Effect of seasonal drawdown variations on groundwater quality in Nigeria

Water samples from twenty (20) shallow wells in Akure were analyzed during the wet and dry seasons in 2009 to ascertain the effect of drawdown on their qualities. Twenty (20) parameters consisting of five physical, twelve chemical and three heavy metals were tested for in the samples. The parameters included dissolved and suspended solids, turbidity, the pH, alkalinity, calcium, sulphate, nitrate, magnesium, electrical conductivity, lead, iron and manganese, were determined using standard procedures. Preliminary findings showed that 40% of the wells had poor drainage system and water levels expectedly varied significantly with seasonal change. However, most of the analysis showed significantly negative and weak correlations for the observed parameters during the two seasons of study. The revealed water quality was independent on drawdown but dependent on other parameters such as hygiene, pollution due to usage, underlying rock formation materials and proximity to polluting sources peculiar to emerging African cities.


INTRODUCTION
Lack of clean water and fresh water pollution are among many plausible causes of potable water shortage and outbreak of water related infections especially in the developing nations.Increase in pollution and its necessities have led to deterioration in both surface and sub-surface water (Shyamala et al., 2008).In Nigeria for instance, Jaji et al. (2007) reported that only 48% of the inhabitants of the urban and semi urban areas and 39% of rural areas inhabitants have access to potable water supply.This confirms the World Health Organization (WHO) report which stated that one-fifth of the urban dwellers in developing countries and three quarters of their rural dwelling population do not have access to reasonably safe water supplies (WHO, 2004).This has placed tremendous pressure on groundwater supplies as alternative to the shortage being experienced, thereby resulting in rapidly-changing drawdown in most wells *Corresponding author.E-mail: sanusko@unisa.com.Tel: +27 11 471 2829.(Howard and Bartram, 2003).
Moreover, due to this pressure, increased usage has not only put pressure on the rate of drawdown but also responsible for pollution (Ikem et al., 2002).Groundwater pollution is believed to be mainly due to the process of industrialization and urbanization that has progressively developed over time without any regard for environmental consequences (Longe and Balogun, 2010).However, recent studies have shown that poor drainage especially around wells, over-exploitation, high hydraulic conductivity of soil and indiscriminate use of herbicides and inorganic fertilizers for subsistence agricultural practices also played prominent role in its pollution (Kannan, 2008).Igbinosa and Okoh (2009) reported the damaging consequences of leachate infiltration into groundwater bodies on life expectancy of such water consumers, while Quinn et al. (2006) enumerated its effect and that of delayed drawdown on moist plant productivity and wetland ecology.
Several studies have been conducted by Al-Awad (2004), Paredes and Lund (2006), Quinn et al. (2006), Kannan (2008) and Al- Sabahi et al. (2009) on excessive pressure rate of drawdown on groundwater qualities and quantities, aquatic biota, recharge-discharge relationships and health implications with little or no consideration on the probable effect of seasonal drawdown variation and its associated factors on water quality.The objective of the study therefore was to access probable effects of seasonal drawdown variation on groundwater quality in Akure, Nigeria.The study would also consider other associated factors that may be responsible for groundwater contamination within the region of interest.

Study area
The study area was Akure, Ondo State, Nigeria.Akure is located on latitude 7° 15 ' 0" N and longitude 5° 12 ' 0" E, Southwestern part of Nigeria.It has a tropical humid climate with two distinct seasons: a relatively dry season from November to March and rainy season from April to October, though now modified and unpredictable due to the effects of climate change.Akure has an average annual rainfall within the range of 1405 and 2400 mm of which rainy season accounts for over 90% and the month of April marks the beginning of rainfall.

Experimental procedure
Twenty (20) shallow wells in Akure were randomly selected for use in this study which indicated two well in one zone.Their depths and diameters were measured using measuring rule and line sinker.Physical inspection of wells was carried out to gather information such as well development and completion.Type of completion whether ringed or not, age of well, distance from polluting source example abattoir, landfill, etc and information on diurnal fluctuation and differential local discharges were collected.Oral interviews and structured questionnaires were employed in collecting these and other information about the wells.A zonal classification based on district location was done for easy data collation and reporting.The description and associated information on the selected wells are shown in Table 1.The selected wells were measured in the two distinct seasons in Nigeria; dry and wet.The first round measurements were conducted in February 2009, while second round measurements took place in August, 2009.

