Abstract: This study was conducted in 2004 to investigate variability and relationship between sodium concentration and erodibility of soils formed over different lithologies. A free survey technique guided by the geological map of the study area was adopted in field sampling, which was followed by routine laboratory analyses. Soil data were subjected to Analysis of Variance (ANOVA) using PROC Mix-model of SAS computer software and correlation analysis. Results showed that soil groups had very low sodium concentration (Exchangeable Sodium Percentage = 0.3-1.4) and this trend was followed by depth distribution (Exchangeable sodium percentage = 0.5-1.2). Soils were highly erodible spatially (Dispersion Ratio = 28.7-83.7%) and with depth (Dispersion ratio = 62.7-65%). While soil dispersability had good relationship with bulk density (R = 0.51; p = 0.05), clay (R = -0.62; p = 0.05) and sand (R = 0.66; p = 0.05), it had non-significant relationship with exchangeable sodium percentage at the same level of probability. It becomes necessary to consider other edaphic and soil-related factors for more reliable assessment of erodibility factors in studied soils.
INTRODUCTION
Among several forms of environmental degradation occurring in the rainforest belt of the humid tropics, soil erosion by water is the most prevalent. It has reduced the productive capacity of soil through unfavourable changes in soil texture and nutrient losses. Soil erosion is associated with high silt-clay ratio of soil since it does not promote weathering and pedogenesis (Igwe et al., 1995). Physical damages caused by soil erosion include increased gravel and decreased silt and clay contents, increased bulk density, reduced total and macro-porosity, infiltration capacity and saturated conductivity (Mbagwu, 1988). Loss of soil productivity following soil erosion by water also manifested in unfavourable change in soil chemical fertility. Plant essential nutrients, such as nitrogen, phosphorus, potassium, calcium, magnesium and others are lost with runoff and sediments.
Biological parameters of soil fertility are adversely affected by soil erosion, Eroded soils contain less micro-and macro-organisms (Mbagwu, 1988) and this drastically reduces activity and transformations within the pedosphere.
Vulnerability of soils to erosive forces depends on among other factors the nature of soils (Igwe et al., 1995; Wang et al., 2001), land use and soil management practices (Fu et al., 1999; Hontoria et al., 1999) and a combination of land use and lithological formations (Kosmas et al., 2000). Organic matter is also considered a major binding agent that stabilizes soil structure (Haynes et al., 1991) and this is possibly achieved by chemical bonding of soil particles by fungal hyphae and plant Roots (Miller and Jastrow, 1990; Angers, 1998), a combined bonding mechanisms between clay and organic matter (Mikha and Rice, 2004). The above interactions and relationships with soil erodibility have been well documented in soils of southeastern Nigeria by Akamigbo (1983) Mbagwu (1988) Igwe et al. (1995) Igwe et al. (2002) Mbagwu and Auerswald (1999) and Igwe and Stahr (2004). While literature on the extent and severity of erosion of soils of southeastern Nigeria is voluminous, there is paucity of information on specific relationships between soil erodibility and sodium concentration of soils and their variability among different soil groups of the rainforest agroecosystem. But this knowledge would be helpful and fundamental in introducing effective and corrective measures aimed at preventing and minimizing catastrophic soil losses in the region. Based on the above, we investigated the differences and relationships between sodium concentration and soil erodibility in lithologically different soils of southeastern Nigeria. We hypothesized that these soils vary in their susceptibility to erosion due to variability in sodium concentration.
MATERIALS AND METHODS
Study Area
The study was conducted at central southeastern Nigeria comprising Abia
and Imo States in 2004 and 2005. The site lies between latitudes 4°4011011
210 and 8°1512011.110 N and longitudes 6°4010511.
210 and 8°1513011.310 E. The area is dominated by
lowlands with few scarpy landscapes in a Northeast orientation. Soils are formed
over six major parent materials namely Alluvium, Coastal Plain Sands, Falsebedded
Sandstones, Lower Coal Measures, Shale and Upper Coal Measures. Annual rainfall
ranges from 1800 to 2500 mm and characterized by 9 months of rainy season and
3 dry months. Rainforest vegetation dominates the study area with patches of
marshlands along natural water bodies. Low-Input-low-output agriculture is the
main socio-economic activity.
