Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Accessibility, poverty and land cover in Hambantota district, Sri Lanka : incorporating local knowledge into a GIS based accessibility model : a minor field study

Ahlström, Anders (2008) In Lunds universitets Naturgeografiska institution - Seminarieuppsatser
Dept of Physical Geography and Ecosystem Science
Abstract (Swedish)
Tillgänglighet har funnits starkt relaterat till socio-ekonomiska indikatorer och
markanvändning i många studier. Termen tillgänglighet har många definitioner. I
grunden beskriver termen möjligheterna eller med vilken lätthet man kan nå,
förskaffa eller delta i en plats, vara eller en aktivitet beroende på spatial eller social
separation. Denna studien har fokuserat på den spatiala separationen. I denna studien
har kontinuerliga data av uppskattad tillgänglighet framställts genom GIS baserad
modellering. Modellen är baserad på en kvalitativ och kvantitativ intervjustudie och
geografiska data som erhölls i området för fältstudien, Sri Lanka. Det var avgörande
att uppskatta reshastigheter av olika landskapsentiteter, friktionerna för... (More)
Tillgänglighet har funnits starkt relaterat till socio-ekonomiska indikatorer och
markanvändning i många studier. Termen tillgänglighet har många definitioner. I
grunden beskriver termen möjligheterna eller med vilken lätthet man kan nå,
förskaffa eller delta i en plats, vara eller en aktivitet beroende på spatial eller social
separation. Denna studien har fokuserat på den spatiala separationen. I denna studien
har kontinuerliga data av uppskattad tillgänglighet framställts genom GIS baserad
modellering. Modellen är baserad på en kvalitativ och kvantitativ intervjustudie och
geografiska data som erhölls i området för fältstudien, Sri Lanka. Det var avgörande
att uppskatta reshastigheter av olika landskapsentiteter, friktionerna för kostnadsytan
6
som användes i tillgänglighets-modelleringen. I andra liknande studier används
vanligtvis en av två vanliga metoder för att uppskatta de kritiska friktionerna. Den ena
är att använda nyckelinformatörer, den andra är att utgå från hastighetsbegränsningar
på vägar. Den andra går bara att använda där trafikflödets medelhastighet är lik
hastighetsbegränsningen på vägarna i fråga, och den första är osäker på grund av
osäkerheterna i svaren från nyckelinformatörerna. I denna studien har en fungerande
metod utvecklats för att integrera lokal kunskap med fysikaliska geografiska data i ett
GIS. Data och metoder som vanligtvis används inom sammhällsvetenskap och
humaniora har integrerats med data och metoder som vanligtvis används inom
naturvetenskap. Detta har uppnåtts genom att att beräkna de dörr till dörr hastigheter
av tre vägklasser som bäst stämmer överäns med intervjudata. Resultatet av dessa
beräkningar har använts som friktioner i kostnadsytorna, som använts i
tillgänglighetsmodelleringen. Resultatet av modelleringen har sedan jämförts med
marktäckning och fattigdomsindikatorer. Jämförelseanalysen visade på starka
förhållanden mellan marktäckningsdata och uppskattad tillgänglighet, samt mellan
fattigdomsindikatorer och uppskattad tillgänglighet. Resultaten tyder på att ett fåtal
människor som bor nära städer överger den tradionella risodlingen. Och de tyder på
att människor med stor uppskattad tillgänglighet till marknadsplatser är fattigare än
människor med låg tillgänglighet till marknadsplatser eller stor tillgänglighet till
städer. Trots osäkerheter på grund av lågupplöst statistisk data tyder resultaten på att
stor tillgänglighet till marknader inte kan minska fattigdomen. De modellerade
tillgänglighetsmåtten visar även starkare relationer till marktäckningdata och
fattigdomsindikatorer än vad det vanliga, enkla, tillgänglighetsmåttet Euklidiskt
avstånd gör. (Less)
Abstract
Accessibility has been found closely linked to socio-economic indicators and to land
use in many studies. The term has many definitions, basically the term describes the
ability or the ease in which a location, opportunity or ware can be reached, obtained
or be participated in by either spatially separated or socially separated people. This
study has focused on the spatial separation. In this study accessibility have been
modelled in a raster based approach in a GIS, the results are continuous estimated
accessibility raster layers. The study is based on a qualitative and quantitative
interview survey and geographical data attained in the country of the field work, Sri
Lanka. It was crucial to estimate the travelling speed of... (More)
Accessibility has been found closely linked to socio-economic indicators and to land
use in many studies. The term has many definitions, basically the term describes the
ability or the ease in which a location, opportunity or ware can be reached, obtained
or be participated in by either spatially separated or socially separated people. This
study has focused on the spatial separation. In this study accessibility have been
modelled in a raster based approach in a GIS, the results are continuous estimated
accessibility raster layers. The study is based on a qualitative and quantitative
interview survey and geographical data attained in the country of the field work, Sri
Lanka. It was crucial to estimate the travelling speed of different landscape entities,
the frictions of the cost surface used in the accessibility model. Similar studies
normally use one of two common approaches to attain the crucial frictions. The first
is to use data from key informants, the second is to use the speed limits of the roads.
