the data after these analyses than can be obtained by analysis of the very same data using less robust methods.

We are unable to reject the null hypothesis that tornado damage to an old-growth stand of oak forest growing in a floodplain situation has no effect on small mammal populations in the area, as popula tions of P. leucopus were similar in both tornado-damaged and undamaged areas of the Overcup Oak RNA two growing seasons after the storm. Shortly after the sampling reported here, a flood inundated the site completely. We are uncertain of the specific effect of that flood on the populations of small ma mmals in Overcup Oak RNA without further sampling effort.

We are able to reject the null hypothesis that responses of small mammals to different techniques of afforestation are similar among species, based upon frequency of occurrence data across the 13 seasons of the sample. We can further examine the response of one of the species within an individual season, taking advantage of the detailed population estimates provided by the Program Capture analy ses. Clearly, P. leucopus is more widespread on the site now than in the beginning of the experiment (Table 1). The great variability in distribution of P. leucopus populations across the Sharkey Large- Scale Afforestation Experiment as exemplified by the capture data for November 2008 (Table 3) in dicates that additional explanations must be sought to explain the variations in distribution, explana tions specific to the Sharkey Large-Scale Afforestation Experiment rather than to the distribution of P. leucopus in afforested locales in the Missis sippi Alluvial Valley. We offer the following as a hypothesis for some of the variability, subject to test in additional work. The explanation reflects specific events that occurred within Block 1 of the experiment, as follows. For reasons external to the design of this experiment, the natural re generation treatment plot in Block 1, adjacent to the cottonwood nurse crop treatment plot, has a very dense stand of Cornus sp., swamp dogwoods, which provide abundant cover, a situation not re flected in the natural regeneration treatment plots in the other blocks in the study. Furthermore, in the fall of 2005 a portion of the stand of cottonwoods in the cottonwood-nurse crop treatment plot, whose trees were already weakened by attack of wood- boring insects (Saperda calcarata), was damaged by the passage of a hurricane. By the spring of 2007, we could detect qualitatively a response of P. leucopus to the increase in downed woody debris produced by the storm breaking the susceptible trees (Figure 1, where sampling plots are indivi dually identified). All 3 animals in NURM and all 5 animals in NURW were in the NW corner part of plot where the storm damaged the vegetation and one animal moved between the plots during the sampling week. Of two animals captured in NURS, one was a female captured several days in the plot, and the other a male who had been captured in NURM and NURW as well. Harvesting activity took place shortly after and eliminated the potential source of coarse woody debris.

CONCLUSIONS

Capture-mark-recapture methods are in widespread use to sample small mammals in oak forests throughout the range of the Quercus and related species. Robust analytical methods have been employed in North America, Europe, and Asia, but less so in Central America and South America. These analytical methods permit extraction of more information from the same field data sets than do earlier methods. We document succession in small mammal communities in oak restoration areas, lack of effect of tornado on populations of a small mammal in primary oak forest in Mississippi, and complex responses of the same species to intratreatment variations in environmental conditions within a successional study. Our work may prove useful to others interested in study. of small mammals in oak forest systems in Central and South America.

ACKNOWLEDGEMENTS

We are grateful for the support of the staff of the Center for Bottomland Hardwoods Research at each phase of this work. Fieldwork on the Sharkey Large-Scale Afforestation Experiment involved a very large number of colleagues, too numerous to mention individually. Keith Willis and Chris Wo odson deserve special mention. Crews of colleagues from the University of Memphis Department of Biology helped us in the early years of the experiment. Sampling designs were greatly impro ved through discussion with Bob Cooper. Plínio Gonçalves de Oliveira and Raquel Partelli Feltrin assisted with the work done on the Redgum Research Natural Area. Raquel Partelli Feltrin and Fernanda Scheffer Alves de Lima worked with us on the Overcup Oak Research Natural Area. John Stanturf stimulated us to undertake the work on the Sharkey Site. Margaret Devall and Ralph Pearce made the work on the Research Natural Areas on Delta National Forest possible. Nathan Schiff iden tified the wood boring insects for us. Rosa Isabel Ojeda Martinez kindly translated the abstract into Spanish for us. The manuscript received very useful comments from Tom Dell and an anonymous reviewer.

