Assessment and Mapping of Wind Erodibility of Aridisols and Entisols
in the River Nile State, Sudan
Abdelmonim Ahmed Hassan* and Mukhtar Ahmed Mustafa
Desertification and Desert Cultivation Studies Institute,
University of Khartoum, Shambat, Sudan
ABSTRACT
Wind erodibility (WE) is a major determinant of wind erosion under a given
climatic condition. This study is part of a national project for assessing and mapping of
wind erodibility in Sudan. It was undertaken to generate WE data for the River Nile
State. Three replicate surface soil samples were collected, randomly, from fifty georeferenced farms spread in the State, and non-erodible soil particles (NEP > 0.84 mm)
and selected soil properties were measured using standard procedures. The mean NEP
values ranged from 4.3% to 98.1% with an overall mean coefficient of variation of
replicate determinations equal to 7.4%. The equivalent WE ranged from 0.0 to 470.4
ton/ha. The results showed a highly significant (P ≤ 0.001) power increase of NEP with
increase of clay (C), CaCO3 and organic matter (OM), and decrease with increase of sand
and sand plus silt (S+Si) expressed, successively, as ratios of clay, clay plus OM and clay
plus CaCO3. The reverse trends were obtained for the relations of WE and the various
soil properties and their ratios. Clay, (Si+S)/C, (Si+S)/(C+OM) and (Si+S)/(C+CaCO3)
accounted for 72%, 72%, 72% and 73% of the variation of NEP, and 69%, 68%, 68% and
69% of the variation of WE. Multiple regressions relationships of NEP or WE with clay,
sand, CaCO3 and OM gave coefficients of determinations equal to 65% and 69%,
respectively. Thus, it is recommended that clay or (Si+S)/ (C+CaCO3) should be used for
predicting NEP, and the standard table should be used to get the equivalent WE. A table
for wind erodibility groups was developed.
Key words: Wind erosion; Aridisols; Entisols; River Nile State
*
Current address: National Institute of Desert Studies, University of Gezira, Al-Dabba,
Northern State, Sudan
INTRODUCTION
The Nile State lies between latitudes 16º and 22º N and longitudes 31º 88' and 35º 70' E.
Its total area is about 121,000 km2. The climate of the State is desert and semi-desert with
mean annual rainfall less than 200 mm. Normally, the temperature exceeds 49 ºC in
summer, but in winter temperatures may decrease to as low as 1.5 ºC. Winds prevail from
the north with a mean maximum velocity of 17.8 km hr-1 (Izzeldin and Ahmed 2004).
The soils are Aridisols and Entisols.
Wind erosion is the predominant desertification process in the Nile State. It
depends on two main factors; namely, soil erodibility referred to hereafter as wind
erodibility of soil and wind erosivity. The former is an indicator of the soil vulnerability
to detachment, and the latter is an indicator of the erosive energy of the wind to cause
transport of the soil particles. Wind erodibility (WE) is a major indicator of wind erosion
and it was used for the prediction of wind erosion by the soil loss equation (Woodruff and
Siddoway, 1965). In general, soil erodibility depends on many factors such as
topographic position, slope steepness and soil management that causes soil's disturbance,
e.g. tillage, but the most important factors are the soil properties (Morgan, 1995).
Previous research showed that WE is determined, mainly by the soil properties, e.g.
particle-size distribution, structural stability, organic matter content, nature of clay
minerals, and chemical constituents (Harris et al., 1966; Wishmeier and Mannering,
1969; Romken et al., 1977; Lyles and Tatarko, 1968; Lal, 1988; Black and Chanasyk,
1989; Medani and Mustafa, 2003; Mustafa and Medani, 2004). Soil texture has the major
impact on WE because it affects both the detachment and transport phases of the process.
