Austin, Z., Cinderby, S., Smart, J., Raffaelli, D. & White, P. Mapping wildlife: integrating stakeholder knowledge with modelled patterns of deer abundance by using participatory GIS. Wildl. Res. 36, 553-564 (2009).

This paper reports on a study that sought to integrate stakeholders into the issue of wildlife management by involving them in the GIS process used to model deer abundance in the East of England.  In this study area, where feral deer populalations are viewed as a nuisance or danger by some and as a resource by others, it can be difficult to reach a concensuus on management plans.  This is made more difficult by incomplete records on species numbers and distributions throughout the area.  The researches employed a strategy of ‘participatory GIS’ to fill in the gaps in their knowledge of the demography of these deer populations and to ground truth the models they developed to estimate density and distribution based on reported traffic accidents.  The model was used to create abundance maps in ArcGIS 9, which were used in conjunction with stakeholder interviews to incorporate their local knowledge of deer occurrences  with the predictions of the model.  The study found this approach flawed because each interviewee had ‘expert’ knowledge of a limited physical area of the study site,but were prone to alter parts of the map that were outside of this range.  Thus those alterations could not reliably be used in their final mapping.  Hoevere, it was a good tool for evaluating what the limiting knowledge is in creating real-time occupancy maps for a far ranging species.  This paper is a good example of using GIS to integrate the public into wildlife research.  Also, it describes some good techniques for incorporating non-spatial data into GIS work.

Broseth, H. & Pedersen, H.C. Hunting effort and game vulnerability studies on a small scale: a new technique combining radio-telemetry, GPS and GIS. Journal of Applied Ecology 37, 182-190 (2000).

This study combined the use of GPS devices and GIS mapping to look at the effects of hunting on Willow Ptarmigan in Norway.  The novel approach in this study was that in addition to the birds being tracked with radio-collars, the hunters were also tracked via GPS units given to them by the researchers.  The study collected about 50 days of data from the hunters and combined this in a GIS with the locations of radio tagged birds in the hunting area as well as their probable mortality based on radio-tagged birds that were collected by hunters.  This technique was modified from the use of satellites tracking fishing vessels to deduce the impact commercial fishing has on fish populations and inform decisions about fisheries management.  Garmin GPS recivers aere mounted on the hunters backpacks at the beginning of each hunt and were programmed to automatically create a way-point at 1-minute intervals; the hunters also took waypoints at each site of a ptarmigan kill.  These tracks were integrated into a gridded map of the study site and buffered to trepresent the path of a hunter and dog.  The whole area was overlayed with polygos representing individual home-ranges for radio tagged birds, and these intersections were clipped in ArcGIS, allowing a visual representation of the expended effort of a hunter to kill one bird.  The GPS tracks of the hunters were also used to create maps in ArcGIS that display the spatial distribution of hunting pressure for the study area.  This study was a very clever way to link GPS data from two different sources with GIS and take advantage of GIS tools to associate these in a meaningful way.  I will also be looking for the overlap of disparate features that may be influencing what areas of an island that birds use for breeding.

Rubin, E., Stermer, C., Boyce, W. & Torres, S. Assessment of Predictive Habitat Models for Bighorn Sheep in California's Peninsular Ranges. JOURNAL OF WILDLIFE MANAGEMENT 73, 859-869 (2009).

This study compared 3 models for predicting the habitat use of Bighorn Sheep in southern California’s Peninsular Ranges.  Two of these models are specific to use with GIS, the Ecological Niche Factor Analysis(ENFA) and the General Algorithm for Rule-set Production(GARP).  These framework models were tailored to Bighorn sheep in the study area using day time GPS coordinates collected from GPS-collared individuals in 5 subpopulations, and the resulting analyses were comnpared to analyses obtained from the non-GIS model that is recommended by the U.S. Fish and Wildlife recovery plan for Bighorn sheep.  Both GIS models proved to yield results comparable to the expert-based model, with the GARP model being slightly more statistically powerful than the ENFA model.  This is another good example of using GIS models to prove existing models and the power of GIS to offer various methods of achieving the same result.