BIBLIOGRAPHIC REFERENCES

Aragon, E.E., A. Garza & F.A. Cervantes. 2009. Structure and organization of rodent assembles of a forest of the Sierra Madre Occidental, Durango, Mexico. Revista Chilena de Historia Natural 82: 523-542.        [ Links ]

Barko, V.A. & G.A. Feldhamer. 2002. Cotton mice (Peromyscus gossypinus) in southern Illi nois: Evidence for hybridization with white-footed mice (Peromyscus leucopus). American Midland Naturalist 147: 109-115.        [ Links ]

Barrett, G.W. & G.A. Feldhamer. 2008. The golden mouse: ecology and conservation. New York: Springer-Verlag.        [ Links ]

Bowman, J.C., D. Sleep, G.J. Forbes & M. Edwards. 2000. The association of small mammals with coarse woody debris at log and stand scales. Forest Ecology and Management 129: 119-124.        [ Links ]

Corredor Prado, J.P. & D.A. BejaranoBonilla. 2009. Pequeños mamíferos no voladores de la reserva natural Ibanasca (Tolima, Colombia). Revista Tumbaga 4: 121-134.        [ Links ]

Cujar-Tovar, A. 2006. Uso del hábitat del venado (Mazama rufina) en la Reserva Biológica Cachalú y su área de influencia, pp: 101 - 118. En: Solano, C & N. Vargas. (eds.) Memorias del I Simposio Internacional de Robles y Ecosistemas Asociados. Fundacion Natura-Pontificia Universidad Javeriana Bogotá        [ Links ]

Devall, M.S. & P.F. Ramp. 1992. U.S. Forest Service Research Natural Areas and protection of old growth in the south. Natural Areas Journal 12: 75-85.        [ Links ]

Dickson, J.G. (ed.)2001. Wildlife of southern forests: habitat and management. Blaine, Washington: Hancock House Publishers. 480 p.        [ Links ]

Fantz, D.K. & R.B. Renken. 2005. Short-term landscape-scale effects of forest management on Peromyscus spp. mice within Missouri Ozark Forests. Wildlife Society Bulletin 33: 293-301.        [ Links ]

Fleming, T.H. 1974. The population ecology of two species of Costa Rican heteromyid rodents. Ecology 55: 493-510.        [ Links ]

Gardiner, E.S., D.C. Dey & J.A. Stanturf. 2010. Approaches to restoration of oak forests on far med lowlands in the eastern U.S. Revista Colombia Forestal Vol. 13 (2): 223-236.        [ Links ]

Greenberg, C.H. 2002. Response of white-footed mice (Peromyscus leucopus) to coarse woody debris and microsite use in southern Appala chian treefall gaps. Forest Ecology and Management 164: 57-66.        [ Links ]

Hadley, G.L. & K.R. Wilson. 2004. Patterns of density and survival in small mammals in ski runs and adjacent forest patches. Journal of Wildlife Management 68: 288-298.        [ Links ]

Hamel, P.B. 2003. Winter bird community differences among methods of bottomland hard wood forest restoration: results after seven growing seasons. Forestry 76(2): 189-197.        [ Links ]

Hammond, E.L. & R.G. Anthony. 2006. Mark-recapture estimates of population parameters for selected species of small mammals. Journal of Mammalogy 87: 618-627.        [ Links ]

Harmon, M.E., J.F. Franklin, F.J. Swanson, P. Sollins, S.V. Gregory, J.D. Lattin, N.H. Anderson, S.P. Cline, N.G. Aumen, J.R. Sedell, G.W. Lienkaemper, K. Cromack Jr. , K. W. Cummins. 1986. Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research 15: 133-302.        [ Links ]

Hodges, J.D. 1997. Development and ecology of bottomland hardwood sites. Forest Ecology and Management 90: 117-125.        [ Links ]

Loeb, S.C. 1999. Responses of small mammals to coarse woody debris in a southeastern pine forest. Journal of Mammalogy 80: 460-471.        [ Links ]

Lopez-Barrera, F. & R.H. Manson. 2006. Ecology of acorn dispersal by small mammals in montane forests of Chiapas, Mexico, pp. 165- 176. En: Kappelle, M. (ed.). Ecology and conservation of Neotropical montane oak forests. Springer-Verlag Berlin        [ Links ]