Field Studies
A reconnaissance study of the site was conducted in the early months of
2004 and this was quickly followed by field sampling which was guided by a base
map derived from a geological map of the study area. A free survey approach
involving a target sampling was used. On each soil group, 5 profile pits were
dug, described and sampled according to the procedure of FAO (1990). Soil samples
were collected from both surface and subsurface horizons although ‘soil
erosion is more of a surficial phenomenon. Collected soil samples were air-dried,
gently crushed and sieved using 2 mm sieve.
Laboratory Determination
Particle size distribution was determined by hydrometer method according
to the procedure of Gee and Or (2002) using both water and sodium haxametaphosphate
as dispersants. Exchangeable sodium was obtained through inductively coupled
plasma spectroscopy on 1: 10 soil/Mehlich-3 extracts (Mehich, 1984) while cation
exchange capacity was estimated using the procedure of Darmody et al.
(2000).
Calculations
Dispersion ratio was used as an indirect measure of soil erodibility (Middleton,
1930) and computed as follows:
Where,
DR = Dispersion Ratio
Exchangeable sodium percentage was used as an index of sodium concentration and was calculated as follows:
Where,
| ESP | = | Exchangeable Sodium Percentage |
| CEC | = | Cation exchange capacity (NH4OAC) |
| Na | = | Sodium |
Statistical Analyses
Soil data were subjected to Analysis of Variance (ANOVA) using the PROC
Mix-model of SAS (Little et al., 1996). Means were separated using Standard
Error of the Difference (SED) at 5% level of probability. Thereafter values
of DR were correlated with some soil properties to ascertain degree of relationship.
RESULTS AND DISCUSSION
Soil Physical Properties
Soils were generally sandy although sandiness varied significantly (p<0.0001)
among parent materials (Table 1). Similar variations were
found in silt-and clay-sized particles (Table 1) and in their
Silt-Clay Ratio (SCR). Soils were highly weathered as indicated in their low
SCR (SCR = 0.023-0.64) (Table 1). There was non-significant
(p = 0.05) relationship among soil groups in Bulk Density (BD) values (Table
1). Particle size distribution varied possibly due to differences in lithological
materials at the regional scale while recorded similarities in sandiness and
bulk density can be attributed to climate, land use and land use history of
the study area. Least SCR encountered in soils derived from. Upper Coal Measures
implies higher weathering and stability in line with the observation of Igwe
et al. (1995) that the higher the SCR, the younger the soils and that
higher SCR values are associated with landscapes devastated by soil erosion.
But the SCR of Upper Coal Measures is in contrast with its BD value which is
the highest among soil groups (BD = 1.47 mg m-3), suggesting higher
possibility of aggregate instability and erosive impact. This is because higher
BD values suggests reduced porosity and higher build-up of runoff water and
consequent deterioration of macro-aggregates (Park and Smucker, 2005) and intraaggregate
pores (Paustian et al., 1997).
Sodium Concentration
Sodium saturation in soil groups differed but were generally very low in
studied soils (Exchangeable Sodium Percentage <1.5%), while soils are said
to be saline if ESP is greater 15% (Michael, 1985). Low sodium content of soils
is attributable to the high rainfall amount and duration.
| Table 1: | Selected physical properties of studies soils (n = 150) |
| SED: Standard Error of the Difference ; SCR: Silt-Clay Ratio; BD: Bulk Density; NS: Not Significant | |
| Table 2: | Distribution of ESP among soil groups (n = 15) |
| ESP = Exchangeable Sodium Percentage, SED = Standard error of the difference in means | |
| Table 3: | Change in distribution of ESP with depth (n = 150) |
| ESP = Exchangeable Sodium Percentage, SED = Standard error of the difference in means | |
| Table 4: | Parent material-depth interactions in ESP of soil of the study site (n = 150) |
| SED: Standard error of the difference in means | |
As rainwater percolates through the pedosphere, it dissolves and leaches away sodium cations which may accumulate in ground water, implying that the amount of Na-concentration in ground water may be proportional to the amount of soluble Na-cations leached out of top and sub soils. This is consistent with the observation of Dupriez and Deleener (1992) that rain water falling on the surface of a field causes soils to hardly be associated with any saltness.