Sampling and analysis techniques
Water sample collection was done using standard techniques described by AOAC (2000).The physical analyses conducted included, temperature, turbidity and dissolved solids (DS).For chemical analysis, the following parameters were analyzed: pH, electrical conductivity (EC), chloride (Cl), sulphate (SO 4 ), nitrate (NO 3 ), Iron (Fe), suspended solids (SS), total solids (TS), potassium (K), total alkalinity (TA), total hardness (TH), sodium (Na), calcium hardness (CH) and magnesium hardness (MH).Manganese (Mn) and lead (Pb) were also investigated for heavy metals and free carbon dioxide (CO 2 ) was also determined in all the samples considered using standard methods for the examination of water (APHA, 2005;AOAC, 1990).The pH was determined using a Mettler Toledo pH meter by direct measurement, analog mercury thermometer was used in making temperature measurements and a Hach 2100A turbidimeter was used for turbidity determination.
To ensure a high degree of accuracy, samples collection was done in the morning before the wells were put into use and laboratory sampling bottles were used for sampling.The samples were covered with cork to prevent spillage and contamination and were kept in the laboratory at 4°C before the commencement of analysis.The wells under construction or development were not considered, two well-spaced (20 m distance) wells were chosen in each of the 10 selected zones and special attention was given to two abattoirs in Akure during the investigation.

Statistical analysis
Paired sample t-tests were used to analyze significant differences between the observed parameters during different seasons and also between the drawdown rates observed.The probability levels used for statistical significance were P < 0.05 for the tests.