Under a given climatic zone, wind erodibility is the main determinant of wind erosion. It
is a good indicator for the prediction, assessment and mapping of wind erosion. For
quick assessment of WE, the concept of wind erodibility groups (WEGs) was proposed
(Chepil, 1962; Chepil and Woodruff, 1959; Hayes, 1965; Black and Chanasyk, 1989)
The present research is part of a national project designed to assess and map soil
erodibility in Sudan. Its specific objectives are (i) estimation of wind erodibility of soil
samples selected from different locations spread in the State, (ii) identification of
appropriate soil indicators for wind erodibility, (iii) establishment of wind erodibility
groups (WEGs) for the State and (iv) mapping of wind erodibility classes.
MATERIALS AND METHODS
Three replicate surface (0-3 cm) soil samples were collected at random from each of fifty
farms spread in the Nile State, starting from Al-Sabalogha (16º 36' N and 32º 84' E) and
ending at Abu Hamad (19º 53' N and 33º 33' E). The farms were geo-referenced using a
GPS. The samples were collected after completion of land preparation for crop
cultivation. The samples were carefully saved in bags to avoid fragmentation of natural
aggregates. They were air-dried and stones and straw, if present, were removed. Wind
erodibility was determined by the dry sieving method proposed by Chepil and Woodruff
(1959). One kilogram of each sample was sieved through 0.84 mm sieve, and the
percentage of particles greater than 0.84, referred to hereafter as non-erodible soil
particles (NEP) was determined. A standard table developed by Woodruff and Siddoway
(1965) was used to obtain WE. The soil samples were, then, crushed, passed through 2-
mm sieve and saved for physical and chemical analysis. Soil particle-size distribution
was measured by the hydrometer method (Black et al., 1965) and the texture class of
each sample was determined using USDA texture triangle. Saturated soil paste was
prepared for each sample and the pH of the soil paste was determined. The electrical
conductivity of the saturation extract (ECe) was measured using a conductivity meter.
Calcium carbonate content was determined using a calcimeter. Calcium and magnesium
were determined by titration against EDTA according to the method described by
Chapman and Pratt (1961). Organic carbon was determined by the dry-ashing method
proposed by Fredrick as reported by Ibrahim (1991) and organic matter (OM) was
calculated. Sodium was determined using a flame photometer, and sodium adsorption
ratio was calculated by the following equation: SAR = [Na+]/ {[Ca+2+Mg+2]/2}, where
the ionic concentrations were expressed in me/l.
The mean percentage of NEP values of each texture class was calculated and a table of
WEGs was presented. Simple statistical parameters and the equations of trendlines were
obtained using Microsoft Excell package. GIS was used to map the spatial variation of
WE in the Nile State, Sudan.
RESULTS AND DISCUSSION
The mean clay percent of the soil samples from the fifty fields ranged from 5.7 to 47.8
with an overall mean coefficient of variation of replicate measurements of each field
(CVr) equal to 6.2%. The mean silt percent ranged from 4.1 to 32.8 with a CVr equal to
5.5%. The mean sand percent ranged from 34.6 to 87.5 with a CVr equal to 3.4%. The
very low CVr values reflect a very high precision of replicate measurements of the three
particle-size fractions. These primary particles are durable and not susceptible to
significant change by management. The particle-size distribution resulted in a dominant
sandy clay loam texture, which characterized the surface soils of thirty farms. The surface
soils of five farms were clay, five were sandy clay, two were clay loam, seven farms were
sandy loam, and one was loamy sand.
The mean CaCO3 percent ranged from 0.3 to 7.9% with a CVr equal to 6.2%. This
compound is slightly soluble and relatively durable and this explains why it has a very
low spatial variation indicative of precision of replicate measurements.
The mean organic matter percent in these arid-zone soils is low ranging from 0.01 to 0.6
with a CVr equal to 15.2%. The moderate spatial variation may be explained by the fact
that OM is greatly affected by non-uniform soil and plant residue management.
The mean ECe values ranged from 0.4 to 26.2 dS/m with a CVr equal to 23.5%. The
mean SAR values ranged from 0.9 to 49.1 (me/l)1/2 with a CVr equal to 39.6%. The high
CVr values are due to the inherent high spatial variation of salts, which is profoundly
affected by interactive effects of land micro-relief and field water management (Ibrahim
and Mustafa, 2001).
