Annotated Bibliography

Oregon State University: GEOG 560, GIScience I: Introduction to Geographic Information Science)

Annotated Bibliography


Skye Kreisler

Posted: December 1, 2019


My capstone project for the Master of Natural Resources is titled “Evaluation of managed habitat during the non-breeding period for mallard duck (Anas platyrhynchos) at Bald Knob National Wildlife Refuge in central Arkansas“. The goal is to determine which biophysical attributes of a cover type may affect habitat selection or avoidance by mallard ducks, if any. Cover types include afforestation, cypress brake, mudflat, marsh (natural wetland), crops such as rice, corn, millet and milo, and layout (corn or rice laid out due to failure of crops that year). Ducks and habitat characteristics were surveyed over thirteen wintering periods (November through March; 2006-2019) and entered into an Access database. My project is to analyze the data and address the question as to whether ducks are using a habitat type more or less than expected based on availability or other biophysical variables. While other species such as American black duck (Anas rupripes) are on the decline, suggesting a higher sensitivity to habitat loss through land use/land cover change, mallards are considered to be more of a generalist when it comes to habitat selection, and population trends remain stable or increasing despite being a popular game species.

Image of federal duck stamp available in Texas, U.S.A.

Mallards are economically and culturally important as a game species in Arkansas, and my project will help to inform refuge management decisions in letting parcels to farmers for agricultural use which will sustain or increase wintering mallard populations on the refuge. This annotated bibliography provides examples of ways in which a geographic information system (GIS) might be used to find relationships between ducks and habitat selection at different spatial and temporal scales. With anthropogenic land cover change and shifting vegetation from climate change, finding such relationships in support of conversation for declining species has growing importance. Using a GIS to depict such relationships can help with land management and land use decisions, particular for national wildlife refuges which manage for multiple species.


Below is an annotated bibliography of fifteen articles from peer-reviewed, scientific journals, plus one unpublished M.S. thesis (Although the latter is not available to the public online, I included it because it has been the primary source for the GIS shapefiles used in my graduate project for Bald Knob NWR). Each study has either involved using a GIS or has an application for GIScience relating to my graduate research project on habitat selection by ducks. Although most of the articles discovered during this process pre-dates modern usage and availability of computers, it does not pre-date GIScience which is attributed to the mid-to-late 20th century. Additionally, even literature within the past decade often does not incorporate utilization of a GIS in the data analysis. As such, I have included a few such articles and described how a GIS might have benefited either the study itself or the presentation/reporting of the information to the reader. This annotated bibliography therefore only represents a selection of the full bibliography for my project.

Creating and posting this annotated bibliography has several benefits. The first of which is to demonstrate how GIScience and the use of a GIS benefits research, particularly that which pertains to my project. The second is that the literature research unsurfaced additional resources for my project. While reading article for relevance, I explored citations within the papers themselves and found even more literature! Completing this assignment in the graduate-level GEOG560 course has converted me into a fan of annotated bibliographies for future research project. Not only does this force a person to more thoroughly read multiple scientific articles on their topic of choice, it encourages comprehension and retention of the information in the writing of a summary of the study.


Abd Rahman, A. (2015). The Importance of Geographic Information System (GIS) and remote sensing in protecting migratory bird habitat population at Matang Mangrove Forest, Perak. Sains Malaysiana44(9), 1229–1239.

The premise of this article is to illustrate the efficient and accurate methods of using a GIS for modeling and analyzing relationships between wildlife and habitat. In the study, variables such as temperature, precipitation, food availability, elevation, land use and land cover were entered as layers in a GIS. Migratory bird counts and locations had been recorded over a ten year period in the Matang Mangrove Forest in Malaysia. Overlaying the distribution of birds on a map, and using statistical analysis, the author was able to identify the most significant parameters which were affecting migratory bird distribution. As a scientific paper from a region other than North America, this paper demonstrates how visual representation and modeling of data/results can be useful not only in improvement in reporting to a lay audience, but for overlaying multiple variables and finding relationships between multiple variables. This information would have otherwise been presented in a data table, and is therefore limited to a scientific community. This article is relevant to my research as a resource for me to analyze multiple factors in habitat selection by mallard ducks.

