Assignment 1: Landscape Ecology and GIS
For this assignment, I reviewed the paper Modeling direct and indirect climate change impacts on ecological networks: A case study on breeding habitat of Dutch meadow birds by J. Van Dijk, R. van der Vliet, H. de Jong, M. van Emmichoven, H. van Hardeveld, S. Dekker, and M. Wassen.
This study investigated four species of meadow birds (Black-tailed godwit, Common redshank, Eurasian oystercatcher, and Northern lapwing), and how climate change affected their habitats in the Netherlands. The authors used three metrics to predict future densities: landscape openness, land use, and groundwater levels. They examined scenarios both including and excluding climate change impacts, and they found that in both scenarios there was a significant decrease in habitat. They found, however, that the indirect results of climate change (human actions resulting from climate change as opposed to the natural effects like increase ocean temperature) had a much larger effect than direct results.
This paper was well written and structured. However, one of the major flaws is that the authors predicted to 2200 CE. They do not explain how they felt comfortable predicting this far into the future. As well, the bird data was over 10 years old; using more recent data would have improved the utility of their study.
[Van Dikj, J., van der Vliet, R., de Jong, H., van Emmichoven, M., van Hardeveld, H., Dekker, S., Wassen, M. 2014. Modeling direct and indirect climate change impacts on ecological networks: a case study on breeding habitat of Dutch meadow birds. Landscape Ecology (Published online 6 January 2015): 1-12.]
Another presentation I found particularly interesting was A modelling approach to infer the effects of wind farms on landscape connectivity for bats, by Federica Roscioni, Hugo Rebelo, Danilo Russo, Maria Laura Carranza, Mirko Di Febbraro, and Anna Loy. The authors created a series distribution model, then a connectivity map, and finally added an overlay of the turbines to determine how much risk wind turbines created to a migratory bat species. The team successfully identified risk areas and suggested ways to improve conditions. However, the authors assumed that the bats were following commuting, daily movement patterns, but the species actually follows migratory, annual patterns. As well, the final map was presented as a binary — areas either contained risk or did not. Perhaps it would have been better to use a gradient or multiple classes to represent varying amounts of risk.
[Roscioni, F., Rebelo, H., Russo, D., Carranza, M. L., Di Febbraro, M., & Loy, A. (2014). A modelling approach to infer the effects of wind farms on landscape connectivity for bats. Landscape Ecology, 29(5), 891-903.]
Assignment 2: Health Geography and GIS
I reviewed the article Geographical accessibility to healthcare and malnutrition in Rwanda for this assignment. This paper analyzed the relationship between travel time from houses to health centers and height-for-age z-scores in children under 5 years old in the eastern province of Rwanda. This paper is one of the first of its kind in that it used travel time as opposed to Euclidean distance to measure the distance between households and health centers. Stunting due to malnutrition is a big problem in Rwanda, so this study sought to investigate whether improving access to health centers would help to combat this issue.
The team found that travel time was a statistically significant factor for stunting, but it was not the most significant in a pool of socioeconomic factors. The maps provided a good visualization of the locations of houses and health centers as well as travel times. The largest issue in this paper, however, was their reliance on height-for-age z-score data from the World Health Organization. This data is global, so it does not take into account average heights of Rwandan people specifically. This may have skewed their results. The team also considered this an initial study on the subject, so further research would be useful.
[Aoun, N., Matsuda, H., & Sekiyama, M. 2015. Geographical accessibility to healthcare and malnutrition in Rwanda. Social Science & Medicine. 130, 135-145.]
The paper Mapping violence and policing as an environmental–structural barrier to health service and syringe availability among substance-using women in street-level sex work, by K. Shannon, et al., provided interesting insight into some of the problems facing sex workers in Vancouver. The paper sought to examine the relationship between sex workers and health resources in the city. The paper found that because of risk, sex workers would avoid places that had health resources. This was a very interesting used of GIS to solve health and social problems, but there were several issues with the study. The authors did not take into account johns or the behavior of clients as factors affecting movements of sex workers. As well, the study looked only at cis-women and ignored other populations such as trans* and male sex workers. Finally, the maps produced were extremely poor.
[Shannon, K. et al. 2008. “Mapping violence and policing as an environmental–structural barrier to health service and syringe availability among substance-using women in street-level sex work”. International Journal of Drug Policy 19.2:140-147.]
Assignment 3: Crime Analysis and GIS
For the final assignment I reviewed the paper Geographies of identity theft in the U.S.: Understanding spatial and demographic patterns, 2002-2006, by G. Lane and D. Sui. There had been no previous studies investigating the geographic spread of identity theft in the US, so the authors wanted to see whether regional patterns existed and to examine the effects of demographic and social factors on these geographic patterns. The team produced maps showing that identity theft, although it is not constrained by physical boundaries, demonstrates similar spatial patterns to traditional larceny. They also found correlation between some of the socioeconomic variables, such as Hispanic population, and identity theft. As well, there were temporal shifts, one of which may have resulted from population displacement resulting from Hurricane Katrina.
Despite being well-presented, this paper had several issues including a failure to discuss possible error, illegible maps, and a lack of discussion on the choice of socioeconomic variables. The legends of the maps were unreadable, rendering the maps useless. As well, three out of eight variables had to do with Hispanic people, and there was no discussion as to why these should be so heavily weighted. There were no other variables on race, and the idea that nearly half of the variables were based on being Hispanic was very curious.
[Lane, G., & Sui, D. (2010). Geographies of identity theft in the U.S.: Understanding spatial and demographic patterns, 2002–2006. GeoJournal, 75(1), 43-55.]
The paper Predictive crime mapping by J. Fitterer, T.A. Nelson, and F. Nathoo interested me because it provided a more broad perspective on the use of GIS in crime, but concentrated on the city in which I currently reside. The authors used a case study of events of residential and commercial breaking and entering (B & E) in Vancouver, BC, to inform a larger discussion of the use of GIS in predictive crime mapping. The study focused on space-time clustering analysis as a way to predict where these crimes would happen, so that a mobile GIS could be developed. The were many limitations due to the nature of this type of data, but the paper is a good initial study of this topic. Unfortunately, the authors did not thoroughly explain why they chose specific modelling and statistical techniques; this would have enhanced their discussion.
[Fitterer, J., Nelson, T. A., Nathoo, F. 2015. Predictive crime mapping. Police Practice and Research 16 (2): 121-135.]