Week 9: Final Revisions

Geoda, R, ArcMap

k=1
knn1 = knearneigh(afghan.centers,k,longlat=T)
afghan.knn1 = knn2nb(knn1)
plot(afghan)
plot(afghan.knn1,afghan.centers,col="blue",add=T)
k=2
knn2 = knearneigh(afghan.centers,k,longlat=T)
afghan.knn2 = knn2nb(knn2)
plot(afghan)
plot(afghan.knn2,afghan.centers,col="red",add=T)

k=3
knn3 = knearneigh(afghan.centers,k,longlat=T)
afghan.knn3 = knn2nb(knn3)
plot(afghan)
plot(afghan.knn3,afghan.centers,col="green",add=T)
k=4
knn4 = knearneigh(afghan.centers,k,longlat=T)
afghan.knn4 = knn4nb(knn4)
plot(afghan)
plot(afghan.knn4,afghan.centers,col="darkblue",add=T)
k=5
knn5 = knearneigh(afghan.centers,k,longlat=T)
afghan.knn5 = knn5nb(knn5)
plot(afghan)
plot(afghan.knn5,afghan.centers,col="darkblue",add=T)
d = 100
afghan.dist.100 = dnearneigh(afghan.centers,0,d,longlat=T)
plot(afghan)
plot(afghan.dist.100,afghan.centers,add=T,lwd=2,col="red")
summary(afghan.dist.100)
d = 150
afghan.dist.150 = dnearneigh(afghan.centers,0,d,longlat=T)
plot(afghan)
plot(afghan.dist.150,afghan.centers,add=T,lwd=2,col="2")
summary(afghan.dist.150)
d = 200
afghan.dist.200 = dnearneigh(afghan.centers,0,d,longlat=T)
plot(afghan)
plot(afghan.dist.200,afghan.centers,add=T,lwd=2,col="2")
summary(afghan.dist.200)
d = 50
afghan.dist.50 = dnearneigh(afghan.centers,0,d,longlat=T)
plot(afghan)
plot(afghan.dist.50,afghan.centers,add=T,lwd=2,col="2")
summary(afghan.dist.50)
plot(afghan)
plot(afghan.lags[[2]],afghan.centers, add=T,lwd=3,col="green",lty=2)
afghan.lags = nblag(afghan.knn2,3)
plot(afghan)
plot(afghan.lags[[3]],afghan.centers, add=T,lwd=3,col="green",lty=2)
w.cols = 1:32
w.rows = 1:32
w.mat.knn = nb2mat(afghan.knn1, zero.policy=TRUE)
w.mat.knn
image(w.cols,w.rows,w.mat.knn,col=brewer.pal(3,"BuPu"))
w.mat.dist = nb2mat(afghan.dist.250, zero.policy=TRUE)
image(w.cols,w.rows,w.mat.dist,col=brewer.pal(9,"PuRd"))

Moran I statistic standard deviate = -1.5163, p-value = 0.9353

alternative hypothesis: greater

Week 7 Assignment

GIS and R

GIS and R

GIS and R

Boxplot

Assignment Week 5

Open Gallery

http://ecosystems.wcp.muohio.edu/studentresearch/climatechange02/agriculture/images/LATIN.jpg

http://www.biochange-lab.eu/images/10.jpg

http://images.google.com/imgres?imgurl=http://www.grida.no/_res/site/Image/series/vg-lac/large/10.jpg&imgrefurl=http://www.grida.no/publications/vg/lac/page/2741.aspx&usg=__3Ci_pTOoCRlHiucjdUycUAmYh7c=&h=717&w=595&sz=35&hl=en&start=42&um=1&itbs=1&tbnid=U1PRExk88QVreM:&tbnh=140&tbnw=116&prev=/images%3Fq%3Dclimate%2Bchange%2Blatin%2Bamerica%26start%3D40%26um%3D1%26hl%3Den%26sa%3DN%26rlz%3D1W1RNWO_en%26ndsp%3D20%26tbs%3Disch:1

