Research Journal of Environmental and Earth Sciences 3(4): 393-399, 2011 ISSN: 2041-0492 © Maxwell Scientific Organization, 2011 Received: February 14, 2011 Accepted: March 21, 2011 Published: June 05, 2011
Spatio-Temporal Variability of Rainfall Distribution in the Western Region of Ghana
1C.B. Boye, 1I. Yakubu and 2D.S. Pokperlaar 1Department of Geomatic Engineering, Faculty of Mineral Resources Technology, University of Mines and Technology, P.O. Box 237, Tarkwa, Ghana 3Ghana Meteorological Agency, Accra
Abstract: The Western region of Ghana experiences the highest rainfall. The predominant activity in this region includes agriculture and mining. Due to the good climatic conditions coupled with the concentration of mining companies in the area, people from the various parts of the country migrate to this region. The study was carried out to determine the rainfall distribution pattern over a thirty year period from 1975 to 2005 in the western region of Ghana. Ilwis, ArcGIS and Microsoft excel software were used for the data interpolation and trend of the rainfall pattern. The data used for this included monthly and annual rainfall data for selected districts within the region and topographic map. The results revealed that there is a general rise in recorded rainfall quantities from 1975 through 1985, 1995 to 2005 in all the selected meteorological stations within the study area, except Tarkwa which showed an erratic trend. There are other isolated reductions in rainfall pattern over the period. The rated environmental degradation should to check to improve on the situation within the region.
Key words: Rainfall Distribution Pattern, Western Region of Ghana
INTRODUCTION significant increase in heavy rainfall events has been observed (Anonymous, 2010), including evidence for The effect of Climatic change is gradual but has changes in seasonality and weather extremes pronounced consequences on the environment resulting in (Anonymous, 2010). rising sea levels, extreme rainfall and excessive drought. The interest in rainfall distribution pattern for the The IPCC 1995 report anticipated intensification of the study area stems from the desire to investigate the impacts hydrological cycle which would increase global rainfall of climate change on the hydrological cycle as recorded by 7 to 15% (Ramos, 2001). Therefore an increase in by IPCC 1990 for Ghana and West Africa as a whole. extreme events is assumed which may be destructive to Among various climatic variables, precipitation is mainly natural and human systems. However, while some areas required for applications like natural resource will have increased rainfall other areas will suffer management, agricultural management, mining operation decreases in rainfall. A 10% increase in annual rainfall scheduling, ecosystem modeling, and hydrological along the Guinean coast during the last 30 years has been modeling. Understanding its temporal and spatial observed by the Inter governmental Panel on Climate distribution is also important for undertaking climate Change (IPCC). In West Africa a decline in annual change impact studies on various systems (Anonymous, rainfall has been observed since the end of the 1960s with 2010). Changes in climatic conditions are regional in a decrease of 20 to 40% noted from 1931 to 1960 and nature and considered as continuous geographic fields from 1968 to 1990. In the tropical rain-forest zone, measured at selected points in the study area. Spatial declines in mean annual precipitation of around 4% in interpolation procedure of estimating the value of West Africa, 3% in North Congo and 2% in South Congo properties at unsampled sites within the area covered by for the period 1960 to 1998 have been noted existing observations are usually applied with the (Anonymous, 2010). In other regions, such as Southern rationale that points close together are more likely to have Africa, no long-term trend has been noted. Increased inter similar values than points far apart (Tobler's Law of - annual variability has, however, been observed in the Geography). Both deterministic and statistical methods of post-1970 period, with higher rainfall anomalies and more interpolation have been applied for precipitation in earlier intense and widespread droughts reported studies. The geo-statistical interpolation techniques such (Anonymous, 2010). In different parts of southern Africa as kriging (Marco and Andrea, 1997) have been applied (Angola, Namibia, Mozambique, Malawi and Zambia), a to spatial analysis of precipitation. Previous experience
Corresponding Author: B. Cynthia, Department of Geomatic Engineering, Faculty of Mineral Resources Technology, University of Mines and Technology, P.O. Box 237, Tarkwa, Ghana 393 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
