Lauren Gritzke
Final Project Paper
Impact of Climate Change in Cambodia
Sea Level Rise Estimates
Climate change will cause sea level rise, tidal variation, increased frequency of
tropical cyclones, and increased precipitation.1 Scientists expect that continued growth of
greenhouse gas emissions will lead to 1 to 3 meters of sea level rise in this century and
the unexpected breakup of the Greenland and West Antarctic ice sheets could lead to 5
meters of sea level rise.2
Climate Change Effects in Cambodia
Scientists predict that climate change will profoundly affect Cambodia’s coastal
residents, ecosystems, and economy. All climate change models show that rainfall in
Cambodia will increase, but there is a range in the magnitude of this increased rainfall.
Estimates range from 3% to 35% and this increased rainfall will likely occur during the
wet season. Rainfall is likely to increase more in the lowlands than in the highlands.
Runoff in the Mekong Basin is projected to increase by 21% by 2030. There are also
estimates that by 2030 climate change will raise the wet season level of the Tonle Sap
Lake by 2.3 meters.3
Cambodia’s high level of poverty will make it harder for its population to adapt to
the effects of climate change. Cambodia is considered a Least Developed Country (LDC)
with about 36% of its population living below poverty line and it has a GDP per capita of
less than US$300.4 In addition, many economic activities are concentrated within lowlying and coastal areas, such as agriculture and aquaculture. Cambodian farmers are
already facing troubles related to rising sea levels because the seawater is destroying
crops and salting the soil.5
Project Description
This project geospatially shows the areas that would be affected by up to 6 meters
of inundation due to sea level rise as well as increased precipitation in Cambodia. It
shows the villages, schools and hospitals that would be at risk. It also demonstrates the
land cover in the potentially affected areas. Land cover is especially important, as
agriculture is important for maintaining livelihoods in Cambodia.
1
IPCC, Fourth Assessment Report, 2007.
Dasgupta et al. “The Impact of Sea Level Rise on Developing Countries: A Comparative Analysis”.
February 2007. World Bank. Available from http://wwwwds.worldbank.org/servlet/WDSContentServer/WDSP/IB/2007/02/09/000016406_20070209161430/Rend
ered/PDF/wps4136.pdf
2
“Climate Change Vulnerability and Adaption in Cambodia.” December 2009. Available from
http://www.thefishsite.com/articles/805/climate-change-vulnerability-and-adaption-in-cambodia
4 Kingdom of Cambodia, “Cambodia and Climate Change.” 2009 Available from
http://www.mlit.go.jp/common/000053789.pdf
5 Tom Price, “Cambodian Farmers Battle Rising Sea Levels.” Available from
http://crs.org/cambodia/cambodian-farmers-battle-rising-sea-levels/
3
Data Sources
Data source 1: Census
1. Name of data set: General Population Census of Cambodia
2. Year represented: 2008
3. Data description: This data consists of a series of map layers and associated
databases with aggregated counts, ratios and rates. It includes provincial, district,
commune, village, and operational layers as well as water bodies and roads.
4. Agency that produced the data set: National Institute of Statistics of Cambodia
5. URL: This data set was acquired from the M drive, but it is originally from
http://www.nis.gov.kh/
6. Format: Vector
Data source #2: Land cover
1. Name of data set: Global Land Cover 2000: South East Asia
2. Year represented: 2000
3. Data description: The sub-regional vegetation map of South East Asia has been
produced from SPOT Vegetation S10 data. It includes forest cover, cultivated
land, and water bodies.
4. Agency that produced the data set: Global Environment Monitoring Unit (GEM)
– European Commission Joint Research Center
5. URL: http://bioval.jrc.ec.europa.eu/products/glc2000/products.php
6. Format: Raster
Data source 3: Elevation
1. Name of data set: STRM (Shuttle Radar Topography Mission)
2. Year represented: 2000
3. Data description: Digital elevation models on a near global scale from 56 degrees
S to 60 degres N. The resolution is three arcsecond or 90 m data.
4. Agency that produced the data set: NGA/NASA
5. URL: http://www2.jpl.nasa.gov/srtm/index.html
6. Format: Raster
Data analysis
First, I selected UTM Zone 48 for my projected coordinate system.
I conducted analysis for the area that would be affected to up to 6 meters of sea level rise.
I downloaded STRM elevation data. I then clipped it so that it was a manageable size. I
used the Reclass tool to reclassify the STRM data to set everything up to 6 meters to be 1,
and everything else to be 0.
I used the Spatial Analyst tool – Zonal – Tabulate Area tool to calculate the total area of
each type of land cover in the inundation zone (reclassed to be 1). This reports the area
of each type of land cover in the zone in meters. I converted these numbers to square
kilometers in Excel and showed this as a table on my poster. I also clipped the land cover
area by the most affected districts to show the land cover on a map.
I used the Select tools to select out the census districts within the affected zone and used
that selected set to report in table format the population, schools and health centers in the
affected zone.
