Infiltration Measurement: Permeability and Available Moisture
The measured permeabilities (Ksat), introduced in the Soil Water Characteristics (SWC) software (Saxton et al. 1986) for the determination of the characteristic soil moisture, allowed to obtain, in combination with HWSD (Nachtergaele et al. 2012), several soil textures for the same permeability but with almost equal available moisture (AM) for the obtained soil textures. The different textures obtained on SWC, were clay-silt texture covering 41% of the valley area (which can also correspond to the clay), clay-silt, clay-sand or loamy-clay-sand texture, with an AM of 12.22 ± 1.9. The silty texture covers 37% of the valley bottom area, and can correspond to the following textures: loamy-sandy-clay, loamy-silty, loamy with an average AM of 3.06 ± 0.02. Finally, the soils of silty-sandy texture cover 22% of the valley bottom area, but also fits the following textures: sandy-loamy and silty-loamy with an average AM of 13.2 ± 4.10.
Variability of the Water Table During the Year
The recharge of the aquifer calculated by the Thornthwaite balance is equal to 68.48 mm, i.e. 7% of the rainfall. It is higher than the values found for the altered aquifers in the Ouagadougou area, ranging from 19.95 to 49.88 mm/year (Hie 2009), which can be explained by a higher value of the runoff coefficient. In fact, in irrigated rice field such as in Boulbi valley bottom, the use of delineation dikes of farm plots decreases runoff and subserve infiltration.
Agricultural Input Inventory
The survey provided an overview of the types of treatments applied by rice farmers in the study area. The most frequent treatments were herbicides, due to the quantity of weeds, and insecticide, due to crop pests. This situation fits with that described generally in Burkina Faso (Toe 2007). The different types of inputs used on the Boulbi valley bottom are NPK (14-23-14) and urea. Organic manure use is very rare on the plain mainly due to its high cost. Twenty (20) types of phytosanitary products are used on the irrigation scheme Table 1, among which 35% are composed with the controversial glyphosate (Lajmanovich et al. 2015; Sarkar and Das 2017). Several studies indicate glyphosate to be a highly toxic and carcinogenic substance. It is also interesting to note that paraquat chloride – denounced to be responsible for magnesium and potassium depletion in soils (Mbuk et al. 2009) – represents 30% of the herbicides used. The World Health Organization recommends that the use of these products be reserved for trained persons who respect the precautions required (Gougo 2016).
Table 1
List of phytosanitary products used on the Boulbi valley bottom
TRADE NAME
|
PESTICIDE TYPE
|
ACTIVE SUBSTANCE
|
ADWURA WURA
|
Herbicide
|
GLYPHOSATE 360 g/l
|
ADOPA WURA
|
Herbicide
|
GLYPHOSATE 360 g/l
|
BIBANA
|
Herbicide
|
GLYPHOSATE 360 g/l
|
GROWNSATE
|
Herbicide
|
GLYPHOSATE 480 g/l
|
GANORSATE
|
Herbicide
|
GLYPHOSATE 480 g/l
|
SUNPHOSPHATE
|
Herbicide
|
GLYPHOSATE 350 g/l
|
ROUNDUP 360 SL
|
Herbicide
|
GLYPHOSATE 360 g/l
|
BENAXONE SUPER
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
GRAMOQUAT SUPER
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
GRAMOSHARP SUPER
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
GRAMODA SUPER
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
GRAMOKING 276 SL
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
PARAKIN 276 SL
|
Herbicide
|
PARAQUAT CHLORIDE 276 g/l
|
EMACOT
|
Insecticide
|
EMAMECTINE BENZOATE
|
KAPAASE
|
Insecticide
|
EMAMECTINE BENZOATE 20 g/l; ABAMECTINE 20 g/l; ACETAMIPRIDE 40 g/l
|
DECIS 25 EC
|
Insecticide
|
DELTAMETHRINE 25 g/l
|
ALLIGATOR 400 EC
|
Herbicide
|
PENDIMETHALINE 400 g/l
|
PYRICAL
|
Insecticide
|
CHLOPYRIPHOS-ETHYL
|
SAMORY
|
Herbicide
|
BENSULFURON METHYL 100 g/kg
|
TOROL
|
Insecticide
|
LAMBDACYHALOTHRINE 16 g/l
|
Analysis of method parameters
In general, the water table is shallow, with static levels varying from 2.22 m to 3.55 m (Table 2), thus assigning high elevation values to the parameter D in all the three methods. Net recharge (R) on the Boulbi plain is low, averaging 68.48 mm/year. The Boulbi rice plain has a free water table. Analysis of the logs of boreholes drilled in the area (Ouandaogo/Yameogo 2008) shows that the lithological nature of the Aquifer (A) is composed of alluvial deposits, essentially clay and granite. The type of soil (S) in place are essentially clay, clay loam and loamy sand. The calculated slopes (topography parameter T) give very low values, ranging from 0.0 to 4.0%. The vadose zone impact (I) class include mainly clay, metamorphic and igneous rocks. The permeability test using double rings has made it possible to determine the hydraulic conductivity (C) of the soils. The plain is essentially an agricultural area (land use parameter O) with irrigated rice fields, irrigated vegetables and plantations (only the SI method is concerned by this parameter).