Ashley, P., K. A. Hobson, S. L. VanWilgenburg, N. North, and S. A. Petrie (2010). Linking Canadian harvested juvenile American black ducks to their natal areas using stable isotope (δd, δ13c, and δ15n) methods. Avian Conservation and Ecology 5(2):7.

Focused on American black duck, this article demonstrates connectivity between breeding and harvest areas based on genetic analysis. The researchers compared stable isotopes in feathers from harvested juvenile ducks with those of hatchlings. Using a combination of Bayesian statistical and a geographic information system (GIS), the researchers mapped the probability of origin for each bird.  Of the four subregions of Canada which were analyzed, fewer than between 5% and 32% (M=15%) of individuals originated from and moved through agricultural areas, suggesting that A. rubripes avoids stopover and use of agricultural landscapes as habitat. The information supported the hypothesis that regional differences in land cover influence distribution and abundance of this species. Variation in habitat quality (low quality = sink; high-quality = source), may affect the population growth or decline of ducks, as well as indicate selection versus avoidance. Modeling for source-sink population dynamics, habitat attributes and landscape variables during breeding and non-breeding periods may serve as a means to inform habitat management on wildlife refuges and hunting areas. American black duck populations have been observed to be declining while mallard ducks have been increasing, which supports that mallards may be more habitat generalists than black ducks and therefore more resilient from impacts of land conversion to agriculture. 

Booms, T. L., F. Huettmann and P. F. Schempf (2010). Gyrfalcon nest distribution in Alaska based on a predictive GIS model.(Report). Polar Biology33(3), 347–358.

While this article did not involve ducks, it does produce an example of extrapolation of species information using a predictive GIS model. For gyrofalcon nest distribution across a large remote landscape in Alaska, the researching employed twelve variables including slope and aspect, mean temperature and precipitation, and cover type. Using published literature on general habitat characteristics for gyrofalcon nest sites, the predictive model produced probability of nests in a given locale based on the interpolated twelve variables. The methodology is described in detail within the article. This type of model and data analysis as described in this study is useful for conservation and management of wildlife. Many species have distribution and abundance for which quantification is inexact due to density (swarms and flocks), behavior (elusiveness, hostility, mimicry, rapid movement, migration, etc.), inaccessibility (jurisdictional boundaries, terrain hazards, remoteness, fossorial, marine, etc.), or other traits. In the case of mallard duck, estimates of population are generally rounded to thousands or hundreds. Field research is tedious and requires multiple observers, and is therefore inconsistent and imprecise for scheduling and weather events. Another disadvantage to manual surveys is double-counting if the same individuals move from one observation point to another during a survey. In order for refuge management to make decisions regarding crop types and when/where to flood at Bald Knob NWR, using a GIS to develop a predictive model for the species of interest could be a useful tool.  Although in remote areas such as the gyrofalcon habitat in Alaska, ground trothing would not be possible, the model does remove many of the aforementioned uncertainties as well as the time-consuming and sometimes impossible surveys.

Callicutt, J. T., H. M. Hagy, and M. L. Schummer (2011). The food preference paradigm: a review of autumn-winter food use by North American dabbling ducks (1900-2009). Journal of Fish and Wildlife Management 2(1):29–40.