McCay, T.S. 2000. Use of woody debris by cotton mice (Peromyscus gossypinus) in a southeastern pine forest. Journal of Mammalogy 81: 527-535.        [ Links ]

MacDonald, P.O., W.E. Frayer & J.K. Clauser. 1979. Documentation, Chronology and Future Projections of Bottomland Hardwood Habitat Loss in the Lower Mississippi Alluvial Plain. Vol. 1. HRB-Singer. Inc. State College. Pennsylvania.        [ Links ]

McMinn, J.W. & D.A. Crossley Jr. (eds.). 1996. Biodiversity and coarse woody debris in southern forests, proceedings of the workshop on coarse woody debris in southern forests: effects on biodiversity; 1993 October 18-20; Athens, GA. Gen. Tech. Rep. SE-94. Department of Agriculture, Forest Service, Southern Research Station. Asheville, N.C. U.S.        [ Links ]

Mills, J.N., T.L. Yates, J.E. Childs, R.R. Parmenter, T.G. Ksiazek, P. E. Rollin & C. J. Peters. 1995. Guidelines for working with rodents potentially infected with Hantavirus. Journal of Mammalogy 76: 716-722.        [ Links ]

National Weather Service. 2008. Storm Data and Unusual Weather Phenomena. Acceso: March 2008. Disponible en: http://www.srh.noaa.gov/jan/stormdata/data/mar2008.pdf.        [ Links ]

Navarro Arquez, E. 2005. Abundancia relativa y distribución de los indicios de las especies de mamíferos medianos en dos coberturas vegetales en el santuario de flora y fauna Otún Quimbaya, Pereira-Colombia. Trabajo de Grado. Pontificia Universidad Javeriana, Facultad de Ciencias. Bogotá. 78 p.        [ Links ]

Osbourne, J.D. & J.A. Anderson. 2002. Small mammal response to coarse woody debris in the central Appalachians. . Proceedings of the Southeastern Association of Fish and Wildlife Agencies 56: 198-209.        [ Links ]

Osbourne, J.D., J.T. Anderson & A.B. Spurgeon. 2005. Effects of habitat on small-mam mal diversity and abundance in West Virginia. Wildlife Society Bulletin 33: 814-822.        [ Links ]

Otálora Ardila, A. 2003. Mamíferos de los bosques de roble. Acta Biologica Colombiana 8: 57-71.        [ Links ]

Otis, D.L., K.P. Burnham, G.C. White & D.R. Anderson. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monographs 62: 1-135.        [ Links ]

Ramírez, H.E. & W.A. Pérez. 2007. Mamíferos de un fragmento de bosque de roble en el Departamento del Cauca, Colombia. Boletin Cientifico Universidad de Caldas 11: 65-79.        [ Links ]

Rexstad, E. & K.P. Burnham. 1991. Users Guide for Interactive Program CAPTURE. Colorado Cooperative Fish & Wildlife Research Unit. Colorado State University. Fort Collins. Colorado.        [ Links ]

Rojas Rojas, L. & M. BarbozaRodriguez. 2007. Ecología poblacional del ratón Peromyscus mexicanus (Rodentia: Muridae) en el Parque Nacional Volcán Poás, Costa Rica. Revista de Biología Tropical 55: 1037-1050.        [ Links ]

Saitoh, T., J.O. Vik, N.C. Stenseth, T. Takanishi, S. Hayakashi, N. Ishida, M. Ohmori, T. Morita, S. Uemura, M. Kadomatsu, J. Osawa, K. Maekawa. 2008. Effects of acorn abundance on density dependence in a Japane se wood mouse (Apodemus speciosus) population. Population Ecology 50: 159-167.        [ Links ]

Sánchez-Cordero, V. 2001. Elevation gradients of diversity for rodents and bats in Oaxaca, Mexico. Global Ecology and Biogeography 10: 63-76.        [ Links ]

Savage, L., J. Anthony & R. Buchholz. 1996. Rodent damage to direct seeded willow oak in Louisiana. Proceedings of the Annual Conference of the Southeastern Association of Fish and Wildlife Agencies 50: 340-349.        [ Links ]