Wong, S., Ronconi, R., Burger, A. & Hansen, B. Marine distribution and behavior of juvenile and adult Marbled Murrelets off southwest Vancouver Island, British Columbia: Applications for monitoring. CONDOR 110, 306-315 (2008).

 The authors sampled locations of adult and juvenile Marbled Murrelets off the coast of Vancouver Island, British Columbia, and used GIS to calculate distances and densities associated with various environmental and behavioral predictors.  These predictors include associations with juveniles and kelp beds, proximity to shoreline, proximity to other adults and juveniles and emigration from the study area.  The locations were obtained in two ways: through boat transect surveys 300m from the shoreline, with GPS waypoints of boat location being recorded at 1 minute intervals and with a theodolite from cliff tops overlooking the study area.  The authors used a kernel density analysis in GIS to look at the study site at a coarse scale and determined that the daily tendency of juveniles to associate closely with adults was not repeated at an annual level.  For their fine-scale analysis they looked at the theodolite data and found that adult densities were higher in areas with juveniles and also that juveniles were aggregated significantly closer to shore than adults were.  They also found that using counts of adults versus juveniles may bias your overall productivity estimates if not accounting for emigration of adults away from the study area.  This study is important to me because I am also using a theodolite to record the coordinates of seabirds in the field.  There are good resources in the methods section and literature cited section of this paper.

Doherty, K.E., Naugle, D.E. & Walker, B.L. Greater Sage-Grouse Nesting Habitat: The Importance of Managing at Multiple Scales. Journal of Wildlife Management 74, 1544-1553 (2010).

This paper looks at the differences in scale that affect habitat use in animals, and how this difference can be integrated into GIS.  Habitat can vary from a local scale to a landscape scale, and the whole gradient can influence where an animal will or can utilize space.  GIS is heavily relied upon when designing habitat models for myriad species and environments, but most often this considers only landscape scale vegetation patterns because of the extreme resolution required to extract local level vegetation characteristics from aerial photography or satellite imagery.  This study used field collected data to create local-scale habitat selection models as well as traditional landscape scale models and ground truthed their ability to predict breeding grounds for sage-grouse with the known locations in their study area.  They also created mixed scale models and these ended up being the best predictors of sage grouse nesting areas, proving that sage grouse use both fine and coarse scale landscape features to make decisions about where to nest.  This will allow management to more time and cost effectively select areas for reintroduction or habitat rehabilitation to increase sage-grouse nesting success.  This study was a good assessment of what people were not evaluating using GIS and how they can circumvent apparent stumbling blocks of GIS technology and produce more precise analyses.

Montgomery, R.A., Roloff, G.J., Ver Hoef, J.M. & Millspaugh, J.J. Can We Accurately Characterize Wildlife Resource Use When Telemetry Data Are Imprecise? Journal of Wildlife Management 74, 1917-1925 (2010).

 This study employed GIS to test the 5 most common techniques in wildlife science of correcting for inaccuracies of positional data obtained from telemetry.  This is important because error in positional accuracy can lead to the wrong classifications being made in relation to habitat preferences, home ranges, dispersion patterns and landscape use.  The authors created different covariate rasters to represent different types of landscapes and also different types of predictor variables.  All methods of telemetry correction were simulated with randomly generated points and were tested against both ‘categorical’ rasters, representing a land cover type predictor, and a ‘continuous’ raster representing distance or elevation.  Also, field collected positions of elk in South Dakota were used in the same method to validate the findings.  They found that patch size significantly affects the probability that the telemetry error will skew results of an analysis.  Also, it showed that of all techniques to correct for telemetry inaccuracy, ignoring the inaccuracy is just as good as making corrections.  This paper was a good reminder to account for inaccuracies in positional data collected in the field, and that validation of techniques is something that should be preformed often and incorporate the latest technology.