The mean NEP values ranged from 4.3 to 98.1% with a CVr equal to 7.4%. The CVr
values are low because NEP is constituted from durable primary soil particles. The
equivalent WE ranged from 0 to 470.4 ton/ha.
Relationships between NEP and soil properties
The results show a highly significant (P<0.001) power increase of NEP with increase of
clay (r = 0.8494), CaCO3 (r = 0.5389) and OM (r = 0.5617), and decrease of sand (r = 0.7426) (Table 1). Both the clay platelets or domains and derivatives of organic matter
act as cementing agents, promote soil aggregation and, thus, they increase NEP. The
calcium ions in the CaCO3, in spite of its low solubility, promoted flocculation of the clay
domains and, hence, enhanced aggregation. The effect of silt was not significant. Sand,
being inert had a dilution effect on the impact of the cementing agents and hence reduced
NEP. Clay, sand, OM and CaCO3 accounted for 72, 55, 32 and 29% of variation of NEP,
indicating the overriding impact of clay. However, the lack of perfect accountability in
the case of clay may be attributed to the interactive effects of variables other than the one
in question. This is an inherent limitation of regression analysis.
Table 1. Parameters of trend lines showing the relationships of non-erodible soil
particles (NEP, %) or wind erodibility (WE, ton/ha) as a function of various soil
properties
Property Trend a
b
c
r2
r*
(%)
line #
NEP
Clay
Sand
OM
CaCO3
Power
Power
Power
Power
0.3191
510.01
118.09
23.259
1.4262
- 0.0445
0.4121
0.4189
WE
-
0.7214
0.5515
0.3155
0.2904
0.8494
- 0.7426
0.5617
0.5389
Clay
Quad. 0.2046
- 20.7260 559.71 0.6864
- 0.8285
Sand
Quad. 0.1343
- 8.9103 184.09 0.5607
0.7488
OM
Log.
- 73.498 - 69.209
0.3392
- 0.5824
CaCO3
Quad. 9.9948
- 98.548
333.83 0.3680
- 0.6066
# Trend lines: Log. (Logarithmic): a ln X + b; Quad. (Quadratic): aX2+ bX + c; Power: Y
= a Xb;
* Level of significance: r0.05 = 0.2789; r0.01 = 0.3613; r0.001 = 0.4519
Attempts were made to correlate NEP with compound indicators that express noncementing agents as a ratio of cementing/flocculating agents (Abd Elwahab et al., 2009).
Fig. 1 shows highly significant power decrease in NEP with increase of (Si+S)/C,
(Si+S)/(C+OM) and (Si+S)/(C+CaCO3). The three successive compound indicators
accounted for 72, 72 and 73% of the variation of NEP. It is evident that these compound
indictors did not improve the accountability of clay alone. Thus, for this State clay or
(Si+S)/(C+CaCO3) ratio may be used as indicators of NEP. However, we recommend the
use of the compound indicator because it incorporates the
120
A
y = 98.655x-1.0766
R2 = 0.72
NEP (%)
100
80
60
40
20
0
0.00
5.00
10.00
15.00
20.00
(Si+S)/C
120
B
y = 98.368x-1.0766
NEP (%)
100
R2 = 0.7203
80
60
40
20
0
0.00
5.00
10.00
15.00
20.00
(Si+S)/(C+OM)
120
C
-1.0941
y = 88.232x
NEP (%)
100
2
R = 0.7327
80
60
40
20
0
0.00
5.00
10.00
(Si+S)/(C+CaCO3)
15.00
20.00
Fig. 1. Non-erodible soil particles (NEP) as a function of (A) (Si+S)/C (B)
(Si+S)/(C+OM) and (C) (Si+S)/(C+CO3)
three primary particles and the slightly soluble CaCO3. The use of CaCO3 is in agreement
with the criteria used for delineating WEGs (Black and Chanasyk, 1989). This agrees
with the conclusion of Abd Elwahab et al. (2009) for the Northern State.