Table 2
Rating (r) values obtained for each parameter in the 3 vulnerability methods
Rating (r) computed for the three vulnerability methods
|
Parameter
|
Class
|
DRASTIC
|
GOD
|
SI
|
Water table D (m)
Recharge R (mm/an)
Lithology A
|
2,22 − 3,55
68,48
Free water table/fractured granite
|
9
3
3
|
1
-
0,7
|
90
30
30
|
Type of sol S
|
Laterite
Clay
Clay loam
Loamy sand
|
3
1
4
6
|
-
-
-
-
|
-
-
-
-
|
Topo slope T(%)
|
0–2%
2–4%
|
10
9
|
-
-
|
100
90
|
Vadose I
|
Clay,
Metamorphic and igneous rock
|
3
1
|
0,55
0,60
|
-
-
|
Conductivity C (m/s)
|
1,5.10-7–5.10-5
|
1
|
-
|
-
|
5.10-5–15.10-5
|
2
|
-
|
-
|
Land use O
|
rice, irrigated vegetables, plantations
|
-
|
-
|
90
|
Vulnerability Characterization
A class of low vulnerability occupying the entire surface of the plain, with indices oscillating between 83 and 88, shows on the map of the vulnerability to agricultural pollution by the DRASTIC method (Fig. 3) indicating that the plain is more likely well protected. The nature of the dominant soil with a very high clay content is a factor limiting the infiltration of pollutants into the aquifer, hence the relatively low vulnerability class. The figure yielded by the GOD method also shows moderate vulnerability class (Fig. 4): it extends over the entire study area with indices between 0.35 and 0.42. It guarantees a less severe pollution in case of contamination. This degree of medium vulnerability may be related to the nature of the vadose zone, essentially composed of clay and granite, and the dominant soil on the plain, composed of clay, which is not very permeable (Khakhar et al. 2017). Finally, the SI method, gives values that vary between 57.5 and 60 (Fig. 5). The analysis of this map highlights a class of medium vulnerability, which occupies the entire plain. These vulnerability indices guarantee a less severe pollution in case of contamination. The results of SI map were in agreement with the laboratory analysis of NO3− concentration in the groundwater for the entire 11 locations (Fig. 7) for which all the values were found below 50 mg/l, the maximum defined by WHO (Hamza et al. 2007; Stigter et al. 2006; Ward et al. 2018). However, it should be noted that the results of the validation of the vulnerability maps could be more complete if a larger number of nitrate measurements well distributed over the surface of the studied aquifer were carried out, as the data only concerned the water from wells present on the plain.
Determination of the physico-chemical characteristics of the collected water
The parameters measured in the field concerned 32 samples, 21 of which were from surface water (plots of land, earthen drains, and in the dam) and 11 for water from large diameter wells present on the plain.
