Development of a Rapid Assessment and Monitoring Technique for Soil Erosion in Malawi Based on Analysis of the Occurrence of Broken Ridges
III. Validation of a Method for Qualitative Assessment of Soil Erosion from Smallholder Farmersí Fields in Malawi.
Dr. Yusuf Mohamoud
III. Validation of a Method for Qualitative Assessment of Soil Erosion from Smallholder Farmersí Fields in Malawi.
To validate whether ridge breakage is a reliable indicator of soil erosion occurrence and damage, two validation methods were used. The first method is based on the relationship between broken ridges and factors influencing soil erosion. The second method is a more direct method and is based on the relationship between broken ridges and suspended sediment concentrations measured from streams and soil loss data measured from small experimental plots that are adjacent to farmersí fields. The first validation method revealed that there is a significant relationship between number of broken ridges and 30-minute maximum rainfall intensity. Number of broken ridges and rainfall intensity had a Spearmanís correlation of r = 0.59. Data from farmersí fields also showed an exponential relationship between broken ridges and slope. More broken ridges were recorded on farmersí fields located on areas with higher slopes than on areas with lower slopes. We have also observed that broken ridges decrease with increase in crop canopy development and more ridge breakage and soil erosion occurs during the early part of the rainy season. The second validation method also showed that broken ridges are closely related to suspended sediment concentrations measured from streams and soil loss data measured from small experimental plots. Based on the results from both validation methods, it can be concluded that broken ridges are closely related to soil erosion and can therefore be used as qualitative indicators for assessing soil erosion occurrence and damage.
In Malawi, over 85 percent of the population live in rural areas where agriculture provides both income and employment (Kapila et al. 1995). Soil erosion and loss of soil fertility and consequently reduction in crop yield are threatening the livelihoods of the rural population. Several factors are contributing to increased soil erosion, but human activities particularly those that influence land use and land management practices are often responsible for the accelerated soil erosion. The Malawi National Environmental Action Plan (NEAP) (DREA, 1994) listed nine environmental problems and ranked soil erosion as the number one environmental problem. Soil erosion is a serious threat to land productivity as evidenced by reduction in crop yields. It also contributes to the degradation of surface water bodies and loss of reservoir storage capacity and consequently loss of hydro-electric power generation capacity in Malawi.
Despite its ranking as a major environmental problem in Malawi, the extent of the soil erosion problem and its geographic distribution are not known. In addition, sufficient research that can lead to the selection of proven technologies for soil erosion control has not been conducted. For instance, contour ridging has been widely adopted and despite its widespread use, accelerated soil erosion is occurring. To control soil erosion, it is important to know which conservation practices are effective against soil erosion under different soils, topography and rainfall regimes. Lack of suitable methodologies to assess soil erosion rates and evaluate effectiveness of conservation practices is a major constraint in Malawi. There is therefore a need to develop and validate simple and reliable methods for assessing soil erosion. A method based on counting the frequency or the number of times broken ridges are recorded and number of broken ridges after each rainfall event has been developed and tested on 30 farmersí fields in Kamundi watershed. The objective of this part of the study is to validate whether broken ridges can be used as qualitative indicators of soil erosion. This has been accomplished by relating broken ridges and factors influencing soil erosion or by relating broken ridges and soil loss from experimental plots and suspended sediment data from streams.
To validate the relationship between broken ridges and soil erosion, broken ridges were recorded from 30 farmersí fields located at Kamundi watershed and soil erosion data was measured from streams and experimental field plots.
The Kamundi watershed is located near Mangochi (latitude 140 33' S and longitude 350 03' E). The watershed has a sandy loam soil with shallow depths characterized by relatively high permeability layers overlaying layers with low permeability. It has an area of about 7 km2 and consists of three small subbasins. In this study, data was collected only from subbasin 1 and subbasin 2 (Figure 3).
To study the relationship between broken ridges observed from farmersí fields and soil erosion occurrence and damage, we used two validation methods. The first method relates broken ridges to factors influencing soil erosion such as field slope, rainfall intensity, and crop height in lieu of crop canopy cover. The second method relates broken ridges to suspended sediment concentrations measured from streams draining farmersí fields and from small experimental plots located near farmersí fields (Figure 3). Detailed discussion of each method is given in the following paragraphs.
Frequency of broken ridges on a farmerís field or number of broken ridges recorded after a rainfall event can be influenced by several factors that include rainfall characteristics, farmersí conservation practices, field slope, and antecedent soil moisture. The method relates the number of broken ridges to rainfall characteristics namely 30-minute maximum rainfall intensity, crop height as a substitute for crop canopy, and field slope. Based on these relationships, the suitability of the method for assessing soil erosion can be determined. Soil erosion increases with increase in slope and rainfall intensity and decreases with increase in crop canopy. A reliability test of the validation method is therefore to show that broken ridges increase with increase in slope and rainfall intensity and decrease with increase in crop height and thus crop canopy.