Fig. 1: Map of western region has shown that kriging is preferable to other rainfall variability of monthly and yearly rainfall distribution interpolation methods, at least for monthly rainfall or pattern in the study area over a thirty year period from storm totals (Marco and Andrea, 1997). Among three 1975 to 2005 at ten year intervals in view of the IPCC spatial interpolation methods, deterministic and stochastic expectation. methods used for assessing seasonal rainfall variability in Guinea Savanna Part of Nigeria, ordinary kriging was Study area: The Western Region of Ghana lies between found to be suitable for the study because it allows the latitude 4º00! to 7º00! North, and between longitude sharpest interpolation rainfall data and it was the most 3º07! West and 1°07! East of the Green Wich Meridian representative (Ayanlade and Odekunle, 2009). The (Fig. 1). The region is located in the south-western part of results of a study on different interpolation models Ghana and shares boundaries with the Central, Ashanti, generated in GIS environment to show fine scale and Brong - Ahafo regions. To the West it shares a border precipitation surfaces from precipitation data showed that with the Republic of Cote D'Ivoire. The region has 192 the multivariate extension model of Ordinary kriging that km of tropical beaches on the Atlantic Ocean and a uses elevation as secondary data was the best model tropical climate characterised by moderate temperatures especially for monsoon months (Ashiq et al., 2009). all year round. The Region occupies an area of 238,537 Again, the need to keep track of the changes in km2, which is about 6.6% of the land area of Ghana rainfall pattern is imperative for agricultural purposes, (Egan, 1975). It has an estimated population of 19, since agricultural practices in most developing countries 403,792 (2010 projection) and an annual population like Ghana depend heavily on rainfall. Other impacts on growth rate of 2.6% with 13 administrative districts. It is humanity e.g. water management and flooding is also of the second most densely populated region in the country concern. next after Greater Accra with a population density of In this paper a combination of statistical methods and about 79.3 person’s per-square kilometer and 63% of the Kriging (a regionalise spatial interpolation technique) region is rural (Anonymous, 2005). The region has the were applied to determine the spatial and temporal highest rainfall in Ghana and has lush green hills and
394 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
fertile soils. Some of the large rivers in the region are the
) l 20000
a
m Ankobra River, the Bia River, the Pra River in the east u
m
n
(
n 15000
s
A
e and the Tano River partly forming the western national
l
u
a
l
t 10000 a
o border. There are numerous small and large-scale gold V
T
l
l y 5000
t a
i
f mines (Anonymous, 2005). l
n
a
i
u a 0
The native people of the Western Region are mostly Q R 1970 1975 1980 1985 1990 1995 2000 2005 2010 Akans-speaking with various dialects including Ahanta, Years Nzema, Sefwi, Wassa, Brosa, and Pepesa. Principal religions are Christianity, African Aminism and Islam. Fig. 2: Graph of total annual rainfall in western region The principal economic activities include agriculture
(cash crops and food crops), fishing (commercial and ) 3000 1975 1995
m
m 1985 2005
( subsistence), and mining and manufacturing. The main 2500
l
l
a exportable produce are; cocoa, timber, copra, coffee, f
n i 2000
a
rubber/latex, gold, manganese, and bauxite. R
l
a 1500
u
n
n
A
MATERIALS AND METHODS 1000
f
o
y
t
i 500
t
Materials: The area of study is the Western regions of n a
u 0 Ghana. To determine the rainfall pattern within the Q a ta i o i o i l i o w n h s a s im d u n s k a c a w w x a f si n r n o k a or o s e western region, secondary data consisting of monthly and a w E G e i A k kr B T k B W a A N i i T N lf yearly rainfall values were obtained from the w w a ef ef H Meteorological Service for thirty years duration that is S S from 1975 to 2005 at ten years intervals from selected Fig. 3: Graph of total annual rainfall for the selected districts meteorological stations in the study area. A digital topographic map of the study area was also acquired from the Survey and Mapping Division of the Lands 1200 Commission. Data obtained was processed and analysed 2005 1995 using Microsoft Excel, Arc GIS and Ilwis software. This study was conducted in May, 2010. 1000 1985 m 1975
m
n
i
Methods: The methods employed in the production of s e 800
u
l monthly distribution patterns and the ten year interval a
V
l rainfall distribution variability within western region are l
a
f
n
i discussed in the following subsections. a 600
R
y
l
h
t
Monthly rainfall distribution pattern: Rainfall data n
o
M 400 measured and recorded from selected meteorological f
o
y station in each district in the region and beyond were t
i
t
n acquired from the National Meteorological Service office a
u for the study. Statistical processing was carried out using Q 200 Microsoft Excel and appropriate graphs generated from the scatter plots of each station. The quantity of rainfall for each stations per given month was plotted for each 10 0 Jan Mar May Jul Sep Nov year period. Month of the year
Ten year interval rainfall distribution variability: The Fig. 4: Graph of monthly rainfall at Benso values topographic map of the area was loaded in ArcGIS software environment and the locations and the rainfall RESULTS AND DISCUSSION quantities (z) imported. Spatial correlation graphs were generated together with experimental variograms which The total annual rainfall quantities over the thirty were modelled with a spherical model from the data. The year period of the study was observed to drop slightly at points were finally interpolated using ordinary kriging the end of the first ten years by an average of 0.02% per (the regional interpolation technique which is capable of year but increased steadily by an average of 2.25% per showing the errors inherent in the estimation). year from 1985 to 2005 (Fig. 2, 3).
