Ecological Archives E094-084-D1
D. A. Kelt, P. L. Meserve, J. R. Gutiérrez, W. Bryan Milstead, M. A. Previtali. 2013. Long-term monitoring of mammals in the face of biotic and abiotic influences at a semiarid site in north-central Chile. Ecology 94:977. http://dx.doi.org/10.1890/12-1811.1
Small Mammal Metadata
Introduction
Our long-term research program was established in 1989 to evaluate the relative role of competition and predation in structuring small mammal assemblages, as well as cascading effects on the plant community. In the 3rd and 4th years of this study our site experienced an El Niño / Southern Oscillation (ENSO) warm (high rainfall) event which resulted in a near-doubling of the annual rain (233 and 229 mm in 1991 and 1992, respectfully; the long-term mean from 1989 to 2005 was 139.4 mm). Another ENSO warm event occurred in 1997, and a prolonged one in 2000–2002. Reflecting our developing recognition of the priority of abiotic influences at this site the focus of our research transitioned to one of the relative importance of biotic (predation, competition) and abiotic (rainfall) influences on biotic composition and functioning. In recent years we have begun to focus on long-term stability and dynamics of the biotic assemblage at our site in the face of likely climate change. This metadata file outlines our field methods and the structure of the small mammal data set.
Metadata
Class I. Data set descriptors
A. Data set identity: Long-term monitoring in the face of biotic and abiotic influences at a semiarid site in north-central Chile.
B. Data set identification code:
Fray_Jorge_Small_Mammals_19890-200512.txt
C. Data set description
Current Principal Investigators:
Douglas A. Kelt, Department of Wildlife, Fish, & Conservation Biology, University of California, Davis, CA 95616 USA
Peter L. Meserve, Department of Biological Sciences, Northern Illinois University, DeKalb, IL 60115 USA
Julio R. Gutiérrez, Departamento de Biología, Universidad de La Serena, La Serena, Chile
Abstract: The data set covers a 16-year period (1989–2005) of monthly surveys of small mammal. The small mammal fauna is trapped every month on all experimental plots and information on each captured small mammal is contained in this file. These data have been applied to diverse population- and community-level analyses and remain useful for investigations of population dynamics, species interactions, and the relative influence of biotic and abiotic influences. Sampling is ongoing and data will be added over time.
D. Key words: climate; competition; LTREB data; population dynamics; predation; semiarid ecosystem; small mammals; species interactions.
Class II. Research origin descriptors
A. Overall project description
Identity: Capture data for every small mammal caught in our study site in the Quebrada de Las Vacas located in Bosque Fray Jorge National Park, IV Region, Chile.
Originators: See main metadata file metadata.php.
Period of Study: 1989–2005, ongoing.
Objectives: To monitor the community- and population-level dynamics of semiarid-zone small mammals and characterize the influence of competition, predation, and rainfall on small mammal assemblage structure and composition, and cascading influences on vegetative structure and composition.
Sources of funding: See main metadata file metadata.php.
B. Specific subproject description
1. Site description: See main metadata file metadata.php.
2. Experimental/Sampling design: See main metadata file metadata.php.
Treatments: See main metadata file metadata.php., Table 1, for details on treatment assignments for each plot.
Data Collection Period, Frequency: Small mammals are trapped for four days/month/grid (5 × 5 stations, 15-m interval, two traps/station). As noted in the Master Metadata File, degus (Octodon degus) are removed from degu exclusion plots (e.g., “-D-P” and “-D+P” treatments), and all small mammals are removed from small mammal exclusion plots (“-SM”).
3. Research Methods
Field: We have a small house on site, and our field crew lives there for nine days every month (second–tenth day of the month). They trap eight grids for four nights followed by the other eight grids for four nights (from 1997–2002 the crew added two +F grids to each 4-d cycle). Two Sherman-type live traps (9 × 11 × 30 cm) are placed at each station (hence, 50 traps per grid) and checked 2–3 times daily (morning, late afternoon, midday under hot conditions). All individual small mammals are uniquely marked with numbered ear tags or leg bands (National Band and Tag Co., Newport, Kentucky, USA), and standard data are recorded (see Table 1). Because our site experiences fog or high clouds frequently, we do not target sampling to a particular lunar phase.