Comparatively, soils formed over Alluvium had the highest ESP value, (Table 2) possibly due to marine influences, as sampled sites belong to the River Niger delta region of Southeastern Nigeria thus proximal to the influences of the Atlantic Ocean. Generally, Na-concentration varied and increased significantly (p<0.05) with depth (Table 3). Variability in depth distribution of Na-concentrations could be attributed to precipitation-dissolution reactions (Khoshgoftar et al., 2004). The distribution of pore sizes in the soil and boundary conditions (Hamlen and Kachanoskil, 2004), presence of restrictive zones (Shaw et al., 1997) and orientation of tubular pores in soils (Ezeaku and Anikwe, 2006).
Exchangeable sodium percentage values were mainly only significant (p<0.05) at epipedons when parent materials interacted with depth (Table 4) except in soils derived from Shale and Upper Coal Measures, possibly due to high micro-porosity of Shale derived soils and high bulk density in soils formed over Upper Coal Measures.
| Table 5: | Distribution of dispersion ratio among soil groups (n = 150) |
| DR = Dispersion Ratio, SED = Standard error of difference in means | |
| Table 6: | Changes in dispersion ratio with depth (n = 150) |
| DR: Dispersion Ratio; SED: Standard error of difference in means | |
| Table 7: | Parent material-depth interactions erodibility of soils of the study site (n = 150) |
| SED = Standard error of the difference in means | |
Erodibility
All investigated soils were erodible since dispersion ratio was greater
than 15% in all soil groups (Table 5). Highest mean erodibility
value was recorded in soils formed over Upper Coal Measures although it had
least SCR (SCR = 0.23) and ESP (ESP = 0.4%) in line with reports from Mbagwu
and Auerswald, 1999) indicating that the concentration of Na is not a major
determinant in dispersion and erodebility of soil of the study area. In a similar
study Mbagwu and Auerswald (1999) attributed soil structural instability and
vulnerability to erosive forces to land use. However, the presence of various
forms of sesquioxides in the region may be main factors determining erodibility
of soils since iron, aluminum and manganese form bridges between clay and organic
matter in the formation of stable aggregates (Igwe and Stahr, 2004).
Most studies in the area investigated surficial soil erodibility (Mbagwu, 1988; Igwe et al., 2002; Mbagwu and Auerswald, 1999), but results of this study (Table 6) show that sub-surficial layers are erodible and that erodibility increased somewhat with depth (p>0.0001). These values suggest that removal of epipedal layers paves way for rapid dispersion and erosion of deeper soil layers and this could be responsible for deep gullies in Southeastern Nigeria. There were significant (p<0.0001) interactions between parent materials and depth in erodibility of soils (Table 7), pointing to the need for both aspects in soil conservation modelling in the study area.
| Table 8: | Correlation coefficients for linear relationships between dispersion ratio and some soil properties (n = 150) |
| DR: Dispersion Ratio; BD: Bulk Density; *: Significant at p = 0.05, NS = Not Significant | |
However, there was non-significant relationship between DR and ESP in contrast to values obtained when DR related with BD, clay and sand, (Table 8) showing that particle size influences erodibility with clays being aggregating agents (Igwe and Stahr, 2004).
In conclusion, this study has revealed that low SCR may not serve as a good indicator of structural stability in the study area. Secondly the study has shown that ESP is not the major component determining variable in the erodibility of studied soils Again soils differed in their vulnerability to erosive forces in both space and depth, suggesting the need for their considerations in conservation modelling ventures. Large scale studies may be necessary in future investigation for increased accuracy of predictions. Finally, more attributes of soil resource should be investigated to create greater confidence.