http://www.mja.com.au/public/issues/179_11_011203/smi10724_fm-1.gif

As we discussed last week, visualization is a key technology and method for analyzing as well as presenting climate change data. Geovisualization is especially important for such issues as climate change because it is used by decision makers as well as the general public, which makes it even more crucial to display the data in an easily-understandable way.
I decided to delve further into the question of climate change after our discussion last week, particularly analyzing geovisualization techniques for my regional focus: Latin America. First of all, although we have seen different examples in our class/readings of various geovisualization techniques, choosing a specific case study about a particular region, made me realize the role of the type of data (i.e. region-based, temporal, spatial, multi-variate, station-based, etc.). As we have learned, the type of data also determines the type of visualization (i.e. 2D maps vs. 3D globes).
In the first map of South America, rather than 2D/3D being a factor, it is the two types of visualization techniques that we should be discussing. One is the dot problem we discussed a couple of weeks ago, which in this case also does not seem to be the best way to convey population/major cities and the colors which depict elevation, which in my opinion is also not the best way to show this information.
The second example using remote sensing and it is a common image used to show climate change in Latin America but it does not have a legend. The third, bar graph showing CH4 emissions in Latin America and the Caribbean is very simple and straightforward. On the other hand, although example 4 looks simple, I do not quite understand it and would like to further discuss it in class.
We should also take into consideration that climate data visualization faces different user groups with different skill levels, purposes qualifications, interests, and from different disciplines as is demonstrated in this class. Another lesson I have also learned from class exercises and further exploring data visualization and its application is that applying visualization to scientific data, like climate change is not straightforward. This is due to the various available tools, techniques as well as parameters. Sophisticated technologies, such as graphical user interfaces, visualization design (as discussed in class in the context of simple v. more complex) are essential for bridging the gap between such systems and users. At best, they reduce the hurdles for applying the full functionality of advanced, interactive data visualization systems.

Hockey Stick Graph

Review Essay: The Climate Science Debate and Role of Data, Information, and Visualization