I chose this article specifically because it did not utilize geographic information science, but would have benefited from an application of a geographic information system (GIS). Valuations of plant taxa as foods for dabbling ducks (such as mallards) has largely not been investigated thoroughly. The authors assert that 89% of plant taxa prescribed values in management guides have not been researched for energetic values and metabolization in ducks. It is further suggested that previous methods used to evidence habitat selection have been biased, and yet management guides continue to rely on that data. By using literature review on the food preference paradigm for dabbling ducks, the authors attempt to evaluate habitat selection at scales used by wintering dabbling ducks. After reading this article and looking at several confusing data tables, here is my own assertion for consideration. While locating energy and Kcal values in previous literature is certainly an admirable feat, particularly since it relies on laboratory analysis of specific plant taxa, compiling such data into a GIS for spatial analysis would provide the greatest benefit to management decisions. On private land, landscapes and agriculture are largely dictated by economic and social values and not by ecological or duck-friendly considerations. On many wildlife refuges in the southeastern United States, parcels are leased out for agricultural use and then flooded post-harvest in order to provide wintering habitat for migrating birds. If energetic value of plant taxa, namely crop plants like rice, millet and milo or endemic moist-soil plants such as smartweed (Polygonum spp.) were compiled into a table within a GIS, these could then be mapped. In a raster map, patterns of habitat selection would be readily identifiable with a high to low energetic value “relief” across a landscape. Annual bird counts would be depicted visually according to the map, and either support or fail to support the current food preference paradigm prevalent on wildlife refuges. If managing for non-game species of migrating birds for conservation or even for multiple-use areas, this application of using a GIS is promising.

Cramer, D., P. Castelli, T. Yerkes, and C. Williams (2012). Food resource availability for American black ducks wintering in southern New Jersey. Journal of Wildlife Management 76(1):214–19.

In this study, Cramer et al. used a GIS to map the predicted habitat selection of wintering American black ducks, whose populations are declining while mallard populations are not. The researchers compared soil core samples to determine seed availability across five predicted habitat selections of American black ducks. The seed densities were used to determine the habitat type and the availability of those habitats were then mapped and compared to duck counts from those areas. A flaw of this study, which reduces its credibility, is that sample periods occurred over only two wintering periods (2006-07; 2007-08). This study was therefore on too small of a temporal scale to provide conclusive data. However, it demonstrates how even a simple study can be a launching point for future research; a replicated study over time and space using similar methodology and a GIS would add additional credibility to this research.

Davis, B. E., and A. D. Afton (2010). Movement distances and habitat switching by female mallards wintering in the lower Mississippi alluvial valley. Waterbirds 33(3):349–356.

Surveying for mallard movement and distribution in the Mississippi Alluvial Valley, which is considered part of the central flyway used by many migratory waterfowl, the researchers had to plan out how to successfully obtain the information for their study. The female mallards first had to be captured and either tagged with a GPS device which would send out cardinal coordinates to a central receiver periodically over multiple years (until the battery expired), or collared/marked so as to be detectable with radio telemetry. Using a combination of aerial imagery and ground truthing, the researchers ascertained different habitat types around the study area. The employment of a GIS was used in a rudimentary sense, rather than as part of their study or analysis. The authors of this paper used a GIS to map out private land parcels versus public lands (such as national forest, state or county park, or wildlife refuge) and accessibility by vehicle or by foot. Additionally, using the same criteria, the researchers discerned on the public lands which areas were open or closed for hunting at the time of observation. As movement to avoid hunters was presumed to be an external factor in duck movement which would affect the habitat study, the researchers were able to plan their observation dates and locations accordingly.  The conclusion of this study showed that female mallard movement during the non-breeding period was minimal if a single habitat type met both dietary and roosting needs. Female mallards selected for forested wetlands they would remain in that environment, whereas those which selected for flooded agricultural plots would not later move significant distances during the winter. Male mallards were not studied.

Forcey, G. M., G. M. Linz, W. E. Thogmartin, and W. J. Bleier (2007). Influence of land use and climate on wetland breeding birds in the prairie pothole region of Canada. Canadian Journal of Zoology 85(3), 421–436.