Suspended sediment samples were collected from sub-basin 1 and sub-basin 2 after each rainfall event. Water samples were taken to the Central Water Laboratory at Lilongwe, Malawi for suspended sediment, phosphate, and nitrate analysis. Data collected from streams draining sub-basins where farmersí fields are located was used for validation purposes. Higher suspended sediment concentrations mean more soil erosion. If more suspended sediments concentrations correspond to more broken ridges, it can concluded that broken ridges are related to soil erosion.
To validate whether broken ridges are closely related to soil erosion, we set up three small experimental plots (5m by 10m) and measured soil loss from the plots. The experimental plots were located at sub-basin 1 (Figure 3). During the 1997/98 rainy season, runoff and sediment data were collected from the experimental plots in the Kamundi watershed. The three plots received three different treatments. Plot 1 was tilled and left bare through out the rainy season, Plot 2 was tilled with 50% corn residue cover, and plot 3 was tilled with maize planted on top of ridges. The ridges were 90 cm apart and 30 cm high. To avoid water entering the plot from outside, galvanized iron sheets were inserted about 15 cm into the ground. Runoff and sediment from each plot were collected at a pit located at the lower end of the plot after the end of each runoff producing rainfall event.
Rainfall was measured using standard and recording rain gauges and runoff was collected at the pit after each rainfall event. Supernatant samples were collected from the pit by thoroughly stirring the water and taking samples for suspended sediment and nutrient analysis. After the water samples were collected, the supernatant material was carefully removed using a bucket. When the water level was lowered just 2 cm above the sludge, water was sucked from the sludge using a spongy material. The sediment deposited at the pit was carefully levelled and the depth of the sediment was measured. A sediment sample was removed from a section of the pit and the area of the section was measured. After the sample was taken, the pits were left clean and free of debris. The sample was put in a plastic bag and was taken to the laboratory. It was dried in the oven at a temperature of 105 0C. The weight of the sediment was determined by gravimetric methods. Total sediment weight in the pit was estimated using the ratio between the area of the pit where the sample was removed to the total area of the pit. Among the three plots only plot 3 had a treatment that closely resembles farmersí conservation practices. To validate the method, soil loss measured from Plot 3 during rainy days was compared with the number of broken ridges observed from farmersí fields that are adjacent to the experimental plots.
In areas with ridges aligned on the contour, ridge breakage occurs when rainfall intensity exceeds the infiltration capacity of the soil and the depression storage on the soil surface. Maximum 30-minute rainfall intensity contributes to ridge breakage and soil erosion. It is defined as the maximum intensity recorded in 30 consecutive minutes during a rainfall event. Figure 4 shows the relationship between the number of broken ridges and maximum 30 minute intensities recorded during the early part of the rainy season. Within a given watershed where soils have similar soil erodibility, maximum 30-minute intensity is an important factor that causes ridge breakage and soil erosion occurrence.
The relationship between number of broken ridges and maximum 30 minute intensity has a Spearmanís rank correlation of rs =0.59 which was significant at the 5% probability level. As the rainy season progressed, crop canopy developed and protected the soil from the impact of raindrops. For example, two rainfall events with similar characteristics may not result in similar number of broken ridges if one rainfall event occurs at the onset of the rainy season and other occurs one month later. Because of changes on the soil surface, the relationship between rainfall intensity and number of broken ridges may be valid only during the early part of the season when the soil is bare. Maximum 30 minute intensity is related to soil erosion and is an input parameter for the Universal Soil Loss Equation (USLE), (Wischmeier and Smith, 1978). It is not therefore surprising to have significant correlation between broken ridges and maximum 30-minute rainfall intensity.
To evaluate the effect of slope steepness on number of broken ridges and thus on soil erosion, farmersí fields were grouped into slope classes and number of broken ridges were recorded in each slope class. Among the 30 farmersí fields, the number of farmers under slope classes 0-5%, 6-9%, and greater than 10% were 40 percent, 40 percent and 20 percent, respectively (Figure 5). In general, about 58% of the land area of Malawi have slopes that are greater than 8 percent (Newson, 1993). Results obtained by monitoring the number of broken ridges after an intense rainfall event, which occurred on December 12, 1997, show that the number of broken ridges increase with increase in slope steepness. The fact that more broken ridges were recorded on higher slopes indicates that more erosion occurred on fields with higher slopes than on fields with lower slopes. It also indicates that farmersí current conservation practices are not effectively controlling soil erosion on steep slopes.
The degree of association between slope and number of broken ridges was determined using a Spearmanís rank correlation. Because the two sub-basins have different slopes, a separate test was conducted for each sub-basin. Farmersí fields located at Sub-basin 1 had lower slopes and Spearmanís rank correlation between slope and broken ridges was low and not significant at the 5% level. On the contrary, farmersí fields located at Subbasin 2 had relatively higher slopes and the correlation between the number of broken ridges and slope was high and was significant at the 5% level.