Regression analysis yielded a highly significant (P < 0.001, R = 0.8056) correlation
between NEP and multiple soil variables as shown in the following empirical
relationship:
NEP = 21.78 + 1.44 clay% - 0.51 sand % + 2.59 CaCO3% + 14.10 OM%
According to this relationship, the four soil properties account for 65% of the variation of
NEP. It is evident that clay alone or the compound indicators gave better accountability
than the four properties expressed in this relationship.
Relationships between WE and soil properties
The results show highly significant (P<0.001) decrease in WE with increase of clay (r = 0.8285), CaCO3 (r = - 0.6066) and OM (r = - 0.5824) and increase with increase of sand
(r = 0.7488) (Table 1). These single soil properties, in sequence, accounted for
approximately 69, 37, 34 and 56% of the variation of WE. The first three soil properties
that promoted aggregation and increased NEP reduced WE, while increase in sand
increased WE.
Fig. 2 shows highly significant logarithmic increase in WE with increase of (Si+S)/C,
(Si+S)/(C+OM) and (Si+S)/(C+CaCO3). The three successive compound indicators
accounted for 68, 68 and 69% of the variation of NEP. The results were nearly similar to
those obtained for NEP. However, the accountability of the compound indicators for the
variation of WE were slightly lower than that for NEP, because determination of WE
included the additional error inherent in the standard table. In view of the lower
accountability it is recommended to predict NEP from knowledge of clay percent or
(Si+S)/(C+CaCO3) and then read the equivalent WE from the standard table.
Regression analysis yielded a highly significant (P < 0.001, R = 0.8313) correlation
between WE and multiple soil properties as shown in the following empirical
relationship:
WE = 373.59 – 7.26 clay% + 0.64 sand% - 13.35 CaCO3% - 58.34 OM%
500
A
WE (ton/ha)
400
300
y = 166.33Ln(x) - 9.4434
R2 = 0.6756
200
100
0
0.00
5.00
10.00
15.00
20.00
(Si+S)/C
500
B
WE (ton/ha)
400
300
y = 166.39Ln(x) - 9.1093
R2 = 0.6762
200
100
0
0.00
5.00
10.00
15.00
20.00
(Si+S)/(C+OM)
500
C
WE (ton/ha)
400
300
y = 169.41Ln(x) + 7.5005
R2 = 0.6905
200
100
0
0.00
5.00
10.00
15.00
20.00
(Si+S)/(C+CO3)
Fig.2. Wind erodibility (WE) as a function of (A) (Si+S)/C (B)
(Si+S)/(C+OM) and (C ) (Si+S)/(C+CO3)
The four soil properties accounted for 69% of the variation of WE. This level of
accountability was similar to those given by clay alone or (Si+S)/(C+CaCO3).
Wind erodibility groups
Table 2 shows that the NEP of the WEGs of the studied soil samples gave highly (p <
0.001) significant simple linear correlation coefficient with those of N. Dakota (r =
0.6663) and Alberta (r = 0.8526).
Table 2. The mean percentage of measured non-erodible soil particles NEP) for the
various wind erodibility groups (WEG) compared with equivalent values
obtained from other States
NEP
WEG
No. of samples
Clay
5
Clay loam
2
Sandy clay
5
Sandy clay
loam
30
Sandy loam
7
Loamy sand
1
Total
50
Measured
N. Dakota
Alberta
80.5
80.5
67.0
25.0
45.0
40.0
60.8
57.9
38.5
17.3
4.3
40.0
25.0
10.0
0.6665
56.2
43.7
21.5
0.8526
Mapping of wind erodibility
The studied soil samples were classified according to their WE values. Four classes were
delineated according to the following WE class limits: 14.6-30.4 ton/ha (low), 30.4 – 83.6
ton/ha (moderate), 83.6 -270.8 ton/ha (high), 270.8 – 470.4 ton/ha (very high). The
spatial variation of WE was mapped according to GIS (Fig. 3).