The second approach is only feasible where the mean speed of the traffic is similar to
the speed limits of the roads, and the first is vulnerable due to the uncertainties of the
key informant data. In this study a successful method have been developed to
integrate local knowledge and physical geographical data in a GIS-model. Data and
methods commonly used in social sciences and humanities have been integrated with
data and methods from natural sciences. This has been achieved by calculating the
best fit door-to-door travelling speeds of three road classes from interview data. The
results from these calculations have been used as frictions for the cost surface. The
results of the modelling have then been compared to land cover and poverty indicator
data. The comparative analysis showed strong relationships between land cover data
and estimated accessibility and between poverty indicator data and estimated
accessibility. The results suggest that people to a small amount are abandoning the
traditional paddy farming close to towns. And it suggests that people with high
estimated accessibility to markets are poorer than people with high accessibility to
towns. However uncertain due to low resolution statistical data, the results suggest
that high accessibility to markets can not pull the poor out of poverty. It also suggests
that people living in close proximity to towns are better of than people living in close
proximity to the large wholesale markets. The estimated, modelled accessibility
measures also shows stronger relationships to land cover and poverty indicator data
than the commonly used simple accessibility measure of Euclidian distance. (Less)
Please use this url to cite or link to this publication:
author
Ahlström, Anders
supervisor
organization
year
type
H1 - Master's Degree (One Year)
subject
keywords
accessibility, Sri Lanka, Hambantota district, GIS, physical geography, geography, land cover, poverty
publication/series
Lunds universitets Naturgeografiska institution - Seminarieuppsatser
report number
144
funder
SIDA, Minor Field Study programme (MFS)
funder
SAREC - Swedish Agency for Research Cooperation with Developing Countries
language
English
additional info
The field work of this study were funded by SIDA as a MFS scholarship and to a
smaller extent by the SIDA/SAREC funded research project “Regional Development
in an Open Economy – a comparative study of Hambantota and Gampaha districts,
Sri Lanka”.
id
1895471
date added to LUP
2011-12-14 11:20:14
date last changed
2015-06-12 14:52:06
@misc{1895471,
  abstract     = {{Accessibility has been found closely linked to socio-economic indicators and to land
use in many studies. The term has many definitions, basically the term describes the
ability or the ease in which a location, opportunity or ware can be reached, obtained
or be participated in by either spatially separated or socially separated people. This
study has focused on the spatial separation. In this study accessibility have been
modelled in a raster based approach in a GIS, the results are continuous estimated
accessibility raster layers. The study is based on a qualitative and quantitative
interview survey and geographical data attained in the country of the field work, Sri
Lanka. It was crucial to estimate the travelling speed of different landscape entities,
the frictions of the cost surface used in the accessibility model. Similar studies
normally use one of two common approaches to attain the crucial frictions. The first
is to use data from key informants, the second is to use the speed limits of the roads.
The second approach is only feasible where the mean speed of the traffic is similar to
the speed limits of the roads, and the first is vulnerable due to the uncertainties of the
key informant data. In this study a successful method have been developed to
integrate local knowledge and physical geographical data in a GIS-model. Data and
methods commonly used in social sciences and humanities have been integrated with
data and methods from natural sciences. This has been achieved by calculating the
best fit door-to-door travelling speeds of three road classes from interview data. The
results from these calculations have been used as frictions for the cost surface. The
results of the modelling have then been compared to land cover and poverty indicator
data. The comparative analysis showed strong relationships between land cover data
and estimated accessibility and between poverty indicator data and estimated
accessibility. The results suggest that people to a small amount are abandoning the
traditional paddy farming close to towns. And it suggests that people with high
estimated accessibility to markets are poorer than people with high accessibility to
towns. However uncertain due to low resolution statistical data, the results suggest
that high accessibility to markets can not pull the poor out of poverty. It also suggests
that people living in close proximity to towns are better of than people living in close
proximity to the large wholesale markets. The estimated, modelled accessibility
measures also shows stronger relationships to land cover and poverty indicator data
than the commonly used simple accessibility measure of Euclidian distance.}},
  author       = {{Ahlström, Anders}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Lunds universitets Naturgeografiska institution - Seminarieuppsatser}},
  title        = {{Accessibility, poverty and land cover in Hambantota district, Sri Lanka : incorporating local knowledge into a GIS based accessibility model : a minor field study}},
  year         = {{2008}},
}