Taxonomy and systematics: Taxonomy of small mammals is consistent with the third edition of Mammal Species of the World (Wilson and Reeder 2005) with the minor exception that Abrocoma bennetti is mis-spelled therein as Abrocoma bennettii (although the epithet is spelled correctly in the Index).
4. Project personnel: The field crew includes four technicians; over the years we have supported 27 technicians, some in excess of 10 years on the project. Additionally, the crew frequently recruits volunteers to assist with monthly surveys.
Class III. DATA SET STATUS AND ACCESSIBILITY
A. Status
Latest Update: December 2005.
Latest Archive date: December 2005.
Metadata status: The metadata are complete and up to date.
Data verification: Reflecting the complexity of our data set, data verification is a lengthy and iterative procedure.
Definitions:
Record
Line of data.
Master Capture Table
File containing data for all records through the history of the project.
CaptureID
A sequential and unique number assigned to each capture record when entered to the Master Capture Table.
Tag
Refers to uniquely numbered identifiers applied to small mammals. These include ear tags, leg bands, and Michel tags (surgical clips).
AnimalID
A unique number that corresponds to the most recent tag applied to the animal. AnimalIDs are updated through the Master Capture Table whenever tags are replaced.
Identification Change Table
File containing records of all tag changes. The Identification Change Table is not included in this Data Paper.
NewNum
When tags are replaced, this is the new tag number; NewNum becomes the AnimalID throughout the Master Capture Table, until and unless this tag is again replaced.
OldNum
When tags are replaced, this is the old tag number.
Corrections Log
File containing records of all records with data inconsistencies and the corrections made by the data managers (see points 12–14, below). All corrections are referenced to the CaptureID in the Master Capture Table. Corrections Logs are developed for each database update, and are not included in this Data Paper.
Procedures:
1. Field crews record all capture data on standardized field data sheets.a. Each capture record is assigned a sequential and unique number (CaptureID).8. Once the data have passed the visual verification they are appended to the project’s Master Capture Table in a Microsoft Access database. This table includes all monthly capture data since the project began in March 1989.
b. All data fields are reviewed to ensure they are complete and properly formatted.
c. Fields are sorted or organized with Pivot tables to verify that all data are properly formatted, within range of accepted values, and standard abbreviations and codes are used. Inconsistencies are checked against field data sheets or returned to the lead technician for verification if necessary.
CaptureID |
NewNum |
OldNum |
Comment |
34345 |
28995 |
23456 |
23456 replaced with 23987; updated to 28995 after 2nd tag change, below. |
38567 |
28995 |
23987 |
23987 replaced with 28995 |
Queries are run to check for AnimalID attributed to multiple species or genders.13. All records with data inconsistencies are reviewed individually by the data managers.
b. AnimalIDs with multiple records for a single trap revision (same day and hour) are also flagged with a query as they are likely errors (at times we have animals that are released and immediately enter a trap at a different station, but this is uncommon).
c. A list of AnimalIDs with data inconsistencies is generated and all records for these individuals are written to a temporary Microsoft Excel table. This table is sorted by date to see the capture history for each individual with an inconsistency.
a. Most data inconsistencies can be corrected by visual inspection of the data.14. A log of all changes (Corrections Log) is kept. This includes the original data inconsistencies in the records and the correction or change is recorded in a Change Comments field. The Master Capture Table is then updated by incorporating the change to the appropriate record based on the CaptureID, and adding a code in the comments indicating a change had been made and where to find the details in the Corrections Log (i.e., name of the specific Log and the date of the data revision). Currently the Corrections Logs are kept in a set of Microsoft Excel files named by the date of the data revision.i. For each AnimalID the capture dates, location (grid and trap station), weight, sexual and reproductive state, and ancillary observations are used to verify that the records represent a single individual.b. Data inconsistencies that remain unresolved visually require greater effort. In most cases these inconsistencies result from errors in tag reading or data entry.