RESULTS AND DISCUSSION
From the preliminary observations and structured questionnaire, the following were observed (Table 1); Forty percent (40%) of the zones where water samples were collected had poor drainage system.Seventy percent (70%) of the wells whose water samples were analyzed were being used for domestic purposes.Eighty percent (80%) of the wells were protected with iron lids; ten percent (10%) were covered with wooden cover, while the other ten percent (10%) had no cover at all.Eighty-five percent (85%) of the wells had ring-shaped concrete as support and the number of the rings depended on the depth of the well, while fifteen percent (15%) had no casing.
The mean temperature value for all locations during the dry season (Table 2) was 28.4°C, with the highest temperature value of 30°C recorded in samples from FUTA 1 and FANIBI 1 locations, respectively.The mean temperature during wet season (Table 3) was 22.87°C, while the highest value was recorded in sample B1 (FUTA 2).The mean temperature during the dry season was higher due to the prevalent atmospheric conditions.Higher number of sunshine hours would naturally implied lower relative humidity, temperature increase of water bodies due to conduction and convection processes by the earth crust.These were, however, different from the World Health Organization (WHO, 2004) values of 32°C high average value and 27°C ambient temperature under the same conditions.On the other hand, this was similar to the findings reported by Jaji et al. (2007) in his study.Pollutants, among many other factors may cause temperature increase in water in the present situation and similar result was also recorded by Al-Sabahi et al. (2009).
Furthermore, the well depths remained the same in the two seasons but water levels in the well fluctuated due to variation in weather during the seasons.In all the wells sampled, less than 5% had water depth above 2 m, indicating that the wells were mostly shallow wells whose yield was small.The average depth of the wells ranged between 4 and 12 m (Figure 1).The drawdown in water levels during the two seasons was also shown in Figure 1.There were pronounced reduction in water levels during the dry season and obvious increase in water level during the wet season in all the wells; it ranged from 0.2 to 1.8 m between the two water levels in the two seasons considered (Table 4).The changes in water levels may be due to the yield of the aquifer, location of a deep well close to the vicinity, depth and diameter of the wells.This was corroborated by Mohammed et al. (2009).Smaller diameters and deeper wells tend to have higher change in water level than the current observations.
More also, turbidity values ranged from 1.1 to 10 NTU in dry season (Table 2) samples and from 1.2 to 18 NTU in wet season samples (Table 3).Some were higher than 5 NTU, the WHO stipulated values particularly in areas such as H1 (Sijuwade 2), L1, (Oke-ogba 2) and M1 (Obanla 1) during the dry season.In wet season, only M2 (Obanla 1) had values higher that 5 NTU.More locations had higher turbidity values during dry season due to higher demand for water as a result of its shortage.More users had contact with the wells using locally-made fetchers to draw water.This in turn detach the soil    particles from the unstable sidewalls of the well, mostly unlined which increased sediment deposit and turbidity values.These were lower than the values of Longe and Balogun (2002).When the turbidity values for both dry and wet seasons were subjected to statistical test (Figure 2), same pattern was observed in their behavior and significant (r = 0.69) at 95% confidence level (Table 5).However, when the well drawdown was compared with the turbidity values during two seasons, there was no significance with r-values 0.01 and 0.07 respectively (Table 6).The weak correlation showed that the well drawdown did not have any effect on the change in turbidity values observed.Total hardness (TH) values ranged from 73 mg/L in K1 (Oke-ogba 1) to 420 mg/L in samples Q1 (Araromi1) (abattoir) during the dry season (Table 2) and from 80 mg/L in F2 (Ijoka 2) to 420 mg/L at P2 (Oshukoti 2) in wet season (Table 3), respectively.The WHO value of 300 mg/L is the maximum permissible standard for potable water, hence samples M1, N1, O1, P1, Q1, S1 and N2, Q2, P2 and S2 both in dry and wet seasons exceeded the standard and were adjudged as hard.Samples from Obanla 1 and 2, Oshokoti 1 and 2, Araromi 1 and Okearo 1 were hard having exceeded the permissible limit.Underlying rock formations, presence of calcium and magnesium carbonates may be present in the identified samples.The WHO (2004) reported that encrustation in water supply structure and an adverse effect on domestic use was probable in wells above its recommended value.The values were however higher than those obtained by Shyamala et al. (2008).Higher values of TH were observed in wet season when compared with the values in dry season which expectedly was due to higher volume of precipitation, infiltration and runoff which would subsequently have found its way into the aquifer thereby recharging the wells (Figure 3).Analyzing the comparisons, a strong significance (r= 0.91) existed between the values but weak correlation (P< 0.05) existed when the hardness values were compared with well drawdown in all the locations with r-values 0.10 and 0.21, respectively (Table 6).
Meanwhile, the electrical conductivity (EC) values ranged from 180 to 720 µmho/cm in dry season (Table 2) and 180 to 810 µmho/cm in wet season (Table 3) and all were still below 1000 µmho/cm recommended by the WHO for potable water.The EC values were, however, lower than the findings of Al Sabahi et al. (2009) since his groundwater samples were located near landfill and leachate infiltration would be high and hence the very high values.The EC values were surprisingly higher during the wet season when compared with the values obtained during the dry season (Figure 4) and the correlation relationships during the two season (r =0.69) showing relatively high variation.Traditionally, EC values should be lower in wet season due to dilution by increased volume of water.This reason for this observation is unknown; hence further study is suggested on it.Moreover, there was no significance (P< 0.05) when drawdown values were compared with EC values during dry and wet seasons, -0.15 and -0.18, respectively (Table 6), indicating      that the drawdown had no influence on the quality of water samples.
Figure 5 shows the comparisons of well samples with calcium hardness values in dry and wet seasons.As for chloride ion, all the samples were below the WHO limit of 250 mg/L, however, values observed from S1 and S2 (Oke aro 1 and Oke aro 2) had high values 119 and 118 ppm during the dry and wet season, respectively (Figure 6).The values were much lower than the findings of Ikem et al. (2002) and Al-Sabahi et al. (2009), but similar and agreed with the findings of Longe and Balogun (2002).There was significant difference (P< 0.05) between the two pairs with correlation value r = 0.85 (Table 5) and insignificant difference between the chlorides values in dry and wet seasons when compared with well water drawdown (r = 0.26 and 0.22) (Table 6) at P < 0.05.Other parameters analyzed such as sulphate, potassium, magnesium, nitrate and sodium were within the WHO (2003) permissible level for potable water, hence do not pose health risk to the consumers.Heavy metals such as lead, manganese and iron had very low concentration and some were not even detected in all the samples analyzed (Tables 3 and 4).
The bacteriological assay conducted also did not show presence of any pathogenic organisms and/or total bacteria count, which inferred that pollution due to feces either from humans or animals were less likely to occur.This was probably due to the fact that all but one well was covered and none was located near a stream or river which may be carrier of any bacteria organisms (Table 1).The analyses between each sample within the two seasons were significant (P < 0.05), while their correlation with the drawdown in all the wells were not significant at the same confidence level (Table 5).Some were negatively-weak correlated, showing that there was no relationship whatsoever between the samples with respect to drawdown of water level experienced during the dry season in all the wells (Table 6); Al-Awad ( 2004) and Paredes and Lund (2006) findings in their studies pointed to this fact.
A major factor that may be responsible for pollution could be lack of functional drainage systems around the well.Forty percent (40%) did not have drainage and this would result in polluted water finding its way back into the well by infiltration.Though, the concentration of the contaminants may have been reduced by the infiltration process, this however does not prevent pollutants from getting into the well thereby compromising the quality.The mode of wells usage could also be responsible for the pollution observed in the water samples.Since the wells were cited among the communities, indiscriminate domestic usage and discharging used water near the wells in an unhygienic manner could also be responsible for the contamination.Also, location of pit latrines, septic tanks and other polluting sources such as landfills and abattoirs within the communities were most probable sources of pollution of the surrounding wells.Leachate from these pollution sources would easily have found their way into these wells unnoticed.Sangodoyin (1991) therefore suggested the location of domestic wells to be 30 m radially away from any polluting source.The pollution from the underlying rock formation and agricultural pesticides used for subsistence farming could also be responsible for the change in water quality in the area.

