Table of Contents

A method for qualitatively assessing soil erosion from smallholder farmers’ fields has been developed. The method is based on the relationship between ridge breakage on farmers’ fields and soil erosion occurrence and it consists of a rapid assessment technique and a monitoring technique. As the name indicates the rapid assessment technique is based on collecting information from farmers and farmers’ fields. Information from farmers is gathered by interviewing farmers whereas information from farmers’ fields is gathered by visiting farmers’ fields and collecting information about farmers’ practices, location of farmers’ fields in the watershed, slope, and area. The importance of the rapid assessment technique is to give a rough estimate of the severity of soil erosion between different areas or regions in Malawi or among farmers cultivating in a given watershed. The technique can be implemented in a relatively very short time simply by interviewing farmers and visiting farmers’ fields. It is recommended for assessing soil erosion levels when conducting reconnaissance surveys such as the Area Sample Frame (ASF).

On the other hand, the monitoring technique gathers information from farmers’ fields. The technique requires continuous monitoring of broken ridges on each participating farmer’s field after each rainfall event. Reliable assessment of soil erosion occurrence and damage from farmers’ fields requires a field monitoring of at least one or two months depending on the number of storms per month. When testing the effectiveness of farmers’ conservation practices or assessing soil erosion occurrence and damage, it is important to start using the method at planting time when ridges are vulnerable to erosive rainfall.

Soil erosion is a major environmental problem in Malawi and the Government of Malawi ranked soil erosion as the number one environmental problem followed by deforestation and water resources degradation (DREA, 1994). Both on-site effects such as loss of soil productivity and reduced crop yield and off-site effects of soil erosion such as degradation of surface water quality and siltation of reservoirs and irrigation channels were reported by several investigators (Green et al.,1996 and Kapila et al. 1995). Despite being a major environmental problem in Malawi, very little research has been conducted on soil erosion assessment under different land uses and conservation practices. Perhaps the most extensive soil erosion research in Malawi was conducted on small watersheds near Bvumbwe (Amphlett, 1986). Several other investigators have also conducted soil erosion research on small experimental plots (Machera, 1984; Kasambara, 1984; Chome, 1989; Green, 1960; Anon, 1954-70). Limited amount of quantitative information on soil erosion is available in Malawi and constraints encountered when collecting soil erosion data made the results from the experimental plots less accurate. Furthermore, sediment samples obtained from large watersheds may give an estimate of off-site erosion but may not always give reliable results of on-site soil erosion.

Despite widespread use of contour ridging in Malawi, it appears that the effectiveness of conventional contour ridging (without ties or boxes) has not been fully tested. Lack of testing is mainly due to lack of methods to assess soil erosion occurrence and damage. To understand the levels of erosion associated with different management practices and make recommendations for best management practices, we need to develop, test and validate new methods that are suitable for assessing soil erosion from smallholder farmer’s fields. After methods are developed, tested and validated, the methods can be recommended for widespread use by researchers, conservation experts, extension workers, and farmers. The objective of this part of the study is therefore to develop a method suitable for assessing soil erosion from smallholder farmers’ fields by relating soil erosion with the frequency and number of broken ridges.

About 98 percent of the cultivated land of Malawi is grown on annual crops (Saka et al. 1995). In Malawi, planting crops on ridges that are aligned on the contour is the preferred method recommended for controlling soil erosion. Where ridging is practiced, ridges have intervals of approximately 90 cm with a ridge height of about 30 cm. Although ridging is used on almost all the land under cultivation, ridges are not always on the contour and farmers have some difficulty in constructing ridges with proper contour alignment.

There are two areas in Malawi where farmers do not practice contour ridging. Farmers cultivating on the Lakeshore Plain plant cassava on mounds and believe that ridges may create waterlogging and consequently cassava crops may be affected by root rot diseases. Farmers cultivating on the Lower Shire basin also plant crops on a flat soil surface because soils in the Lower Shire basin have poor drainage properties. Because of poor drainage, these soil have reduced infiltration which leads to increased runoff and consequently frequent ridge breakage and soil erosion occurrence. Farmers in both areas recognized through experience that contour ridging is not a suitable practice in these areas.

In the absence of simplified methods for soil erosion assessment, small experimental plots and simulation models are commonly used for assessing soil erosion rates from farmers’ fields. These methods have been extensively used in many parts of the world, but their use in developing countries has met with serious constraints. Some of the limitations of these soil erosion assessment methods are given in the following section.