ii. If the data are consistent with a single individual, logic and majority rule are used to infer the sex and species. That is, we change the record with an inconsistency to the value recorded in the majority of occasions if more than one previous record exists. Records are then updated to reflect the majority decisions.
iii. When no majority exists (e.g., recorded once as male and once as female) the sex and/or species is set to “unknown” (U or UN respectively); future consistency checks may resolve these. That is, if in a later occasion an AnimalID with a U for the sex is recorded as female then this will be detected during data consistency check and the sex for all the records of this animal will be changed to F.
i. These records are reviewed and compared to the original field data sheets to check for data entry errors. It is largely for this purpose that field data sheets are stored as pdf files on a secure ftp server.
ii. If the AnimalID, sex, species, and revision were entered correctly we check to see if the AnimalID is similar to other animals known to be captured at the location with the same species, sex, sexual, and reproductive state and a similar weight. Common errors such as recording a 7 for a 9, the omission of a leading or trailing number, or reading a number backwards, are corrected and logged in a spreadsheet of the Corrections Log.
iii. Inconsistencies in the AnimalID that cannot be resolved are left unchanged for several field sessions.
iv. If these are not resolved within several field sessions the least convincing AnimalID is listed as unresolved by changing the AnimalID to <NULL> and coding the change in the database with a comment (“99”) and the date of the corresponding Corrections Log. These records are removed from future analysis of the data.
B. Accessibility
Storage location and medium: (Ecological Society of America data archives [Ecological Archives], URL published in each issue of its journals). Original mammal data files reside with Douglas A. Kelt and W. Bryan Milstead, and are archived in a separate server. Original datasheets reside with Julio R. Gutiérrez. Small mammal datasheets also are scanned and stored on a server maintained by W. Bryan Milstead.
Contact person: Douglas A. Kelt, Department of Wildlife, Fish, and Conservation Biology, University of California, Davis CA 95616-5270, USA. Phone: +1 (530) 754-0145 E-mail: dakelt@ucdavis.edu.
Copyright restrictions: None.
Proprietary restrictions: We impose no proprietary restrictions on these data, but we do request that authors pursuing analytical efforts with these data communicate with us in advance to prevent redundant efforts. Additionally, we request that authors of publications using these data notify Douglas A. Kelt (contact information above) of publication of their study. This helps us to document the scientific value and use of these data and will contribute to our annual reporting to the funding agencies.
Costs: None.
Class IV. DATA STRUCTURAL DESCRIPTORS
A. Data Set File
Identity: Fray_Jorge_Small_Mammals_198903-200512.txt
Size: 368,759 rows (including header), 17.06 Mb.
Format and storage mode: ASCII text, tab-delimited.
Header information: The first row of the file contains the variable names below.
Alphanumeric attributes: Mixed.
B. Variable information
Table 1. Column information for Fray_Jorge_Small_Mammals_198903-200512.txt. Each row in this data set comprises all data collected for each individual small mammal captured during a survey. Multiple captures of individuals are recorded on separate lines. A period indicates missing data.
Variable name
Variable definition
Variable codes, definitions, and notes
MO
Year and month.
Year and month: e.g., March, 2000 = 200003.
DATE
Julian date, numbered sequentially beginning September 1, 1988.
Integer.
HOUR
Time of trap revision.
0 = Morning;
1 = Regular afternoon revision;
2 = On hot days, a second morning or mid-day revision is added (occurs before Hour = 1);
3 = On a few occasions, a second mid-day revision was added (after Hour = 0 and Hour = 2 but before Hour= 1).GR
Grid number.
1 to 20
ROW
Trapping Row.
1 to 5;
If row unknown entered as 0.COL
Trapping Column.
1 to 5;
If col unknown entered as 0TRT
Treatment
+D -P = Plus Degu Minus Predator
-D -P = Minus Degu Minus Predator
+D +P = Plus Degu Plus Predator
-D +P = Minus Degu Plus Predator
+F = Plus Food
-SM = Minus all Small Mammals (-D-P plots converted to –SM in 2000)AnimalID
A unique number that corresponds to the most recent tag applied to the animal.