Figure 4. Relationship between broken ridges and maximum 30 minute rainfall intensity
Figure 5 shows exponential relationship between number of broken ridges and slope. The relationship between number of broken ridges and slope is similar to the relationship between soil loss and slope (Wischmeier and Smith, 1978; and Elwell, 1978).
In Malawi, farmers burn or bury residue cover after harvest. At planting time, there is very little residue cover on the soil surface and maximum number of broken ridges was recorded at this period. More broken ridges were observed when the crop height was less than 60 cm (Figure 6). At the early part of the rainy season, the soil was freshly tilled and residue cover and crop canopy were not adequately covering the soil surface. As more canopy cover is established and the loose soil becomes consolidated by successive raindrop impact, the soil became less erodible and reduced frequency of broken ridges and number of broken ridges were recorded. Reduction of broken ridges after canopy development demonstrates the importance of canopy cover as a conservation practice.
The relationship between suspended sediment concentrations and number of broken ridges measured after each rainfall event determines the relationship between broken ridges and soil erosion (Figure 7). Number of broken ridges measured from a farmerís field and suspended sediment concentration measured from a nearby stream have a coefficient of determination of R2= 0.57 with a significance level of 5%. The relationship supports the general understanding that the more broken ridges observed on a farmerís field after a rainfall event the higher the suspended sediment concentrations measured from nearby streams. The relationship between broken ridges and suspended sediments is also influenced by the intensity of the storm since more broken ridges and more suspended sediments are observed during intense storms.
Among the three treatments on the experimental plots, the treatment on Plot 3 closely resembles farmersí conservation practices. To validate whether broken ridges have close relationship with soil erosion, soil loss data obtained from Plot was related to number of broken ridges recorded from a nearby farmerís field. Accumulated soil loss measured from Plot 3 and accumulated number of broken ridges recorded from a
Figure 5. Relationship between slope classes and ridges broken by a storm that occurred on December 12, 1997.
Figure 6. Relationship between accumulated number of broken ridges and crop height
Figure 7. Relationship between number of broken ridges recorded from a farmerís field and suspended sediment concentrations measured from a nearly stream
nearby farmerís field have a coefficient of determination (R2 =0.95). Both broken ridges and accumulated soil loss showed that more erosion occurred during the early part of the rainy season as shown by the slope
of the cumulative curves (Figure 8). Figure 8 also shows that more soil loss from Plot 3 after a heavy storm when ridges on Plot 3 were broken. A horizontal shift of the accumulated soil loss curve indicates days when rainfall was not heavy and ridges were not broken. A vertical shift of the soil loss curve indicates days when rainfall was heavy and when broken ridges recorded and soil loss was measured. It appears that accumulated broken ridges measured from farmersí fields and soil loss from Plot 3 have similar trends since days when erosion occurred on Plot 3 correspond to days when more broken ridges were reported from the farmerís field.
A simple and less costly method for assessing soil erosion was validated. The method is based on recording the frequency of broken ridges and number of broken ridges after each rainfall event. The relationship between the number of broken ridges and soil erosion was validated using two different methods. The first method employs the relationship between number of broken ridges and soil erosion by studying the relationship between broken ridges and factors influencing soil erosion such as rainfall intensity, slope, and canopy cover. Results of the validation method I showed that the number of broken ridges increase with an increase in slope and rainfall intensity and decrease with increase in crop height and thus canopy cover. The second validation method directly evaluated the relationship between broken ridges and soil erosion by studying the relationship between broken ridges and suspended sediment concentrations measured from streams and soil loss measured from experimental plots. The second validation method also showed that both soil erosion measured from Plot 3 and Based on the results obtained from both validation methods, we can conclude that broken ridges are closely related to soil erosion. Broken ridges can therefore be used as simple and reliable indicators of soil erosion occurrence and damage from smallholder farmersí fields.
Figure 8. Relationship between accumulated number of broken ridges measured from a farmerís field and accumulated number of soil loss measured from an experimental plot
DREA. 1994. Malawi National Environmental Action Plan. Department of Research and Environmental Affairs, Lilongwe, Malawi.
Elwell, H. A., 1978. Soil Loss Estimation: Compiled Works of Rhodesian Multi-disciplinary Team on Soil Loss Estimation. Department of Conservation And Extension. Causeway, Zimbabwe. August, 1978. 145p.
Kapila, M.I.M., and Mwafongo, W. M. K., 1995. Status of Soil Degradation in Malawi. A Paper Presented at the SADC Regional Workshop on Soil Erosion and Degradation, Monitoring Methodologies and Techniques, Harare, Zimbabwe, 4-7th December 1995.
Newson M. D., 1993. Measuring River Sediment Yields in Southern Africa. In: Subsaharan Africa Hydrological Assessment SADCC Countries. World Bank/UNDP Regional Report.
Wischmeier, W.H., and Smith, D. D., 1978. Predicting Rainfall Erosion Losses- A Guide to Conservation Planning. In Agriculture Handbook No. 537. Washington D. C., USDA, Government Printing Office.