Using broad and small spatial and temporal scales, the authors evaluated environmental variables affecting migratory bird breeding sites in the Canadian prairie pothole ecosystem. The article includes a literature review of habitat associations to construct models of wetlands and predictions of bird abundance under different conditions.  A GIS was used to illustrate general land cover and climate data across the landscape, with habitat-related studies at multiple scales. Multiple species were analyzed, including ducks, grebes, terns, and marsh birds, incorporating historical observations and bird counts into the current data set. Habitat attributes were compared with selection for each species. The study was able to portray species’ distribution across habitat types, suggesting selection and avoidance based on landscape variables.

In this article, the benefits of using a GIS to locate errors in data and spatial correlation is acknowledged, as well as the ease of running Bayesian statistics and population models at different scales. In migratory birds, duration of stopover, brood-rearing, site fidelity, and life history may vary by species, so the ability to analyze habitat selection across a large landscape over time is beneficial to understanding influences and trends in bird behavior. The methods, use of GIS, and reporting from this article may be applicable to my own research of observing patterns in mallard behavior across a larger landscape, particularly if my research expands beyond the single wildlife refuge in Arkansas.  

Green, A. W., and D. G. Krementz (2008). Mallard harvest distributions in the Mississippi and central flyways. Journal of Wildlife Management 72:1328–1334.

Green and Krementz (2008) used a GIS to show density maps of mallard ducks in the central flyway from 1980-2003. Duck hunting is economically important to Arkansas, and mallard duck is the most popular game duck in that state. Arkansas’ hunters had been expressing concern for several decades that mallards were getting more difficult to locate, which both hunters and scientists at the time attributed to “short-stopping”. Shortstopping is a phenomenon where migrating birds stop short of their ancestral destination (wintering grounds or stopover sites) if suitable habitat is located prior to arrival. In this case, the concern was that agriculture and wetlands caused by anthropogenic impoundments and diversion in the Ohio, Missouri and Mississippi floodplains were more attractive to mallards than the natural forested wetlands and marshes of Arkansas. This study mapped out densities and distributions of mallards over the twenty-three year period, and concluded that there was little evidence to support the supposition that a northward shift of wintering mallards had been occurring. This is particularly interesting in the course of my project, because it discredits short-stopping as a reason for perceived bird declines. If there has been any spatial shift in mallard distribution within Arkansas, further research is needed to determine reasons other than land conversion to agriculture.

Hagy, H. M., and R. M. Kaminski (2015). Determination of foraging thresholds and effects of application on energetic carrying capacity for waterfowl. PLoS ONE 10(3).

This is another example of an article which would have benefitted from utilizing a GIS to portray data. In this study, core samples were obtained to ascertain seed distribution across a landscape, and energetic values of each plant were recorded and compared with duck counts and apparent selection. The researchers concluded that numerical counts of dabbling ducks were influenced by factors other than food density (energetic value), and that numerical values of duck counts are not a predictor of habitat quality if that habitat quality is gauged by seed and tuber density. The information provided by these studies is essential to my own research on duck habitat selection, which is progressing into energetic carrying capacity for managed impoundments on the refuge. Scientific journals do provide a resource for other researchers, students, and resource managers, but the audience is inherently self-selecting. It was somewhat difficult to understand this article, as written, and a visual representation on a map might make the information more accessible to a broader audience. While I have read numerous articles by both authors, respectively, very few of them incorporate use of geographic information science beyond basic maps. Perhaps that is due to the length of article submission in journals or other factors, but there are several studies by either or both author for which a GIS might have proved useful for interpreting and reporting the data.

Herbert, J., A. Chakraborty, A. Naylor, L. Beatty, and W. Krementz (2018). Effects of landscape structure and temporal habitat dynamics on wintering mallard abundance. Landscape Ecology 33(8):1319–1334.