I assigned elevation values (meters above sea level) to the point data I had, including
villages, schools and health centers. I used the Spatial Analyst Extraction – Extract
Values to Points tool to do this. This allowed me to geospatially show which villages,
schools and health centers were most at risk.
Limitations
I encountered many difficulties throughout the course of this project. I had
trouble finding data and some of the data that I found did not match well enough with the
other datasets. For example, I had originally used ASTER elevation data and had shown
a 3 meter sea level rise area in addition to close proximity to the coastline. When I added
the census data to this raster analysis, much of the inundation zone lay outside the
country boundaries of the census data. Thus, I was not able to ascertain information about
the inhabitants within the inundation zone.
In order to deal with this problem, I decided to increase my affected zone to
include up to 6 meters of sea level rise. While 6 meters sea level rise is not standard
within the literature, it enabled me to better demonstrate my GIS skills, which was the
main purpose of this project. The literature often goes up to 5 meters of sea level rise,
does not factor in increased precipitation and incorporates a significant amount of
uncertainty so this increase to 6 meters is still somewhat relevant. I also ended up
switching to STRM elevation data because it was easier to work with. STRM elevation
data is 90m resolution instead of 30m.
I abandoned incorporating the distance from the coastline aspect of the project.
First, I had tried to construct a coastline from the ASTER elevation data so that the two
data sets would align, by converting the raster elevation data to polygons. I ended up not
using this constructed polygon coastline, however. I simply used the elevation data
because increased flooding will affect all low-lying areas.
There were additional data limitations. Much of the data that I used is not current.
The land cover data is from 2000 and the census data is from 2008. Thus, there may have
been some significant changes since then. It also would have been nice to have
additional data points for other structures (and infrastructure), but I was not able to find
this data. Some of the census point data fell in areas that were designated as below sea
level (negative numbers), which showed that there was a data quality problem.
Concluding Thoughts
In general this approach worked, but the results may not be that useful. By
extending the inundation zone to 6 meters, this project went against accepted practices in
the climate change literature. It seems as though the worst effects from climate change
in Cambodia will not be related to direct sea level rise, as much of Cambodia’s coast is
not at risk (using the accepted 1 to 3 meter range). The worst effects will likely be
related to flooding in low-lying areas. It also would have been better to use the ASTER
elevation data, which is higher resolution.
Annotated Citations
1. Asian Development Bank, “Addressing Climate Change and Migration in Asia and the
Pacific.” March, 2012. Available from
http://reliefweb.int/sites/reliefweb.int/files/resources/addressing-climate-changemigration_0.pdf
This paper was useful because it is recent and discusses the impact of climate change on
countries in Asia specifically. It identified specific provinces that would be more
affected by climate change within Cambodia. These vulnerable areas of Cambodia
include the southeastern border with Viet Nam as well as the coastal provinces of
Kampot, Koh Kong and Sihanoukville. Battambang Province, in northwestern Cambodia
is also vulnerable because livlihoods in this area are primarily dependent on agriculture
and there are high poverty rates.6
2. Dasgupta et al. “The Impact of Sea Level Rise on Developing Countries: A
Comparative Analysis”. February 2007. World Bank. Available from http://wwwwds.worldbank.org/servlet/WDSContentServer/WDSP/IB/2007/02/09/000016406_20070
209161430/Rendered/PDF/wps4136.pdf
This paper was useful because it uses GIS data, discusses climate change vulnerability
and includes Cambodia. This paper chose to use the following critical impact elements:
land, population, agriculture, urban extent, wetlands and GDP. It used projected sea level
rise from 1 to 5 meters. This paper was global in nature (84 countries were included) so
it did not go into a lot of detail on Cambodia specifically.
3. Clark, George E. et al. 1998.Assessing the Vulnerability of Coastal Communities to
Extreme Storms: the Case of Revere, MA, USA. Mitigation and Adaptation
Strategies for Global Change 3: 59—82
This paper was helpful because it helped me to think about what variables I need to think
about when conducting climate change vulnerability analysis. It lists many vulnerability
variables including, age, disabilities, family structure and social networks, housing and
the built environment, income and material resources, lifelines, occupation, race and
ethnicity. The authors used two methods to obtain a scalar index of coping ability –
averaging and using data envelopment analysis. I can learn from this technique when
conducting my raster overlay analysis.
4. Usery, E. Lynn, Jinmu Choi and Michael Finn, “Modeling Sea-Level Rise and Surge
in Low-lying Urban areas using Spatial Data, Geographic Information Systems and
Animation Methods” Center of Excellence for Geospatial Information Science (CEGIS)
Available from http://cegis.usgs.gov/pdf/sea_level_rise_text.pdf
The authors incorporate elevation, land cover and population into their sea level rise
model. This source was useful because it includes very specific GIS steps on conducting
climate change vulnerability analysis. For example, it explained that a population value
http://reliefweb.int/sites/reliefweb.int/files/resources/addressing-climatechange-migration_0.pdf
6
is assigned to an output pixel only when the location is at an elevation equal to, or lower
than, the specified elevation in the condition. They made separate maps for 1 to 5 meters;
this helped me to think about how I will present my variables.