The “Quantitative Methods in Social Sciences” reading is a nice introduction and provides a helpful example to explain the role of quantitative methodology in a social science study. Because all social science research involves dimensions, measures should be created to evaluate these dimensions. These are the ideas introduced in the context of conceptualization and operationalization.
In the article, “Spin, Science, and Climate Change” the importance of research methodology specifically quantitative visualization is further emphasized. Through describing policy issues and how the debate on climate change and studies led to confusion about the specifics, the article concludes that the problem lies in the methodology and the way the data has been used by politicians to imply that climate change is not a ‘definite’ occurrence.
The authors explain that because data are vexatious, theory is much more straightforward and simple than constructing a set of data that describes “the temperature of the Earth over time.” In other words, yes, it is evident that temperature should be changing, but the problem lies in how to measure and demonstrate these changes. In “Data Visualizations: The State of Art”, Post, Nielson, and Bonneau describe how the degree to which data visualization is currently being used in research, teaching as well as development. This book not only goes into detail about scientific visualization, but I also learned about the more recent field of information visualization.
Through reading chapters of this book, I realized that through using data visualization along with computer-generated images (which in the context of climate change would be images of areas around the globe that show evidence of melting glaciers, etc) we can gain more insight and knowledge from data and its patterns and relationships. “Spin, Science, and Climate Change” revealed that one of the problems with the ‘climate change debate’ is the lack of making connections and using data visualization to show more patterns. The book also discusses methods which can be applied to the ‘hockey stick debate’. For instance, the use of the “broad bandwidth of the human sensory system” can be utilized in map-reading and interpreting complex climate change processes and simulations involving data sets from quantitative/physical geography. I believe these concepts are especially important in research conducted on climate change in terms of methodology of not only computation science but also policy-making as discussed in “Spin, Science, and Climate Change.” Post, Nielson, and Bonneau describe the interplay between various scientific data application areas and their specific problems solving visualization techniques. Specific topics include visualization algorithms and techniques, volume visualization, Information visualization, multiresolution techniques, and Interactive Data Exploration. One of the main methodologies of the book is surveys.
Another helpful book which helped me understand some of the problems discussed in “Spin, Science, and Climate Change” is “Data Visualization 2001” which includes papers (presented at a conference) about visualization of geoscience data, multi-resolution and adaptive techniques, unstructured data, which from my understanding of some of the key issues discussed in “Spin, Science, and Climate Change” is one of the main problems. From the visual diagrams demonstrated in “Spin, Science, and Climate Change” the discrepancies relate to quantifiable factors (i.e. greenhouse gases in the atmosphere- first diagram). Furthermore, the article reveals that most of the disagreements about climate change begin from discussions of quantifiable levels of warming (specifically relating to the rise of carbon dioxide). The authors also describe how other factors (besides carbon dioxide) such as clouds (in the context of water vapour) are not well understood. As result, there should be more attention placed on concepts that people are not as aware of through data visualization, which makes it easier to follow (i.e. patterns of clouds and water vapor forming and how it effects climate change). Although Chart 2 does show how new satellites can now track water vapour in the atmosphere far better than before, it does not contribute to the general understanding of water vapour in the atmosphere.
I personally think it would be more effective to show the pattern of warming
down and cooling up and the role of the greenhouse effects in this process. Although different research still shows different amounts and rates of warming in the lower atmosphere, there should be one overarching graph that shows and makes evident that warming is occurring.
The authors of “Spin, Science, and Climate Change,” explain when and how detailed computer models of the climate need to be called into play. They specifically discuss models (of the atmosphere and oceans) in the context of three-dimensional cells. These models are complex and the authors explain how they are oversimplified. Instead of layering the air, temperature, pressure, etc all within one cell and updating it based on what is happening in adjacent cells and the greenhousing, its properties, and contents, would it be possible to separate them? Despite their limitations, “Spin, Science, and Climate Change” concludes that “climate models do capture various aspects of the real world’s climate: seasons, trade winds, monsoons and the like.”
In “Computer Visualization: Graphic Techniques,” Gallagher et al describe how rapid advances in 3-D scientific visualization have effected the display of behavior. The book shows the role of 3-D in research. The introduction provides a basic overview of the essentials of computer graphics giving reviews of the most recent 3-D graphics display and visualization techniques. Even more thorough about the role of 3-D is “Multi-Perspective Modelling, Rendering and Imaging.” Yu et al describe how “a perspective image represents the spatial relationships of objects in a scene as they appear from a single viewpoint.” The techniques discussed in this article could be applicable to some of the visualization 3-D problems with technique discussed in “Spin, Science, and Climate Change.” In describes how multi-perspective images are combined in terms of how several viewpoints are combined into a single image. Despite their incongruity of view and oversimplification as previously discussed, according to Yu “effective multi-perspective images are able to preserve spatial coherence and can depict, within a single context, details of a scene that are simultaneously inaccessible from a single view, yet easily interpretable by a viewer.”
I would like to now discuss my personal experience with visualization techniques in relation to climate change. I am working on a project called Glacier Research Imaging Project (GRIP) which documents the loss of glacial mass in the frozen “water towers of Asia” through imaging techniques. The GRIP team will deploy resources including archival photographs taken during the past century by the world’s greatest mountain photographers. I can specifically relate to some of the techniques discussed in the “Computer Visualization: Graphic Techniques” article such as how multi-perspective images can be used for analyzing structure revealed through motion and generating panoramic images (see link below) with a wide field-of-view using mirrors through using computer vision.
These images have never been published or publicly displayed. GRIP captures new images that precisely match the early photographic records. Comparing the matched pairs of photographs from each location will starkly reveal the catastrophic glacier loss during the intervening years. GRIP will also follow the downstream journey of the glacial meltwater—from some of the most remote locations in the world to the most heavily populated, from the rural villages of subsistence farmers to sprawling cities bursting with industry, technology, and development. Along the way, the team will document how this essential resource is used and who uses it, in order to better understand the implications of its eventual disappearance.
As briefly mentioned in “Spin, Science, and Climate Change” an immediate threat is the increased incidence of glacial lake outburst floods (GLoFs)—catastrophic discharges of water from lakes that form in depressions previously occupied by glacial ice. These ever-expanding bodies of water are held back by natural dams that are structurally weak and unstable. Dozens of glacial lakes are at risk of bursting their banks, potentially leading to floods that would endanger lives, land, and livelihoods, according to the United Kingdom’s Department for International Development. GRIP is using imagery spanning 150 years to help quantify the loss of glacial mass loss in numerous valley sections. Using this resource in combination with space imagery, the team will estimate the annual fractional volume loss in each catchment area and project future summer-melt scenarios, including estimating when total loss of various glaciers might occur. Here are some examples of some of the images:

