«JM Thamaga-Chitja March 2008 Submitted in fulfilment of the requirements for degree: Doctor of Philosophy (Food Security), African Centre for Food ...»
The screen of the user interface had four sections. The user needs to provide data for three areas to receive input. They are climatic data (annual rainfall in mm); area mean temperature (degree Celsius); annual number of rainfall days and the length of the photoperiod (hours); conducted soil analysis (yes or no); and, manure quantity (number of animals or quantity of manure if it is known or quantity of compost if it is known). The decision support tool can either predict the quantity of manure based on the number of animals for a specified list of crops or a list of specified crops based on the quantity of manure or compost provided. These two outputs cannot be produced 74 simultaneously. This allows one to be sure of which nutrient source the output is based on. In cases where a user desires output based on number of animals, manure and or compost, the output will have to be generated separately. The interface also lists additional factors that are important to consider in organic production including which are: knowledge, skills and literacy (Sligh & Christman, 2007) and an enabling policy environment (Scialabba, 2007). The intention of including these additional risk factors in the user-interface is to inform the decision-maker that it is not only the production elements of organic farming that are important but that other factors have a serious role to play. No output can be generated for these additional factors.
Model outputs are displayed to the right of the user-interface once the required data is filled in (Figures 5.7 and 5.8). The output lists crops that can grow; the number of wheelbarrows (loads* i.e. number of loads of wheelbarrows of manure) required to provide the required amounts of nutrient for each crop; monthly moisture levels and important diseases that are triggered by high moisture presence; and, for disease triggered by low moisture presence.
The use of the tool is simple because once the required information is entered into the Excel spreadsheet, output can be received at the push of the ‘enter’ button. Ideally, the outputs should be displayed on separate screens once the desired input user interface has been loaded. However, that will require extensive computer programming specialist input and time (Voges, 2006). Nevertheless, this step can be carried out in the future. The interface is simple to use and can be used by any literate person, provided they have the required information as specified above. Furthermore, the tool may appear complex to illiterate farmers, but given training it is fairly simple to use for an extension officer or anyone in an advisory role to farmers.
77 In summary, the model provided three desired outputs, these being a list of suitable crops in the chosen areas, an answer as to whether the suitable crops can be grown organically in the chosen areas and possible diseases associated with each crop. Each stage of the model aimed to solve a sub-problem of this study and was described and critiqued. Several assumptions relating to each sub-problem were made. In the first sub-problem a decision was made that single values, instead of ranges, would be used when capturing plant growth data. Absolute values relating to rainfall were used because many smallholder farmers practice rain-fed agriculture and are found in low rainfall areas. Mean values for temperature were used due to the variation in the nature of temperature. On the other hand, it is accepted that the use of mean that certain crop growth will occur values that fall below the mean.
Crop nutrient requirements calculations were based on the most limiting nutrient between N, P and K. It is also accepted that some yield (although below optimum) can be achieved. Other assumptions were related to the use of rainfall as a correlate of moisture for both crop growth and disease onset. Nevertheless, it is accepted that other sources of moisture (mist and humidity) and environmental factors, such as slope, can play a role in moisture levels that can influence disease onset.
Large numbers of wheelbarrows indicated by the model show that organic production based on manure does not lead to optimum yields. The challenge for smallholder farmers is the large number of livestock that is required to produce the manure which is not likely to be possible for poor farmers. Even those with livestock will require unrealistically large amount of manure. This may not be sustainable if livestock numbers drop. Other technologies such as composting and the use of EM are important to consider.
Risk was based on moisture and disease occurrence. Risk is expected to be lower in winter but higher in summer which poses a problem for rain-dependent farmers, therefore the need for supplemented irrigation is heightened. Furthermore, rain is not the only contributor of moisture, but that mist and humidity can also play a role in disease development.
6.1 Introduction In conventional research people are the subjects of research (Tilakaratna, 1990). In this study people were active participants in the collection and processing of the data.
The creation and ownership of knowledge and technology in participatory research is aimed at those who are the ultimate beneficiaries of that knowledge (de Vos, 1998).
Advances in technology related to smallholder farmers are seen as a solution to many of their problems (Duram, 1999; Freyer et al, 1994). However, technology should be appropriate for the environmental, cultural and economic situation it is intended for (NCAT, 2007). Creation of brilliant technologies, without consideration of appropriateness to their beneficiaries, is futile. Participation of beneficiaries in generation of knowledge and/or technology cannot be over emphasised.
One of the constraints in smallholder farming is access to appropriate information (Stephano et al, 2005). Farmers require information to make sound decisions related to production and other areas of farming. Factors such as form, medium of delivery, language and literacy play a role in access to information (Stefano, 2004). Therefore, development of technologies that focus on overcoming these constraints are important.
Previous chapters relating to the model addressed the need for a decision support tool and the development of the decision support tool to address identified production constraints. The purpose of this chapter is to present an application perspective of the decision support tool that was created in partnership with farmers. Actual answers on what crops can be grown organically in the three study areas or whether the farmers agree that the selected crops can be grown organically or not in their areas, are provided for in each case, using a comparative approach. A farmer’s critique of the tool is also presented.
Agro-ecological conditions for each area (being annual rainfall in mm); length of rainy season (days/annum); mean annual temperature (minimum and maximum) and the photoperiod) were entered into the decision support tool (user-interface) to produce output printouts. A stepwise comparison and analysis of the three outputs was conducted, based on each sub-problem.