Experimental plots were initially developed for countries with mechanised agriculture. Its use in countries dominated by subsistence farming has not been successful. There are number of limitations that make the use of experimental plots unsuitable for areas dominated by subsistence farming. For example, time and funding often preclude the use of experimental plots for soil erosion assessment in many developing countries. Other limitations of the method include problems encountered when interpreting results obtained from experimental plots to representative conservation practices on farmers’ fields. These problems arise because smallholder farmers have small size fields and do not use a typical conservation practice but use a modified version of the practice. In addition, the way experimental plots are maintained by researchers is always different than the way farmers maintain their fields. Soil erosion rates obtained from experimental plots may therefore underestimate the true erosion occurring on farmers’ fields. In developing countries such as Malawi where farmers cultivate small parcels of land (Kapila et al. 1995), the use of experimental plots may not therefore give good estimate of soil erosion rates from farmers’ fields. However, in developed countries where the method was developed, it is relatively accurate to assume that soil erosion rates measured from experimental plots are comparable to erosion rates occurring on farmers’ fields. This assumption can be valid because farmers use mechanised agriculture and have large farm sizes with uniform conservation practices throughout the entire field.

Simulation models are also used for assessing soil erosion. Results obtained from models are not accurate unless models are calibrated and validated using actual soil loss data from experimental plots. To validate models, small experimental plots are often set up and model parameters are obtained by relating model results with actual soil loss data obtained from experimental plots. Because of variations in farmers’ conservation practices and management levels, it is difficult to represent all variations of the conservation practices on farmers’ fields when selecting experimental plot treatments.

Empirical models such as the Soil Erosion Estimation Model For Southern Africa (SLEMSA) (Elwell, 1978), Universal Soil Loss Equation (Wischmeier and Smith, 1978), and Revised Universal Soil Loss Equation (RUSLE) (Renard, 1993a) are often used for soil erosion prediction. These models require calibration using actual soil loss data from small experimental plots. Moreover, because of variations in farmers’ conservation practices, it is unlikely that model parameter values obtained from experimental plots capture the variability of all parameter values needed to address the variability in farmers’ conservation practices. Model parameters are also site specific and parameters obtained from a site cannot be extrapolated to other sites without calibration.

An attractive feature of proper contour ridging is its ability to store rainfall in excess of infiltration as surface depression storage. Excess rainfall is stored on the furrow until the storage capacity of the furrow is filled. As rainfall continues, water overtops the ridges and ridge breakage occurs. Although localized soil erosion can occur due to soil detachment by raindrop impact, significant erosion occurs only when runoff occurs due to ridge breakage. Ridges are used almost on all cultivated land in Malawi and the fact that erosion occurs when ridges break presents a strong basis for developing a qualitative method for assessing soil erosion occurrence and damage.

Figure 1 presents a conceptual framework for soil erosion assessment which consists of a rapid assessment and a monitoring technique. The rapid assessment method is based on interviewing farmers and gathering information about farmers and farmers’ fields. Inference about the level of soil erosion or effectiveness of farmers’ conservation practices are made by asking farmers the number of times broken ridges were recorded after a rainfall event. Increased incidence of broken ridges indicates soil erosion occurrence. Unlike the rapid assessment technique where information is gathered through interviews with farmers, the monitoring technique is based on actual counting of broken ridges on farmers’ fields after a rainfall event. After sufficient information is gathered from a location using both the rapid assessment and the monitoring technique, the data are analyzed and decision about soil erosion hazard and effectiveness of farmers’ practices is reached.

The solution of soil erosion control is not to identify suitable conservation practices only but is to understand the farming system as a whole and to identify the political and social issues that are against adoption of better conservation practices. To learn what farmers know about soil erosion and conservation practices, the rapid assessment technique can be used to gather information from farmers’ practices, farmers’ management levels, and farmers’ perception of the causes of soil erosion. Information collected through the rapid assessment technique also include location of the farmers’ field in the watershed, area of the field, slope, type of conservation practices (e.g., contour ridges), contour interval, and ridge height. This information is obtained by interviewing farmers and also by making actual field measurements. When adequate information about farmers’ knowledge of soil and water conservation is gathered, important decision about suitable conservation practices can be reached. When implementing the rapid assessment technique, farmers are asked all the questions given in Tables 1 and 2.