Conclusion
An attempt was made to establish the effect of well water drawdown variations on the quality of water in Akure and its environs, and 20 different parameters were analyzed from samples of 20 wells chosen for the study.Preliminary findings indicated that most of the inhabitants within the well locations on it for domestic purposes despite the unhygienic state of some of the wells, such as lack of cover and casing.All the parameters analyzed showed significant differences at 95% confidence interval when compared between the two seasons considered, but were not significant at the same interval when compared with the water drawdown observed during the dry season.The marginal potability of the water was never in absolute doubt, giving the number that had casings and lid covers since almost all the parameters tested were within the permissible limit by the WHO for portability.However, concern exists when the water quality fluctuates due to diurnal changes in water levels occasioned by change in weather pattern.
This study revealed the independence of drawdown on water quality in the area but on factors such as the hygiene, as less than 40% had poor drainage which will permit inflow of polluted water back into the well, thereby compromising the quality.Other factors could be the mode of usage, proximity to polluting sources such as pit latrines, septic tanks, landfills and abattoirs, which were common occurrence in the city before Government's intervention.This has led to cholera outbreaks and other water-related infections.The pollution from the underlying rock formation and agricultural pesticides used for subsistence farming could also be responsible for the change in water quality in the area.

Figure 1 .
Figure 1.Comparisons of well depths with water depths in dry and wet seasons during the experiment.

Figure 2 .
Figure 2. Comparisons of well samples with turbidity values in dry and wet seasons.

Figure 3 .
Figure 3. Comparisons of well samples with total hardness values in dry and wet seasons.

Figure 4 .
Figure 4. Comparisons of well samples with electrical conductivity values in dry and wet seasons.

Figure 5 .
Figure 5. Comparisons of well samples with calcium hardness values in dry and wet seasons.

Figure 6 .
Figure 6.Comparisons of well samples with chloride values in dry and wet seasons.

Table 1 .
Locations and features of wells form the study area.

Table 2 .
Analysis of results from field samples taken during dry season inFebruary, 2009.

Table 3 .
Analysis of results from field samples taken during wet season in July, 2009.

Table 4 .
Water drawdown between the water levels in dry and wet seasons.

Table 5 .
Correlation analysis of water parameters analyzed in wet and dry seasons with water depth in dry season.

Table 6 .
Correlation analysis of well water drawdown with selected parameters