6.3.1 What crops can be grown organically by the participating farmers?
As explained in Chapter 5, the list of suitable crops resulted from matching the lists of sought-after organic crops from Woolworths (Ferreira, 2004) and agronomic data from the Bioresource Database from the Department of Agriculture and Environment in KwaZulu-Natal (Camp, 1999). Vegetables, fruit and herbs were included in the crop list. The list was used as a base for generating the model outputs. Table 6.1 shows a comparative list of crops per area from model outputs. The list in Table 6.1 was used as a base to match the agro-ecological data (mean annual rainfall, mean area temperature, the photoperiod and the length of the rainy season) supplied in the user interface for the crop requirements of each plant for adequate plant growth. The crops that the model deemed suitable for each area appear in the two-page printouts, per area in Appendix D (Mbumbulu), E (Muden) and F (Centocow).
It must be emphasised that instead of using optimal plant growth conditions, absolute plant growth conditions and means were chosen in the development of the model.
This reasoning was based on the fact that, due to historical reasons, many smallholder farmers are located in agro-ecologically inferior parts of South Africa (Aliber et al, 2006). In addition, means were used for temperature values due to the naturally large variance in temperature across days and seasons. Due to the use of absolute growth conditions, it was expected that lower yield scenarios would be presented by the decision support tool as opposed to higher yield scenarios, if optimal conditions were used. Nevertheless, the modest yield scenario presented is more likely to be
Basil Coriander Note: Where means the crop grows and – means the crop does not grow in the area.
81 According to the model, Mbumbulu’s climatic conditions meet almost all growth requirements for the crops, except for those of amadumbe (taro). Amadumbe was rejected by the model for all three areas due to its high rainfall requirement (FAO, 2003). On the contrary, amadumbe is a popular crop in the area and it is widely grown for consumption and for commercial purposes. A closer look at the reason for this outcome revealed, according to the model, Mbumbulu’s mean annual rainfall was 956mm, as indicated by DAEA’s Bioresources Database (Camp, 1999). On the other hand, the minimum water requirement for amadumbe according to FAO 2003, is 1000mm (FAO, 2003). A shortfall of only 44mm has resulted in the model indicating amadumbe to be unsuitable in Mbumbulu. Due to the shortfall being small, it is reasonable to conclude that amadumbe is suitable for growing in Mbumbulu.
However, this small shortfall also highlights the fact that without supplementary irrigation, Mbumbulu farmers face a significant risk if the rains do not come as expected. Furthermore, the model itself required a wider range of data to be more inclusive to avoid crops being rejected on small margins.
Four crops in Muden (including amadumbe, lemon, peach and mint) were deemed unsuitable for organic production for growth by the model. Only amadumbe and peach were rejected by the model for the Centocow area. Peach was also rejected by the model for both the Muden and Centocow areas due to their relatively short rainy season (Muden 181 days and Centocow 211 days), thereby not meeting the minimum requirement of 240 days for peach’s growth cycle (FAO, 2003). Lemon was rejected by the model on the grounds of the rainy period being too short to fulfil adequate growth in Muden. Fruit trees have a longer growth cycle (and constantly need available water) and take time to bear fruit. Mint was deemed unsuitable for growth in Muden due to shortfall in annual rainfall need (FAO, 2003). It is clear that Muden, compared to Mbumbulu and Centocow, is agro-ecologically less supportive of rainfed smallholder agriculture. Additional irrigation or improvements and the introduction of water harvesting technologies should have a positive impact on crop performance, provided all other important elements are met.
Table 6.2 shows the crops currently grown by the three groups.
There is a discernible difference between what the farmers are currently growing and what is suitable for growth according to the model. It is to be noted that the Mbumbulu farmers (EFO) 82 focus on only three root crops, potatoes, sweetpotatoes and amadumbe and green beans (not root crop and new crop), when the area in fact has the potential to support 19 other crops. The Mbumbulu farmers currently have no supplementary irrigation, which may explain why their farming does not include leafy vegetables, as these crops would require supplemented irrigation. The Mbumbulu farmers stated that they would like to include more vegetables in their production but they were limited by many factors, including organic pest control and a lack of new markets. The Muden farmers’ production focus is currently on vegetables and garlic. Interestingly, garlic is their main cash earner. Although the Muden farmers have supplemented irrigation, they have expressed frustration regarding the inefficiency of the irrigation system (Goba, 2004; Mthembu, 2005). The Centocow farmers focus mainly on three crops despite the potential for many other crops; the farmers revealed that other vegetables were grown in home gardens but not consistently. The farmers expressed a desire to grow more maize and vegetables, such as cabbage, but they are limited by a lack of water and fertiliser.
Note: Where means the crop grows and – means the crop does not grow in the area.
83 The second part of solving sub-problem one was to ascertain what the organic production requirement was for suitable crops and whether farmers in these areas can meet these requirements. According to van Averbeke & Yogananth (2003) it is common knowledge that small-scale farmers in rural areas of South Africa use livestock manure for soil nourishment. Key nutrients for plant growth required in larger quantities for most plants are N, P and K. Manure provides these minerals but only in small quantities. It is important to note that manure’s nutrients are slowly released and subsequent crops would benefit from previous applications. The farmers included manure in varying qualities in their soil nourishment programmes. The Mbumbulu farmers used pen manure exclusively due to certified organic requirements. The Muden and Centocow farmers used manure in conjunction with commercial fertilisers. Both Muden and Centocow farmers expressed a wish to be certified as organic producers but are faced with many constrains with regards to meeting the requirements for organic farming certification. Unless these constraints are addressed, Muden and Centocow do not meet organic certification.
When calculating the amount of manure required for the soil to improve from being nutrient depleted, the nutrient removal rates per crop were used as detailed in Chapter