Danks, F.S. & Klein, D.R. Using GIS to predict potential wildlife habitat: a case study of muskoxen in northern Alaska. International Journal of Remote Sensing 23, 4611-4632 (2002).

This study used GIS to create models of habitat suitability for muskoxen in both winter and summer in the Alaskan Arctic, and then used these models to detect these habitats in the (NPR-A), which is land highly prized for both wildlife and petroleum development.  The muskoxen habitat models were developed by crating layers to represent those environmental attributes required by muskoxen in each season; these attributes include vegetation, elevation, slope and aspect characteristics, and known locations of muskoxen in the past.  When the model was applied to the NPR-A site, maps could then be generated showing the most suitable areas for muskoxen to use during each season, and these are a powerful tool when communicating with opposing stakeholder groups.  Because of the capabilities of ArcGIS, these maps were able to break the landscape up into suitability classes, showing the landscape as a gradient of suitable to unsuitable habitat as related to the underlying geomorphology and hydrology, still represented on the map.  This showcases the power of GIS to account for numerous disparate factors influencing species presence or absence in temporally variable ways.  It was also designed to be used and updated continually with passing seasons as more data is collected on the muskoxen and development in the study area.  I hope that the database I am building in my thesis will be used in much the same way- to understand patterns and the interactive effects of environment on species habitat use.

Sand, H., Zimmerman, B., Wabakken, P., Andrën, H. & Pedersen, H.C. Using GPS technology and GIS cluster analyses to estimate kill rates in wolf ungulate ecosystems. Wildlife Society Bulletin 33, 914-925 (2005).

 The authors used GIS in combination with data from GPS collars to estimate feeding behavior of Grey Wolves in Scandinavia.  This study was used to determine if traditional methods of estimating kill rates from daytime aerial surveys are accurately reflecting prey use by wolves, and if new technology can refine these models to make them more precise.  Traditional models assume 1-2 known locations per day per individual should accurately represent number of large game animals killed by a wolf pack for food.  This study fitted wolves with GPS collars that transmitted a location 1 an hour for the entire study period, day and night.  They buffered these points with 25, 50 and 100 meter buffers, projected them in ArcGIS over a gridded map of the study site, and used overlapping buffers to determine ‘clustering’, each cluster of points marking an individual prey carcass.  These GIS selected sites were then visited by the researchers to examine the effectiveness of the GIS system.  The GPS clustering technique identified about 93% of wolf-killed moose carcasses found in the field if using a 200 m buffer, but only 87% with a 100 m buffer, even with 1 location per hour being recorded by the GPS collar.  The researchers estimate that with the traditional regime of 1 or 2 locations per day, only about 10% of moose kills would be detected.  Also, using clusters recorded during daylight detected about 41% of carcasses, whereas night time clusters detected about 78% of carcasses.  This has serious ramifications for all prey modeling done for wolves using the traditional methods.  This paper showed an excellent way to integrate non-invasive mapping into hypothesis testing and model selection.

Adams, J.R., Kelly, B.T. & Waits, L.P. Using faecal DNA sampling and GIS to monitor hybridization between red wolves (Canis rufus ) and coyotes (Canis latrans ). Molecular Ecology 12, 2175-2186 (2003).

This study used GIS to link spatial data collected on Red Wolves and Coyotes in northeastern North Carolina with non-spatial DNA data to create maps used to determine the presence of hybridization in the study area.  The authors utilized genetic-sequencing techniques that can identify an animal to species from scat samples collected in the field. By recording GPS waypoints at each scat collection site and projecting these onto GIS layers of the study area and incorporated roadways, they could associate the home ranges of known Red Wolf individuals with each scat deposit.  By also associating the DNA species attribute to each scat, the authors were able to identify the home ranges of Red Wolf and Coyote hybrids and keep track of their numbers and whether additional hybridization was going on.  The authors were able to use the study to identify the home of a previously unknown hybrid individual and implement proper management plans, all in an unobtrusive manner involving no handling or trapping of the endangered Red Wolf individuals.  My thesis is also working on non-invasive mapping techniques for sensitive species.  This paper was very interesting because they are approaching their study with different techniques than my own, but working towards the same end result.