In this study, a two-step Bayesian spatio-temporal model with spatial random effects was used in the prediction of mallard abundance in the Mississippi Alluvial Valley (MAV) over a seven-year period. The authors describe this as a spatially- and temporally-dynamic habitat study, and the spatial performance of the model was evaluated to identify where the model might explain local mallard abundance. Within the methods section for this study, it is stated that ESRI ArcGIS was utilized for all geoprocessing and mapping. In that regard, a GIS was used to calculate the area of landscape types, including extent of seasonally-flooded land and seasonal surface water. Overall, there were 25 classifications of landscape structure types included in the study. In the model, a 2 km. by 2 km. grid was overlaid on the different cover types over a 100 x 400 km. area of the Mississippi Alluvial Valley, and each cell in the grid was assigned a value representing mallard abundance or density observed during the seven years of surveys. The map/model was then assigned a color value and a visual representation of mallard abundance across the landscape was portrayed. While more mallards were predicted by the model to utilize the northern areas, mallards utilized the southern extent in the Arkansas portion in highest densities. The authors concluded that the presence of mallards was most influenced by proximity to surface water, wetlands, rice fields, fallow fields, waste rice, and open water.

Osborn, J. M., H. M. Hagy, M. D. Mcclanahan, J. B. Davis, and M. J. Gray (2017). Habitat selection and activities of dabbling ducks during non‐breeding periods. Journal of Wildlife Management 81(8):1482–1493.

In this recent study, water depth was used as a potential factor in habitat selection by dabbling ducks on a national wildlife refuge in Tennessee. Density of each taxa was estimated through ESRI ArcMap 10.1 using a combination of aerial imagery, light detection and LiDAR elevation maps with water depth gauge data. Acreage of flooded areas per managed impoundment was calculated using a GIS. This flooded acreage was considered as potential habitat for the ducks. The mean proportion of each species in relation to total duck estimates was calculated and compared to the flooded acres and habitat selection was inferred based on this comparison. While different variables were used as factors in habitat selection by dabbling ducks, the overarching theme of this study is very similar to that of my project on mallards. This project would be a good basis on which to model my methodology. Furthermore, identification of a potential graduate committee member is strongly informed by this particular study.

Pernollet, A. Guelmami. A. Green, B. Curco Masip, G. Dies, F. Bogliani, A. Tesio, M. Brogi, M. Guahtier-Clerc, and M. Guiellemain (2015). A comparison of wintering duck numbers among European rice production areas with contrasting flooding regimes. Biological Conservation 186(C):214–224.

This article reports on a study of duck regional abundance in Europe in relation to availability of rice plantations. Habitat composition of rice-growing areas of Spain, Italy and France were compared over a ten year period (2002-2012). The initial assessment of the study areas required the mapping of different habitat types, and to track changes in land cover over time. To meet this need, the researchers calculated area of each habitat type using ArcGIS software. Man-made and semi-natural wetlands in four of the five areas studied appeared to be stable throughout the ten year period. Using aerial imagery to map out the natural wetlands in the Carmague region of southern France for the years 2001, 2006 and 2010, the researchers used linear regression to show a declining trend of the wetlands in that region.  The methodology includes a statement that the Spatial Analyst Tool of Reclass/Reclassify was used within the ArcGIS software to identify rice field polygons were flooded in winter versus not flooded. Once the rice fields were mapped for a given year, the researchers used the Math Algebra/Raster Calculator to calculate the total flooded area for all of the previously identified rice fields. This second layer was overlaid atop the rice field layer, and the intersection of the shapefiles was calculated. The resulting map represented the flooded area within all ricefields on a given date at a given site, and the data was available within the attribute table. It was helpful to be offered the exact tools used with a GIS for this study, as this supports replicability in science.

While man-made flooding is part of the agricultural practice for growing rice, post-harvest flooding varies across different regions.  In controlled impoundments, such as the refuge in Arkansas, flooding to the appropriate level may be a factor affecting duck numbers in rice fields. The authors conclude that waterfowl sometimes rely on ricefields for forage during the winter, but only if post-harvest practices make food accessible. If rice fields are harvested in entirety, or man-made flood levels surpass the height of available grains and tubers, dabbling ducks may not select for those areas. If instead, rice fields are flooded post-harvest and seeds are accessible, ducks may select for this habitat type.