Schweitzer, C.J., J.A. Stanturf, J.P. Shepard, T.M. Wilkins, C.J. Portwood & L.C. Dorris ,Jr. 1997. Large-scale comparison of reforestation techniques commonly used in the Lower Mississippi Alluvial Valley: first year results. pp. 313-320. En: S.G. Pallardy, R. A. Cecich, H.S. Garrett & P. Johnson (eds). Proceedings of the 11th Central Hardwood Forest Conference; 23-26 March 1997, Columbia, MO. Gen. Tech. Rep. NC-188. US Department of Agriculture, Forest Service, North Central Forest Experiment Station. St. Paul, MN.        [ Links ]

Shanker, K. 2000. Small mammal trapping in tropical montane forests of the Upper Nilgiris, southern India: an evaluation of capture-recapture models in estimating abundance. Journal of Biosciences 25: 99-111.        [ Links ]

Tietje, W.D., D.E. Lee & J.K. Vreeland. 2008. Survival and abundance of three species of mice in relation to density of shrubs and prescribed fire in understory of an oak woodland in California. Southwestern Naturalist 53: 357-369.        [ Links ]

Tioli, S., F. Cagnacci, A. Stradiotto & A. Rizzoli. 2009. Edge effect on density estimates of a ra diotracked population of yellow-necked mice. Journal of Wildlife Management 73: 184-190.        [ Links ]

Tobler, M.W., E.J. Naranjo & I. Lira-Torres. 2006. Habitat preference, feeding habits and conservation of Baird's Tapir in Neotropical montane oak forests. pp. 347-362 En: M. Kappelle (ed. ). Ecology and conservation of Neotropical montane oak forests. Springer-Verlag. Berlin. Germany.        [ Links ]

Van den Bergh, M.B. & M. Kappelle. 1998. Diversity and distribution of small terrestrial rodents along a disturbance gradient in montane Costa Rica. Revista de Biología Tropical 46: 331-338.        [ Links ]

Van den Bergh, M.B. & M. Kappelle. 2006. Small terrestrial rodents in disturbed and old- growth montane oak forest in Costa Rica, pp. 337-345. En: Kappelle, M. (ed.). Ecology and conservation of Neotropical montane oak forests. Springer-Verlag. Berlin. Germany.        [ Links ]

Vázquez, L.B., R.A. Medellín & G.N. Cameron. 2000. Population and community ecology of small rodents in montane forest of western Mexico. Journal of Mammalogy 81: 77-85.        [ Links ]

Walker, R. & E. Cárdenas. 2004. Evaluación del estado de conservación de la fauna en el municipio de Murillo, Tolima. Boletin Cientifico. Universidad de Caldas : 15-29.        [ Links ]

Wiewel, A.S., A.A. Yackel Adams & G.H. Rodda. 2009. Evaluating abundance estimate precision and the assumptions of a count-based index for small mammals. Journal of Wildlife Management 73: 761-771.        [ Links ]

White, G.C., D.R. Anderson, K.P. Burnham & D.L. Otis. 1982. Capture-recapture and removal methods for sampling closed populations. Los Alamos National Laboratory LA-8787- NERP. Los Alamos, New México. 235 p.        [ Links ]

White, G.C. & K.P. Burnham. 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study 46 Supplement: 120-138.        [ Links ] ]]> 2009 82 523-542 2002 147 109-115 2008 2000 129 119-124 2009 4 121-134 2006 101 - 118 1992 12 75-85 2001 2005 33 293-301 1974 55 493-510 2010 13 2 2 223-236 2002 164 57-66 2004 68 288-298 2003 76 2 2 189-197 2006 87 618-627 1986 15 133-302 1997 90 117-125 1999 80 460-471 2006 165- 176 2000 81 527-535 1979 1 1996 1995 76 716-722 National Weather Service 2008 2005 2002 56 198-209 2005 33 814-822 2003 8 57-71 1978 62 1-135 2007 11 65-79 Colorado State University 1991 2007 55 1037-1050 2008 50 159-167 2001 10 63-76 1996 50 340-349 1997 313-320 2000 25 99-111 2008 53 357-369 2009 73 184-190 2006 347-362 1998 46 331-338 2006 337-345 2000 81 77-85 2004 15-29 2009 73 761-771 1982 1999 46 120-138