Hydrogen potential
The surface water in the Boulbi valley bottom, at the level of the plots, the drains and the irrigation dam are clearly basic. The average pH value is 8.2 ± 0.4. The well groundwater on the plain is slightly basic, even close to neutrality with an average pH of 7.3 ± 0.24 (Fig. 6). Though unclear, attempts have been made to explain why a lake or dam irrigation water tends to be basic. The reduction of sulphate in dam water seems to contribute to the formation of carbon dioxide CO2 which reacts with hydroxide anion OH− to form HCO3− and CO32−, thus increasing the water basicity (Borzenko et al. 2018). It should be noted that the pH of the irrigation water should be between 5.5 and 6.5 for better micronutrient absorption and improved photosynthesis and yield of the crops (Guimarães et al. 2021; Liu et al. 2016). This basic pH may partly explain the low rice yield (less than 4.0 tons/ha) observed in Boulbi valley bottom in spite of the use of fertilizers. The pH values measured for Boulbi waters comply with the WHO recommendations (Ward et al. 2018) for drinking water (6.5–8.5). Additionally, it can be noted that the pH values found are slightly higher than the values found by Ouandaogo (2008) in the waters of Ouagadougou. Finally, other authors found values of 6 to 8.5 in surface waters in sub-Saharan Africa (Abai et al. 2014; Chapman and Kimstach 1996), more in agreement with those of Boulbi.
Electrical Conductivity and Nitrates
The average conductivity is higher for groundwater (195.6 ± 42.2) µs/cm than for surface water (156.9 ± 12.4) µs/cm. These conductivity values are too low to be able to cause any damage to rice. The waters (ground and surface water) are moderately mineralized as they are in the range of (150 to 300) µs/cm (Ferreira et al. 2019). It is known that the conductivity values of surface waters change according to the geological structure and the amount of precipitation (Tepe and Boyd 2003). High EC values indicate the presence of a high concentration of dissolved salts in the water and also correspond to local or point source groundwater pollution during rainy periods.
Nitrate testing in water is a good indicator of raw water quality, and in the long-term nitrate contamination can generally lead to eutrophication of surface waters due to excessive nutrient and biodegradable matter inputs. Nitrate values vary significantly from surface water (4.29 ± 2.25) mg/l to well water (15.61 ± 10.36) mg/l. The values obtained show higher nitrate concentrations in well water which are mainly due to irrigation water running into not-coping protected wells, fish farming, pastoral activities and runoff around and in the Boulbi valley bottom (Fig. 7). These nitrate values are lower than WHO guidelines for surface waters (50 m g/l). Studies conducted by Some (2008) and Tapsoba (2016) also report concentrations lower than the WHO standards for surface water in the Nakambé basin in Burkina Faso, and in dams N°3 of Ouagadougou and DEBE respectively of 34 m g/l; 6.6 m g/l and 9.2 m g/l.
The O-Phosphates (PO43-)
Analysis of the results shows very small phosphate concentrations ranging from (0.067 ± 0.02) mg/l to (0.065 ± 0.04) mg/l for groundwater and surface water respectively (Fig. 8). These phosphorus values are significantly lower than the maximum EPA (US – Environmental Protection Agency) tolerated value of 0.1 mg/l and those found in several other countries like South Africa, Brazil, or New Zeeland (Warrack et al. 2021), but are close to those obtained by Tapsoba (2016) at the level of the dam N°3 of Ouagadougou (0.14 m g/l). A study conducted by Hammami et al. (2005) showed that NO3− and PO43− ions found in surface waters are due to the use of organic fertilizer (organic manure). According to (Zacharia 2002), concentrations of more than 0.5 mg/l and 0.02 mg/l of NO3− and PO43− respectively in surface waters, indicate pollution levels that may cause eutrophication. In situ observations and the low concentration of less than 0.07 mg/l in surface and groundwater of Boulbi show that eutrophication is not currently a major risk.
Sulphate (SO42-) and Potassium K+
Sulphate levels in the water samples ranged from (0.0 to 3.0) m g/l, with an average of (1.18 ± 1.07) mg/l for groundwater and (0.57 ± 0.7) mg/l for surface water (Fig. 9). These values are lower than the WHO standard limits of 250 mg/l. They are close to those obtained by Ayouba (2015) in the locality of Yamtenga in Burkina Faso, values ranging from (0.0 to 9.0) mg/l. A study conducted in the sub-watershed of Tougou dam by Djibo (2020) showed that sulphate values are higher during rainy periods than those without rain, which can be explained by the variation of the fertilizer dose as the plant grows.
The results of the potassium ion concentration in the samples give an average value of (5.05 ± 1.9) mg/l for surface water against (4.44 ± 1.4) mg/l for well water. These potassium concentration values are below the WHO standard limits (12 mg/l) (Meride and Ayenew 2016). Potassium is generally less abundant in water and does not usually exceed 10 m g/l.