Stoleson, S.H., Kirschbaum, K.J., Frank, J. & Atwood, C.J. From the Field: Integrating GPS, GIS, and avian call-response surveys using Pocket PCs. Wildlife Society Bulletin 32, 1309-1312 (2004).

This study utilized the advances in small, personal computing devices, here Pocket PC’s, to digitize the process of conducting call-response, or tape-playback, bird counts, while also integrating GPS locations for these counts and habitat site characteristics.  Bird songs were converted into MP3 files and could be broadcast through the use of small portable speakers attached to the Pocket PC.  Additionally a small GPS unit was integrated with the Pocket PC to collect waypoints in the field.  The use of these small Pocket PCs not only significantly cut down on the amount of equipment carried by each researcher into the field, but allowed all data collected to be easily merged into one database and that database manipulated and analyzed in ArcGIS.  By running ESRI ArcPad, a GIS mapping system developed for use on PocketPC’s, all data collected in the field could be done in real time and stored as a shape file. These shape files could then be loaded into ESRI desktop software and used for spatial analysis purposes.  This short paper was a good example of researchers taking initiative in using the latest technological advances to streamline data collection and enhance the accuracy of field collected data.

Stedman, R. et al. Integrating Wildlife and Human-dimensions Research Methods to Study Hunters. Journal of Wildlife Management 68, 762-773 (2004).

This study incorporated GPS units given to hunters and traditional self-reporting questionnaires with GIS to give game managers a better idea of hunter land-use patterns.  It also explored the differences in GPS recorded hunting tracks and the distances self-reported by the hunters in the traditional questionnaires used to make ungulate management decisions during hunting season.  The authors also were able to link non-spatial data, attitudes about hunting and game management collect3ed via questionnaire, with the spatial data recorded by each GPS unit, through GIS.  This study is an expansion on the paper employing the same techniques on ptarmigan hunters in Norway by Broseth and Pedersen (2000).  The hunter movement data was supplemented with density of hunters data collected during aerial surveys of the study area.  During these surveys technicians were equipped with tablet personal computers and digitizing pens; the computers ran software showing detailed maps of the area from the technicians point of view from the airplane and the digitizing pen was used to plot the location of hunters as they were spotted during each transect of the aerial survey.  This data was used to create hunter-density probabilities for a map of the study site.  This data was then joined using ArcGIS to create detailed datasets linking behavior in the field and hunter characteristics.  My thesis project too seeks to create a dataset of non-spatial attributes linked with locational data, so this paper was helpful in thinking about which characteristics may be most useful in a graphical representation.

Newbold, S. & Eadie, J.M. Using species-habitat models to target conservation: A case study with breeding mallards. Ecological Applications 14, 1384-1393 (2004).

This study sought to improve on practices for choosing habitat restoration sites to benefit wildlife abundance.  To accomplish this they created a habitat preference model and used GIS to locate high priority restoration zones. They then integrated this information with maps characterizing the habitat types surrounding these high priority areas to take into account their influence on the selected site as well as the environmental preferences of the animals to make the best possible decision for locating a new wildlife restoration site.  They applied these techniques to wetland restoration for breeding Mallard ducks in California with great success.  The authors combined Breeding Bird Survey count estimates with land-use datasets for California to build an abundance model for mallards in the breeding season based on surrounding habitat types.  It was found that Mallards are positively associated with wetlands in close proximity to rice fields or other wetlands and negatively associated with proximity to orchards and urban areas.  It was also found that the specific spatial configuration of preferred habitat type in an environment is a significant factor in predicting Mallard abundance, not only total amount of preferred habitat.  ArcGIS was then used to predict what specific areas of restoration would be the most cost-effective to pursue, taking into account cost and location of land parcels, predicted bird abundance, proximity to preferred habitat and local habitat matrix.   This was a good paper for understanding the link between scientific research and effective management strategies.  It has given me some new management application ideas for my thesis project involving the habitat matrix of a colony and how this relates to reproductive success.