395 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
2000 450 2005 2005 1995 1800 1995 400
m 1985
m
1985 m
m
n
1600
i 350
n
i 1975
s 1975
e
s
u
e
l
u
1400 a l 300
a
V
l
V
l
l
a
l
f a 1200
n
f
i 250
n
a
i
a
R
R
y
1000 l
y
h 200
l
t
h
n
t
o
n
o 800
M
150
M
f
o
f
o
y
t
y 600 i
t
t
i 100
n
t
a
n
u
a
u 400 Q
Q 50 200 0 Jan Mar May Jul Sep Nov 0 Jan Mar May Jul Sep Nov Month of the year Month of the year Fig. 7: Graph of monthly rainfall at Tarkwa Fig. 5: Graph of monthly rainfall at Axim 1800 1600 2005 1600 2005
m
1400 1995 m 1995
n
i
1400
m 1985 s
e
m 1985
u
l
n
a i 1200
V
s 1200 1975
l
e
l
u
a
l
f
a
n
i
V
a
l
l 1000 1000
R
a
f
y
n
l
i
h
a
t
n R 800
o
y 800
l
M
h
t
f
n
o
o 600
y
t
M
i
t
f 600
n
o
a
u
y
t 400
i
Q
t
n a 400 u
Q 200
200 0 Jan Mar May Jul Sep Nov Month of the year 0 Jan Mar May Jul Sep Nov Month of the year Fig. 8: Graph of monthly rainfall at half Assini Fig. 6: Graph of monthly rainfall at Nkroful except Tarkwa which showed an erratic trend. The central part of the region around Sefwi Wiawso showed the least Figure 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 and 14 are the recorded maximum rainfall values followed by Tarkwa, scatter plots of the Monthly rainfall distribution for the Enchi, Goaso and Sefwi Bekwai. Whiles highest rainfall selected stations. values are recorded in Axim, followed by Half Assini, The following (Fig. 15, 16, 17 and 18) are the maps Benso, Nkroful, Nkwanta and Takoradi. generated from the interpolation by Kriging of the annual The rainfall distribution pattern for the 1975 map total rainfall values for the selected stations at ten-year shows low values in the central part of the western region; intervals. around Sefwi Wiawso and stretches to wassa Amenfi southwards to the middle portion of the region to Aowin- Observations from the scatter plots and Interpolated Suaman. Gradual increase is shown radially in all maps: There is a general rise in recorded rainfall directions. The forest belt in the south - eastern direction quantities from 1975 through 1985, 1995 to 2005 in all of the southern part of the region shown maximum the selected meteorological stations within the study area, rainfall values, this pattern reduces as it approaches the
396 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
900 500 2005 2005 800 m 450 1995 1995 m
m
n
i
m 400
700 1985 s 1975
e
n
i
u
l
s
a 350
e
V
u 1975
l 600
l
l
a
a V 300 f
l
n
l 500 i
a
a
f
n R 250
i
a
y 400 l
R
h
t 200
y
n
l
o
h t 300
M
n
o 150
f
o
M
y
f 200
t
i
o 100
t
n
y
t
a
i
t 100 u 50
n
Q
a
u
Q 0 0 Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Month of the year Month of the year Fig. 9: Graph of monthly rainfall at Takoradi Fig. 12: Graph of monthly rainfall at Sefwi Wiawso
2005 600 900 2005 1995 1995
m m 800 1985 m
m 500
1985 n
i
n
i
s 1975
s 700 e
e 1975 u
l
u
l
a
a 400
V
V
600 l
l
l
l
a
f
a
f
n
i
n
i 500
a
a 300
R
R
y
l
y
l 400
h
t
h
t
n
n o 200
o
300 M
M
f
f
o
o
y
t y 200
i
t
t
i 100
t
n
n
a
a
u
u 100
Q Q 0 0 Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Month of the year Month of the year
Fig. 10: Graph of monthly rainfall at Sefwi Bekwai Fig. 13: Graph of monthly rainfall at Enchi