Karakoram Panos:
KAR1 http://www.hdrlabs.com/gallery/gigapanos/gigapano.html?karakoram_a&CYLINDER
KAR2 http://www.hdrlabs.com/gallery/gigapanos/gigapano.html?karakoram_b&CYLINDER
KAR3 http://www.hdrlabs.com/gallery/gigapanos/gigapano.html?karakoram_e&CYLINDER

Everest Panos:
EVE1 http://www.hdrlabs.com/gallery/gigapanos/gigapano.html?West_Rongbuk_B&CYLINDER

Using Population Density for Country Case Study: Belize

Using the Population Density Map (previous post), I compiled the information and did further research to analyze the patterns. One of the reasons population was denser in places in Central America like Belize was because of increasing accesbility by the United States (due to US-Central America relations) in ways such as tourism, American investments, etc. Tourism (as well as Americans buying second homes) in Central America (specifically Panama and Belize) until about a decade ago in Belize due to their politics. Recently, however, Prime Minister Dean Barrow’s United Democratic Party (UDP) won the general election in Belize on February 7, 2008 (www.udp.org). As a result, a new Minister of Tourism, Manuel Heredia, was appointed. Heredia is a politician with prior marine biological experience in the fishing industry. With a concentration on fisheries, he served as Chairperson of the Fisheries Advisory Board and Vice-President of the Belize Fishermen's Cooperative (www.udp.org). His current focus is “furthering the growth of tourism by expanding economic opportunities for all stakeholders and by building strong partnerships with the private sector” (www.caribbeanpressreleases.com).
The Population Density map also triggered my interest in demographic informaiton (as mentioned, a layer of demographic data would be very useful to that interface), so I did some research on demographic information on Central American countries, like Belize to draw further inferences.
Demographics
Belize is the most sparsely populated nation in Central America with about half of the population living in rural areas and about one fourth of the population living in Belize City (www.governmentofbelize.gov).
According to the Belize 2000 Housing and Population Census, about 34% of the population is of mixed Maya and European descent (Mestizo), 25% are Kriols, 15% Spanish, and about 10.6% are Mayan. Contrary to the popular believe that a large portion of Belize’s population is Garifuna (personal nterview with Irma Ramos, resident of La Democracia and certified tour guide in Belize) and Bacas (La Democracia resident) only about 6.1% are Afro-Amerindian (i.e. Garifuna). The remaining population of Belize consists of European, East Indian, Chinese, Middle Eastern, and North American groups (Belize Central Statistical Office, 2005).
As a result, it is useful to due further research and look at other patterns that correlate the main variables demonstrated in most graphic data (ie population density) to make connections and delve further into your research question.

Population Density: Central and South America

This map depicts population density in Latin and Central America in 1960. It is of particular interest to me for temporal comparison analysis. It would be a very interesting study to compare population density in Latin and Central America fifty years ago to the present.
From my observations, the population density patterns have not drastically changed. As previously mentioned, humans tend to settle near coastal areas. In terms of population density, itself, there are no dramatic shifts except for population density relatively being more crowded in Central America. This is due to its history (US-Central America relations) during the time period. This data could contribute to my research on Central America (patterns such as tourism, investments, immigration/emigration affecting population density). It would also be helpful to compare Central America population density patterns to South America population density patterns and analyze why Central America’s population was so much denser during certain time periods than South America’s.