To assess the extent of soil erosion on farmers’ fields, farmers’ fields can be randomly selected and frequency and number of broken ridges recorded after each rainfall event. Soil erosion occurrence is characterised by evidence of broken ridges after a rainfall event and soil erosion damage is characterised by the number of broken ridges after each rainfall event. If conservation practices are effective against soil erosion, ridges are not expected to break too frequently. But when conservation practices are not effective against soil erosion, high frequency of ridge breakage and thus soil erosion occurrence and damage is expected. Field assistants and extension workers can use the technique and determine if a rainfall event resulted in broken ridges by walking through farmers’ fields and counting broken ridges after each rainfall event. The total number of broken ridges will be recorded by counting every place where a broken ridge was observed. To link soil erosion occurrence and damage to rainfall characteristics, we recommend the use of a recording rain gauge for measuring rainfall intensity. Rainfall intensity can be defined as the amount of rainfall falling on the soil surface per unit time. When rainfall intensity is low, rainfall can infiltrate into the soil. But when rainfall intensity is higher than the infiltration capacity of the soil, excess rainfall is stored on the soil surface and overland runoff starts when the storage capacity on the soil surface is filled. Contour ridges with proper alignment enhance surface depression storage and may eliminate runoff and soil erosion on days when rainfall has lower intensity. During intense storms when the capacity of the furrow storage is filled, water overtops the ridges and ridge breakage and serious soil erosion occurs. Table 3 gives the data needed for assessing soil erosion occurrence and damage by recording broken ridges on farmers’ fields after each rainfall event.

When gathering information about soil erosion by interviewing farmers, farmers are asked to describe the severity of soil erosion. Without clear understanding of what soil erosion means, farmers may give vague answers according to their perception of soil erosion. In addition, interviewers always ask farmers to discern levels of soil erosion as low, moderate, severe or very severe. But, farmers may not be able to differentiate between moderate and severe soil erosion. Relating broken ridges to soil erosion gives interviewers and farmers a common terminology with which they can understand each other. Therefore, instead of asking farmers the severity of soil erosion, farmers should be asked to estimate the number of times broken ridges were observed in their fields and the average number of broken ridges recorded per each rainfall event. As shown in Tables 1, severity of soil erosion corresponds with the number of times farmers observed broken ridges after a rainfall event. If farmers indicate that ridges in their fields break more than 15 times during the entire rainy season, we can conclude that serious erosion is occurring on

farmers’ fields and farmers’ practices are not effective against soil erosion. This is applicable to farmers whose fields have slopes less than 10 percent.

Monitoring the frequency and number of broken ridges in a farmer’s field is a simple and a less costly method. Unlike experimental plots, the proposed method cannot give numerical soil loss data. It can, however, give an indication of the effectiveness of farmers’ practices in controlling soil erosion and a general understanding of the relative levels of soil erosion under different conservation practices. Because of its simplicity, conservation specialists can use the method as a conservation planning tool for identifying areas with serious erosion and setting priorities for mitigation measures. To use the method, one needs to understand the underlying assumptions of the method. For example, the method is applicable to areas where farmers use ridges (e.g., contour ridges, tied ridges, or boxed ridges) and knowledge of field slope, crop cover, and rainfall intensity is important when assessing soil erosion occurrences or effectiveness of farmers’ conservation practices. Researchers, conservation experts, extension staff, and farmers can use the method by following these guidelines. Depending upon the availability of rainfall intensity, the following guidelines are recommended.

Knowledge of rainfall intensity is important when assessing soil erosion. Ridging enhances depression storage and reduces runoff and soil erosion, but the effectiveness of ridging depends upon the intensity of the rainfall event. When intense rainstorms occur, the storage capacity of the furrows is filled and water overtops and breaks ridges thus causing soil erosion occurrence and damage. Because soil erosion occurs when ridges break, it is important to link rainfall intensity with the occurrence of broken ridges and soil erosion.

Figure 2 shows the relative frequency of maximum 30-miniute rainfall intensities recorded at Kamundi watershed during the 1997/98 rainy season. A storm with 20 mm/hr intensity was observed about 70% of the time whereas a storm with 80 mm/hr was observed less than 20% of the time (Figure 2). Storms with high intensity are rarely observed while storms with low intensity are frequently observed. In other words, low rainfall intensities have high probability of occurrence while high rainfall intensities have low probability of occurrence. The criterion used for evaluating the effectiveness of farmers’ conservation practices is based on linking the frequency and number of broken ridges to the magnitude of the rainfall intensity that caused the ridges to break. Farmers’ practices are ineffective against soil erosion if intensities shown on the lower end of the x-axis of Figure 2 cause ridge breakage.