900 2005 700 2005 m 800 1995
m
1995
n
m
i
1985
s m 600 700 1985 e
n
u
i
l
a 1975
s
e
V
600 u 1975
l
l
500 l
a
a
f
V
n
l
i l 500
a
a
f 400 R
n
i
y
l a 400
h
R
t
n
y
l 300 o
h
t 300
M
n
f
o
o
M
y
200 t 200 f
i
t
o
n
y
a
t
i
u
t 100
n 100 Q
a
u
Q 0 0 Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov Month of the year Month of the year
Fig. 11: Graph of monthly rainfall at Goaso Fig. 14: Graph of monthly rainfall at Nkwanta
397 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
Juabeso Bi a Juabeso Bia
Bibiani/Anwiaso/Bekwai Bibiani/Anwiaso/Bekwai Sefwi Wiawso Sefwi Wiawso
Aowi n-Suaman Wassa Amenfi Aowin-Suaman Wassa Amenfi
Wassa West Wassa West Mpohor W Mpohor W Jomoro Jomoro Nz ema East Nzema East Shama Ahanta Eas Shama Ahanta Ea Ahanta West Ahanta West
Fig. 15: Map of rainfall distribution in Fig. 16: Map of rainfall distribution in the western region for 1975 the western region for 1985
Juabeso Bia Juabeso Bia
Bibiani/Anwiaso/Bekwai Bibiani/Anwiaso/Bekwai
Sefwi Wiawso Sefwi Wiawso
Aowi n-Suaman Aowi n-Suaman Wassa Amenfi Wassa Amenfi
Wassa West Wassa West
Mpohor Wassa East Mpohor Wassa Eas Jomoro Jomoro
Nzema East Nzema East Shama Ahanta East Shama Ahanta East
Ahanta West Ahanta West
Fig. 17: Map of rainfall distribution in Fig. 18: Map of rainfall distribution in the western region for 1995 the western region for 2005 parallel coast east of Ahanta West but increases towards C Low rainfall distribution pattern was observed over the west probably due the nature of the coast line. the thirty year along the stretch of west south west coast perpendicular to the direction of the monsoon CONCLUSION winds from the Atlantic oceans. C The erratic trend of rainfall in Tarkwa area can be C The study shows a gradual increase in the rainfall attributed to high concentration of mining companies values from 1985 to the year 2005 for the study area. and the land use/land cover change over the period.
398 Res. J. Environ. Earth Sci., 3(4): 393-399, 2011
REFERENCES Ayanlade, A. and T.O. Odeyemi, 2009. GIS approach in assessing seasonal rainfall variability in Guinea Anonymous, 2005. Retrieved from: http://www.ghanadis Savanna part of Nigeria. Proceeding of the 7Th FIG tricts.com/region/?r=5, (Accessed on: 25 May, 2005). Regional Conference on Spatial Data Serving People: Anonymous, 2010. IPCC Fourth Assessment Report: Land Governance and the Environment-Building the Climate Change 2007. Retrieved from: www. Capacity, Hanoi, Vietnam, 19-22, October. ipcc.ch/publications_and_data/ar4/.../ch9s9-2.html, Marco, B. and V. Andrea, 1997. On the interpolation of (Accessed on: 10 November, 2010). hydrologic variables: formal equivalence of Ashiq, M.W., Z. Chuanyan, N. Jian and M. Akhtar, 2009. multiquadratic surface fitting and kriging. J. Hydrol., GIS-based high-resolution spatial interpolation of 195: 160-171. precipitation in mountain-plain areas of Upper Ramos, M.C., 2001. Rainfall distribution patterns and Pakistan for regional climate change impact studies. their change over time in the Meditarranean area. Theor. Appl. Climatol., 99(3): 239-253. Theor. Appl. Climatol., 69: 163-170.
399