Integer
SP
Species.
See Table 2.
SEX
Gender of animal.
M = Male;
F = Female;
U = UnknownWT
Weight (Grams).
Integer values (no decimals).
“.” = no data.SST
Sexual State.
0 = unknown;
1 = imperforate (♀)/ testes not descended (♂);
2 = perforate (♀)/ testes descended (♂).RST
FEMALE Reproductive State.
0 = non-reproductive or male;
1 = pregnant;
2 = recently or presently lactating;
3 = pregnant and lactating;
“.” = no dataST
Status of animal
0 = New. Animal not previously captured;
1 = Previously captured and marked;
2 = Remarked. Animal obviously previously marked but tag lost, previous number unknown;
3 = Changed. Animal in danger of losing a tag (ear torn; band tight) so new tag applied. Old tag is recorded and archived;
“.” = no data.F
Fate of animal
0 = Released;
1 = Died in trap;
2 = Injured in trap or during handling;
3 = Escaped;
4 = Repeated capture (during the same trap check; e.g., 2nd capture on a given morning;
5 = Animal removed from the grid. Release site unknown or different from 6 or 7, below; pertains to -D and –SM grids only;
6 = Animal removed from the grid, released between grids 2 and 3; pertains to -D and –SM grids only;
7 = Animal removed from grid, released between grids 11 and 12; pertains to -D and –SM grids only.
Table 2. Species codes employed in Fray_Jorge_Small_Mammals_198903-200512.txt, and general characteristics of these taxa. Codes were established in 1989 and some names have changed (hence, ME for Thylamys elegans, which was Marmosa elegans in 1989).
Code
Species
Common Name
Species Type1
Diet2
Weight in grams (SD)
AB
Abrocoma bennetti
Bennett’s chinchilla rat
Opportunistic
Herb
171 (65)
AL
Abrothrix longipilis
Long-haired akodont
Opportunistic
Insectiv.
47 (9)
AO
Abrothrix olivaceus
Olive-colored akodont
Core
Omniv.
28 (6)
CH
Chelemys megalonyx
Large long-clawed akodont
Rare
Insectiv.
34 (5)
LU
Octodon lunatus
Moon-toothed degu
Rare
Herbiv.
136 (36)
ME
Thylamys [Marmosa] elegans
Elegant fat-tailed mouse opossum
Core
Insectiv.
22 (8)
OD
Octodon degus
Degu
Core
Herbiv.
128(29)
OL
Oligoryzomys longicaudatus
Long-tailed colilargo
Opportunistic
Graniv.
27 (7)
PD
Phyllotis darwini
Darwin’s pericote
Core
Herbiv./ Graniv.
54 (14)
RN
Rattus norvegicus
Norway rat
Rare
Omniv.
128 (44)
RR
Rattus rattus
Black rat
Rare
Omniv.
110 (50)
SC
Spalacopus cyanus
Coruro (fossorial)
Rare?
Herbiv.
80-140g3.
1. Milstead et al. (2007).
2. Meserve (1981).
3. Weight from Muñoz-Pedreros and Gil (2009).
Class V. Supplemental descriptors
A. Publications using the data set: See main metadata file metadata.php.
Literature cited
Meserve, P. L. 1981. Trophic relationships among small mammals in a Chilean semi arid thorn scrub community. Journal of Mammalogy 62:304–314.
Milstead, W. B., P. L. Meserve, A. Campanella, M. A. Previtali, D. A. Kelt, and J. R. Gutiérrez. 2007. Spatial ecology of small mammals in north-central Chile: role of precipitation and refuges. Journal of Mammalogy 88:1532–1538.
Muñoz-Pedreros, A., and C. Gil. 2009. Orden Rodentia. Pages 93–157 in A. Muñoz-Pedreros and J. Yáñez Valenzuela, editors. Mamíferos de Chile, Second edition. CEA Ediciones, Valdivia, Chile.
Wilson, D. E., and D. M. Reeder. 2005. Mammal species of the world: a taxonomic and geographic reference. Third edition. Johns Hopkins University Press, Baltimore, Maryland, USA.