Spatial assessment of NDVI as an indicator of desertification in Ethiopia using remote sensing and GIS

Abstract

Desertification is a serious environmental and socio-economic problem occurring at global, regional and local scale. According to Article 1(a) of the United Nations Convention to Combat Desertification (UNCCD), define the term “desertification” means “land degradation in arid, semi-arid and dry sub-humid (dry lands) areas resulting from various factors, including climatic variations and human activities”. If a trend of vegetation cover is negative, then it can be a sign of ongoing desertification processes. The Normalize Differences in Vegetation Index (NDVI) derived from the long term (1982-2006) NOAA AVHRR satellite’s sensor and observed rainfall data at climate stations are used and analyzed to monitor desertification processes through NDVI and rainfall trends over time in Ethiopia. Vegetation cover naturally depends on precipitation. The suggested methodology called linear regression has been validated using vegetation trends as a proxy for degradation processes.
Results show that at climate stations level, more than 80% of the stations have strong correlation(r) between NDVI and rainfall. A majority of the stations have no trend of NDVI and rainfall over time. Sixteen percent of the stations have experienced an increase and four percent of the stations a decrease both in NDVI and rainfall. The NDVI trend map, which is derived from 8_km AVHRR GIMMS satellite’s sensors, shows that a majority of the study area has no significant NDVI trend (p<0.05). More than 40% of the area has experienced a positive trend in vegetation cover (“greening”) and less than 6% of the area has experienced a decrease in vegetation. This over all positive trends is not in line with the previous study acclaiming widespread and irreversible degradation in Ethiopia.

Popular Abstract

Desertification is a serious environmental and socio-economic problem occurring at
different scale (global, regional and local). According to Article 1(a) of the United Nations
Convention to Combat Desertification (UNCCD), define the term “desertification” means
“land degradation in arid, semi-arid and dry sub-humid (dry lands) areas resulting from
various factors, including climatic variations and human activities”. The process of
desertification and its consequences are devastating for the environments, socio-economic and
political stability of the countries where they occur. More than one billion people globally,
most of them among the poorest in the world are affected by desertification. These people,
account approximately one quarter of the planet, are vulnerable for different challenges
including soil degradation and vegetation loss, and leading to deterioration of arable land and
eventually to chronic food insecurity.
Africa has a great potential to undertake both large and small scale activities in order to
maintain food security and sustainable development which is directly related to the
environment and socio-economic issues. However, many challenges limited the potential of
this continent such constraints are: climate change and variability, extreme weather event such
as drought & floods, lack of sustainable land use management, land degradation and
desertification. “A further threat to Africa realizing the full potential of its land resource is
desertification.
The total area of Ethiopia at risk of desertification is approximately 71.5% of the country.
“The Ethiopian highland covers 44% the country’s total area and some of 27 million ha
representing approximately 50% of the highlands are already significantly degraded”. If this
trend keeps in this way, then production yield per head will be decrease by 30% in highlands
area in 20 years. Land degradation and desertification are the main challenge of this country
since ever and resulting vulnerable to drought and famine in 1972/73 and 1984/85. The main
causes of land degradation in Ethiopia are: Overgrazing, deforestation, and poor sustainable
land management practices and more than 90% of the population and cultivated lands are
located in highland of Ethiopia. Desertification is the continual degradation of dry land
ecosystems by climate variations and human activities.
Rainfall in Ethiopia is highly erratic and mostly very high intensity and its distribution
varies spatially and temporally. Such variability threat to an agricultural industry that relies
heavily on rain feed agriculture since it has been very vulnerable to phenomena caused by
rainfall extremes such as annual droughts and intra-seasonal dry spells as well as floods
particularly in the lowland areas. Highlands and lowlands represent the topographical
elevation of an area above and below 1500 meter a.s.l respectively.
Indicators are useful to assess and monitor desertification. They can be grouped into the
following: physical (measurement) indicator, biomass (vegetation) indicator and social
(settlement) indicators. This study has been limited to use of the common vegetation index
from remote sensing dataset called Normalized Difference Vegetation Index (NDVI) and
rainfall data that have been collected from 25 climate stations in order to visualize the pattern
with natural vegetation. Therefore, this study contributes towards identifying area under
desertification process using trend of NDVI and how correlate to rainfall distribution in
Ethiopia and the report of the result can be used as a background document for further study.
The main aim of this study is to assess desertification in Ethiopia using remote sensing
and rainfall data during the study period 1982-2006. In order to reach the main aim, the
following specific objectives are studied: Visualize pixel wise significance linear trend
(negative or positive) map of NDVI during the study periods (1982-2006). Estimate the
correlation between NDVI and rainfall at climate station level. Produce a NDVI indicator map
of Desertification in Ethiopia.
The result shows, at climate stations level, more than 80% of the stations have strong
correlation(r) between NDVI and rainfall. Majority of the stations have no trend of NDVI and
rainfall over time. Some stations have experience increase and a few stations decrease both in
NDVI and rainfall. The NDVI trend map which is derived from 8km AVHRR GIMMS
satellite’s sensors shows, majority of the study area has no NDVI trend. More than 40% of the
area experience positive trend in vegetation cover (“greening”) and less than 6 % of the area
experience decrease in vegetation. In both cases, the change in NDVI and rainfall per pixel
(64 km2
) is very small in magnitude over time (25 years). This study needs further studies in
order to understand in detail situation at local scale by estimating trend map of NDVI using
high resolution sensors. In addition to that there is also lack of archive long time series of
observed rainfall dataset in the study area. So that, TRMM (Tropical rainfall measuring
mission) / GPM (Global precipitation measurement) satellite datasets are an option
concerning to acquisition of rainfall data. The TRMM mission has been ended but the archive
data is still available.
Keywords:
Climate Change, Land Degradation, Desertification, Land Cover, NOAA AVHRR, NDVI and
Rainfall.

• author: Negash Seboka, Gezahagn
• supervisor: Lars Eklundh; Martin Brandt
• organization: Dept of Physical Geography and Ecosystem Science
• course: GISM01 20161
• year: 2016
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: NDVI and Rainfall, land degradation, climate change, desertification, NOAA AVHRR, land cover, GIS, Physical Geography and Ecosystem analysis
• publication/series: Master Thesis in Geographical Information Science
• report number: 51
• language: English
• additional info: Co-supervisor: Dr. Martin Brandt, Department of Geosciences and Natural Resource Management, University of Copenhagen, Denmark
• id: 8877776
• date added to LUP: 2016-06-08 10:17:29
• date last changed: 2016-06-09 08:31:21