Fig. 3. Spatial variability of wind erodibility in the Nile State
REFERENCES
Abdelwahab, M.H., Mustafa, M. A., and Ganawa, E.S. 2009. Spatial variation of wind
erodibility of soils from the Northern State, Sudan. Sudan Journal of
Desertification Research 1(1): 56-70.
Black, J.M.W., and Chanasyk, D.S. 1989. The wind erodibility of some Alberta soils
after seeding: Aggregation in addition to field parameter. Canadian Journal of Soil
Science 69: 835-847.
Black, C.A., Evans D.D., Ensminger, J.L., and Clark, F.F. 1965. Methods of Soil Analysis
(part I). American Society of Agronomy, L.E. White, Inc., Publisher,
Madison,Wisconsin, USA.
Chapman, H.D., and Pratt, P.E. 1961. Methods of Analysis of Soils, Plant and Water.
University of California, Division of Agriculture, Calfrnia, USA.
Chepil, W.S. 1962. Stubble mulching to control erosion. Proceedings of the Great Plains
Workshop on Stubble Mulch Farming, Linkon, Nebraska (cited in Black and
Chanasyk, 1989)
Chepil, W. S., and Woodruff, N. P. 1959. Estimations of Wind Erodibility of Farm
Fields. USDA Production Research Report No. 25, 21 pp.
Harris R.F., Chesters, G., and Allen, O.A. 1966. Soil aggregation. Advances in
Agronomy 18: 107-169.
Hayes, W.A. 1965. Wind erosion equation useful in designing North Eastern Crop
production. Journal of Soil and Water Conservation 20: 153-155.
Ibrahim, I.S. 1991. Laboratory Test of Soil Fertility. University of Omar El-Mukhtar, ElBaida, Libya.
Ibrahim, Abdel Samad A., and M. A. Mustafa. 2001. Spatial variation of infiltration rate
and related soil properties in a Central Gezira soil.I. Conventional Statistics.
University of Khartoum Journal of Agricultural Sciences 9: 168-182.
Izzeldin, S.I., and Ahmed, S.H. 2004. A proposed plan of action for research on
desertification in the Sudan: Northern and Nile States. In M.A. Mustafa and A.A.
Mahdi (eds.). 2004. Proceedings of the National Forum of Scientific Research on
Desertification in Sudan, 16-18 March, 2004, Al Sharga Hall, University of
Khartoum, Sudan, Published by UNESCO Chair of Desertification Studies,
Sudan, Khartoum University Press, 325pp.
Lal, R. 1988. Erodibility and erosivity. Soil Erosion Research Methods. Soil and Water
Conservation Society. pp 141-161.
Lyles, L., and Tatarko, J. 1968. Wind erosion effects on soil texture and organic matter.
Journal of Soil and Water Conservation 41: 191-194.
Medani, G.H., and Mustafa, M.A. 2003. Wind erodibility of soils from
North Darfur State. University of Khartoum Journal of Agricultural Sciences 11:
369-384.
Morgan, R.P.C. 1995. Soil Erosion and Conservation. 2nd Edition, Longman Group Ltd.,
UK. Essex, England, pp 198.
Mustafa, M. A., and Medani, G. H. 2004. Wind erodibility of soils from Khartoum State.
University of Khartoum Journal of Agricultural Sciences 11: 149-164.
Romken, M.J.M., Roth, C.B., and Nelson, D.W. 1977. Erodibility of selected clay soils in
relation to physical and chemical properties. Soil Science Society of America
Proceedings 41: 954-960.
Wishmeier, W.H., and Mannering, J.V. 1969. Relations of soil properties to its
erodibility. Soil Science Society of America Proceedings 23: 131-137.
Woodruff, N.P., and Siddoway, F. H. (1965). A wind erosion equation. Soil Science
Society of America Proceedings 29: 602 - 608.