When rainfall intensity is known, researchers and conservation specialists can test the effectiveness of farmer’s conservation practices by linking ridge breakage and thus soil erosion occurrence and damage to the magnitude of the rainfall intensity that caused ridges to break. To be effective against soil erosion, ridges on farmers’ fields must not break during less intense rainstorms, but should break only during intense rainstorms. Depending upon the soil, topography, and the rainfall regime of the site, ridges on farmers’ fields must not break when the maximum rainfall intensity is less than 30 mm/hr. For the Kamundi catchment, such a storm corresponds to a 2-year return period. A 10 return period is recommended when testing the effectiveness of mitigation measures on research sites.

Rainfall intensity is not a readily available data and its dependence can make the method difficult to use by farmers and extension officers. To ensure that farmers and extension workers use the method, it is important to develop simplified procedures that can give an approximate assessment when rainfall intensity data is not available. The simplified approach employs the relative frequency of rainfall events and frequency of broken ridges. This can be achieved by counting the number of times that ridges break during the early part of the rainy season and on fields with slopes less than 10 percent. To give adequate number of rainy days, the monitoring technique must be started at planting time and continued throughout the rainy season. For example, a storm with maximum 30-minute intensity of 30 mm/hr was observed 50 percent of the time at Kamundi watershed (Figure 2). If ridges got broken more than 50 percent of the time, it means that low intensity storms are breaking ridges. If that happens, we conclude that farmers’ conservation practices are not effective against soil erosion. The monitoring technique gives good results if it is implemented at planting time and continued throughout the rainy season. It is often difficult to reach a

decision about soil erosion occurrence or effectiveness of farmers’ conservation practices based on limited number of rainy days.

The proposed method for assessing soil erosion is cheap and simple and is suitable for Malawi conditions. Because the method is based on the relationship between soil erosion and ridge breakage, it is not be applicable to areas where ridging is not practiced. The proposed method can also be suitable for soil erosion assessment in other parts of the world where ridging is commonly practiced.

SUMMARY AND CONCLUSIONS

A simple and less costly method that consists of a rapid assessment and a monitoring technique was developed. The method is based on linking ridge breakage with soil erosion occurrence and damage. As the name indicates, the rapid assessment technique is suitable for gathering information by interviewing farmers and collecting reconnaissance information from farmers’ fields whereas the monitoring technique gathers information such as ridge breakage from farmers’ fields and rainfall intensity from nearby rain gauge stations. When time and funding are limited, the rapid assessment technique can be a useful tool for identifying areas where serious soil erosion is occurring. If implemented in conjunction with the Area Sample Frame (ASF), it can provide a means to study the relationship between soil erosion occurrence and damage and crop yield. We also recommend the monitoring technique for widespread use by researchers who would like to evaluate mitigation measures, extension officers who would like to select best management practices, and farmers who would like to improve their soil conservation practices until the frequency and number of broken ridges are reduced and soil erosion is eliminated.

Amphlett, M. B. 1986. Soil Erosion Research Project, Bvumbwe, Hydraulics Research, Report Od78, Wallingford, U.k.

Green, R. I., 1960. Results from runoff and erosion plots from Likuni Agricultural Station, 1956/57 to 1959/60. Report to provincial Agricultural Officer, Lilongwe, Malawi.

Anonymous, 1954-70. Tea Research Foundation of Central Africa. Annual Reports, Mimosa, Mulange.

Chome, V.L.A. 1989. Agroforestry Experience in six years. A Paper presented at the agroforestry National Workshop.

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, Rhodesia. 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.

Kasambara, K.K.K.,1984. Zunde intensive conservation area, Zunde demonstration farm and Zunde runoff plot trials. Paper presented at the Land Husbandry Branch Conference, Mzuzu 4-8 June 1984 (unpublished).

Machira, S. Analysis of soil loss demonstration plots, data at M,mbelwa Farm Institute. A paper presented at the Land Husbandry Branch Conference, Mzuzu 4-8 June 1984 (Unpublished)

Renard, K.G., Foster, G.R, Weesies, G.A., McCool, D.K., and Yoder, D.C. (1993a). Predicting soil erosion by water – A Guide to Conservation Planning with Revised Universal Soil Loss Equation RUSLE " Publication ARS. U.S. Department of Agriculture, Washington, D.C.

Saka, A.R., R.I. Green, and D.H. Ng’ong’ola. 1995. Soil Management in Subsharan Africa. Proposed Soil Management Plan for Malawi.

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.

World Bank. 1992. Malawi Economic Report on Environmental Policy . World Bank, Lilongwe, Malawi.