Pöysa, H. (1986). Foraging niche shifts in multispecies dabbling duck (Anas spp.) feeding groups: harmful and beneficial Interactions between species. Ornis Scandinavica 17(4):333–346.

This paper from 1986 explored whether intraspecific interactions were a factor in habitat selection or avoidance by twelve species of dabbling ducks. The author surmised that body size was a factor in avoidance behavior, in that smaller species avoided larger species. The author attempted to discover foraging niches amongst the species, that is, if one species vacated an area, did another species which had previously avoided the area move in? Additionally, was water depth a factor in the habitat selection, relative not only to body size of the species but to the accessibility of forage plants. For example, dabbling duck with greater reach and larger body size may frequent shallow fields with taller plants, while a smaller duck may prefer smaller seeds or deeper water in flooded fields with taller plants (putting the seeds within reach of the smaller animal). The author reports the results of the study adequately, but it is difficult to follow without a visual representation of the data. If this study were done in a GIS, the author might have used statistical analysis, spatial analysis, presented each species in a different color, and shown different water depths relative to those species on a map. The author might have illustrated niche shifts over time by laying multiple map frames side-by-side. The article does not pre-date the advent of a computerized GIS, but it is understandable to note that in 1986, home computers were uncommon, and even office computers were not as advanced to handle large data files. Applications like ESRI ArcMap, which is not the only GIS program even if prevalent, did not exist nor were they compatible with 1980’s technology.

Robinson, Z. (2009). Research Project Analysis: Bald Knob NWR Waterfowl Habitat Assessment. Unpublished M.S. Thesis, University of Arkansas, Monticello.  

Note: No weblink is available to this unpublished thesis. As my research is centered on Bald Knob National Wildlife Refuge (NWR), the refuge biologist sent me this unpublished thesis of an intern who had served on the refuge in 2009, which I am using in my paper with the permission of Robinson.

Robinson used a GIS to depict cover types within each impoundment/parcel on Bald Knob NWR, calculated the acreage of each and assigned a habitat suitability value of either 0 or 1 for each species of waterfowl. In this manner, the study was able to identify areas of habitat selection for multiple species. The author described methodology for using a GIS to replicate the study, including selecting for particular attributes and finding the strength of relationships (through probability) between species utilizing the same cover types. The methodology described therein will be referenced multiple times in my own research, as this author set a precedent for this work at Bald Knob and I would like to produce a similar high-quality product for the refuge.

Schipper, A. M., K. Koffijberg, M. van Weperen, G. Atsma, A. M. J. Ragas, A. Hendriks, A. Jan, and R. S. E. W. Leuven (2011). Distribution of a threatened migratory bird species in a patchy landscape: a multi-scale analysis. Landscape Ecology26(3), 397–410.

The reason I chose this article is that it includes research on whether nature reserves and refuges are effective in supporting non-game species. Refuges were generally established for wildlife conservation, but are supported by local culture for recreational activities such as sport hunting for game animals such as deer, ducks, and turkeys. My project involves mallard duck, a game species which is not protected nor is of concern for decreasing trends. If managing habitat for a single species of game is prioritized on a refuge, consequences to non-game species should be addressed.  