Sampson, B.A. & Delgiudice, G.D. Tracking the Rapid Pace of GIS-Related Capabilities and Their Accessibility. Wildlife Society Bulletin 34, 1446-1454 (2006).

 This paper categorizes advances in GIS capabilities between 1990-2005 as they apply to the study and management of wildlife, as applied to a long term data set on white-tailed deer.  The authors catalogue the ways in which they have employed GIS technology in the past, new applications that they found to be germane to their branch of research or methods of analysis and stumbling blocks they have experienced related to using GIS in the wildlife field and possibilities of avoiding these in the readers own work.  A detailed time line of their use of GIS programs gives the newcomer to ArcGIS a feel for the incredible advances that have been made possible in the last 20 years.  Additionally, Table 3 is a list of ArcGIS extensions, created by the Minnesota Department of Natural Resources and available for download, that are relevant to wildlife analysis; the table lists them by name and gives a brief description, while the text goes in to more detail about their applicability.  They also list other internet resources for ArcGIS tools, extensions and datasets that are publically available.  This paper is invaluable to me as a source of possible tools to enhance my own analyses.  It will also help me from making common mistakes that may slow my progress toward completion of my degree.

Kauffman-Axelrod, J.L. & Steinberg, S.J. Development and Application of an Automated GIS Based Evaluation to Prioritize Wetland Restoration Opportunities. Wetlands 30, 437-448 (2010).

This study utilizes the ArcGIS tool Model Builder to improve upon previous GIS based tools for evaluating wetland areas for restoration.  This tool allows for the wetland evaluation process to be region or even site specific to obtain the most accurate and relevant ant habitat analysis possible.  The authors also hoped to improve office based assessments by incorporating available environmental and hydrologic data sets publically available to reduce the number of field visits required before a site is declared as a priority restoration area.  The team developed GIS tools to automate the process of environmental evaluation at multiple spatial scales and for both vector and raster datasets.  This study was done on Oregon coastal wetlands and was heavily influenced by and incorporated with the work of Laura Brohpy on developing accessible and ‘transparent’ evaluation methods for prioritizing wetlands rehabilitation and restoration sites.

Ji, W. & Jeske, C. Spatial modeling of the geographic distribution of wildlife populations: a case study in the lower Mississippi River region. Ecological Modelling 132, 95-104 (2000).

This paper presents a case study for using GIS to take a visual approach to spatial modeling and analysis in wildlife management.  By integrating spatial data from radio-tagged Northern Pintails wintering along the Lower Mississippi River with environmental resource records from the area they were able to create a model for predicting pintail movement and distribution that could be viewed through ARC/INFO GIS, rather than being strictly numerical.  The field data derived from the radio-tagged pintails, location for each telemetry survey done was combined with attributes unique to each bird (sex, time/date/location of capture, age) in ARC/INFO GIS to create vector data maps.  This point data was integrated with detailed information about habitat types in the study area and environmental conditions favored by pintails in both summer and winter, and all data layers were related to one another by geography.  After assessing which environmental factors would have the most influence on pintail distribution, these assumptions were layered against actual pintail locations to search for patterns in specific demographic groups found in each habitat type.  Further, the researchers created custom GIS interfaces to provide them with the relevant information for each hypothesis testing.  This paper is germane to the ways in which I will be analyzing the field data collected for my thesis.  There are concept models included in the paper that will assist my decision making process when developing my own spatial analysis models in ArcGIS.