Set in the Rhine River floodplain in the Netherlands, this study reports on distribution of a threatened bird (corncrake; Crex crex) in relation to floodplain vegetation and different cover types. For this species, hypothesized barriers to movement and distribution include open water and manmade line features such as walls, dikes, and roads, while contiguous stream riparian areas serve as migration routes. Preferring marsh, open areas, and edge, corncrakes may only pass through forest patches in quest of a mate rather than for forage. Singing males were counted and used to estimate densities at each observation point. Coordinates and point counts were input into a GIS which contained maps depicting floodplain sections, nature reserves, ecotopes (small-scale landscape features), and line features for dikes, roads, and streams. The nature reserves in this study were maintained grassland which was mowed prior to the migratory songbird’s seasonal arrival, while the other landscapes were generally rural undeveloped, developed, or agricultural. The nature reserves were included in order to illustrate potential importance to this marsh species. Strong relationships were reported between size of suitable habitat patches and corncrake abundance; bird counts increased with increasing patch size and area of available habitat within nature reserves. These findings suggest that this species responds more to habitat quantity rather than its configuration around a landscape. Nonetheless, fragmentation into smaller patch sizes may selection, and barriers to movement would include large distances between suitable habitat types. Considerations for movement and distribution of non-game species must be given in the establishment and placement of nature reserves and refuges. Using a GIS to map distribution and densities of birds in relation to manmade features, barriers, and managed areas is potentially applicable to comparing layers of different species to see these landscape components are affecting multiple species.

Zhu, X., Mao, X., and S. Zhang (2018). Analysis of regional suitable habitats based on remote sensing and Geographic Information System. Journal Of Advanced Oxidation Technologies21(2).;

NOTE: This article was written originally in Chinese; syntax and grammatical errors in the English translation are inconsequential to the quality of its content. The English translation can be found here.

This research utilized a GIS to analyze spatial and temporal distribution of ducks, herons and gulls (four species total) in the Yangtze River estuary on Chongming Island near Shanghai, China. The authors noted that more than 80% of migrating birds in Australia come from mainland China, and Chongming Island is a significant stopover site as well as wintering habitat for some species. Only 40% of the island is considered natural and suitable as habitat for birds. The study concluded that each bird class utilized the habitat resources differently, in that the distribution of each species varied across time and space in distinct patterns from one another. For example, although beach and estuarine habitat was available at multiple locations, the vegetation, cover, soil, water quality, and proximity to man-made features (such as trenches) were used differently by each bird. The island was divided into polygons based on habitat evaluation of multiple biophysical attributes, and the suitability of each unit was calculated according to the suitability analysis model and the geospatial quantification value for the habitat. This used raster data to map out areas of highest to lowest suitability for each species, as a predictor for their use. I see this as applicable to the mallard study in my research, since the refuge biologist wants to inform the leased agricultural parcels as to which crops are most suitable for mallard use. Such analysis would help determine appropriate hunt and no-hunt zones for the refuge during the duck hunting season.


Understanding relationships between mallards and habitat selection relies on the analysis of temporal and spatial relationships with species distribution. Historically, indexes of habitat suitability were painstakingly compiled for each species, demanding resources of time and labor, such as indexes used by the U.S. Fish and Wildlife Service in species status assessments, conservation planning, and refuge management. With the advent and advancement of geographic information science, using a GIS to compile, analyze, and report on data for species is much more efficient. For focused studies like mine on mallards, maps offer a visual representation of the data which is easily understandable by both the scientific community and the public. The interpretative benefits of using a GIS to convey the importance of conservation to the public, or to provide the justification for refuge management decisions in managing for game species, are complimented by the simplicity of visually representing the data set.  

Creating and posting this annotated bibliography has had several benefits. The first of which is to demonstrate how GIScience and the use of a GIS benefits research, particularly that which pertains to my project. The second is that the literature research unsurfaced additional resources for my project. While reading article for relevance, I explored citations within the papers themselves and found even more literature! Completing this assignment in the graduate-level GEOG560 course has converted me into a fan of annotated bibliographies for future research project. Not only does this force a person to more thoroughly read multiple scientific articles on their topic of choice, it encourages comprehension and retention of the information in the writing of a summary of each study.

Thank you for reading!


Suggested Citation:

Kreisler, S. E. 2019. GEOG560 Annotated Bibliography Assignment. Oregon State University, Corvallis. Web. Updated 1 Dec 2019. Available:

Updated December 1, 2019

Print Friendly, PDF & Email

Leave a comment

Your email address will not be published. Required fields are marked *