This study was conducted from 2014 to 2016 (3 years) to evaluate the effectiveness of using the LEAFF tool in the management of BXW and its out scaling would significantly contribute to effective and sustainable adoption of BXW management technologies for improved banana productivity.  The effect of the approach on community awareness, disease incidence, banana yields was evaluated within the participating banana communities of Kiboga and Bushenyi districts.

LEAFF concept and development 

LEAFF is one of the community mobilization approaches for action. It focuses on increasing knowledge and deployment of agro-ecologically sensitive approaches with respect to pest/disease management, nutrient flows and sustainability of the production systems. It fosters better understanding of the variability within and between farms and landscape by farmers so as to match interventions with variability in the production and agro-ecological systems. The approach strengthens skills for observation, analysis and decision making and tap into farmer knowledge and experiences to promote farmer innovation and strengthening horizontal interactions and promote quality engagement within and between partner groups (Figure 1). These were achieved through repeated experimentation, learning and knowledge sharing, then deployment of best-bet practices. The approach is an innovation that was conceived after noticing that many agricultural recommendations to small holder farmers (including disease management) are in most cases made without considering the variability in production systems, agro-ecologies and  socioeconomic  situation  of  different regions. This has had an effect of causing low adoption and discouraging farmers from using indigenous knowledge yet farmers have knowledge of the variability within and between farms and sites with respect to biotic, abiotic and socioeconomic diversity especially the gender attributes which can be exploited to promote farmer-researcher experimentation and learning experiences. This approach was designed to empower farmers to not only be recipients but also originitors of knowledge and technology.

LEAFF is composed of a basic unit called an accountability group (AG), normally of 5-10 farmers who have mutually agreed to set priorities, share plans and lessons to address common objectives on a landscape. The members of AG informally/mutually consent to visit each other’s farms and make observations considering the socio-cultural and gender perspectives. Within the AG, one member assumes responsibility as a contact person for the group and becomes the preliminary link with other AG’s on a landscape and beyond. AG fora are used to identify short and long-term needs and priorities, with respect to technology and/or information while also

capturing feedback. After priority setting by different AGs, experimentation objectives are developed, experiments established, and observations made, recorded, analyzed and interpreted (in participatory manner usually with support of technical persons). Interpretation influences implementation of next agreed actions for next experimentation cycle. In this approach, practices such as integrated pests and disease management (IPDM) only become an integral part of activities but with flexibility for farmers to validate them through experimentation. In addition to experimentation and learning, LEAFF exploits modern communication tools/platforms (including internet and mobile  phones)  to facilitate farmers to access and use improved technologies while enhancing horizontal linkages between farmers and communities across landscapes.

LEAFF is similar to the FFS approach in that both  are group-based adult learning approaches that teaches farmers how to experiment and solve problems independently. LEAFF however attempts to involve farmers in the development of the control measures through farmer action research networks which support experimentation and learning, value chain strengthening, enterprise development and knowledge sharing.  The designs of LEAFF provide for farmers to be exposed to experimentation skills and processes, including constraint identification, prioritization, hypothesis setting and testing, data collection and analysis, participatory monitoring and evaluation and reporting/feedback with emphasis on the specific application to BXW management, while tapping into their existing experiences. This increases farmer-to-farmer interactions to enhance the sharing of experiences and skills in the quest to increase the effectiveness of the control measures. LEAFF farmers are connected through mobile phones and internet thus increase within group and between group interactions which lack within FFS and other community approaches.

Farmers would share notes through visits to each other’s fields and through monthly meetings organized at AG level. The AG contact persons would also collect information via an internet-connected phone (through whatsApp or Facebook) and shares it with other AG contact persons who in turn shared it with members of the AG and other partners in the network. To promote horizontal interactions among farmers, exchange visits to other banana growing areas were organized for group members.

Data analysis

Data from LEAFF and Non-LEAFF farmers were  summarized as means and frequencies/proportions were compared using student t and chi squared tests where applicable. Results were presented in tables and graphs over the years then shared with the communities.

BXW awareness levels and sources of information

At project initiation in 2014, it was established that most of the farmers (91%) were aware of BXW. This high level of awareness was found to have been majorly achieved through radio messages (33%), trainings attended (33%)  and personal observation in their fields (23%) (Table 1).

Changes in BXW incidence and prevalence among LEAFF and non-LEAFF farmers

There was a general observable rduction in  the  number of infected plants per farm over the three year study period (Table 3) although by close of 2016, there were still recordings of presence of infected plants on farmer field among both  the LEAFF and non LEAFF participating farmers (Figure 2). Further analysis showed a significant difference in the percentage of farms with BXW among the LEAFF and non LEAFF participating farmers in 2015 (Table 4), where 31% of the LEAFF participating farmers and 57% of the non LEAFF participating farmers still had BXW on their farms.

Further analysis also showed significant degrees of success in preventing outbreak of BXW among the farms that did not have BXW among the LEAFF compared to non LEAFF participating farmers in 2015 (Table 5). In at least 6 months, 86% of the LEAFF participating farmers and 64% of the non LEAFF participating farmers had not registered any BXW outbreak on their farms suggesting successful control.

Farmer rating of effectiveness of various BXW control options as judged from their experimentation trials also were consistent with above changes in incidence and prevalence (Table 2). They highly rated the effectiveness of avoidance of corm removal (100%), avoidance of pruning (100%), cutting off only dry parts (100%), cutting and burying (79%), debudding (73%) and disinfection of tools in Jik or sodium hypochloride (78%) in reducing the prevalence of BXW on their farms. They however rated having many tools and restricting them for use in specific areas of plantation (33%), avoidance of dumping rubbish in the farm (50%), removal of diseased part only (59%), pouring a liquid mixture of urine, ash and hot pepper on the cut suckers (67%), restricting animals (67%) and avoidance of digging with a hoe in the banana plantation (67%) as not effective in reducing the prevalence of BXW on their farms.

BXW awareness levels and sources of information

The observed high level of BXW awareness is consistent with other studies dating back in  2006  in  Uganda  which showed that by that time, over 85% of farmers had been aware of BXW symptoms and available control options (Bagamba et al., 2006; Tushemereirwe et al., 2006) although they had not been implementing them. This observation further highlighted the need for a more robust approach to cause implementation of the control options. It therefore follows that the observed incidence and prevalence of BXW at the beginning of this study in 2014 cannot be attributed to lack of awareness but due to other factors. It had been noted earlier that many farmers did not appreciate the rationale for these control options. This justified the need for farmers to validate these control option together  with  their  other  related  innovations  for  BXW management.

Changes in BXW incidence and prevalence

It was noted from the farmer managed control practices validation experiments that while some LEAFF participating farmers reported a prevention of outbreak and reduction of BXW prevalence on their farms, others also reported that some practices such as simply having many tools and restricting them for use in specific areas of plantation, avoiding dumping rubbish in the farm, removing diseased/symptomatic part only and pouring a liquid mixture of urine, ash and powdered hot pepper on the cut suckers were not very effective in disease reductions. Pruning with contaminated tools is an established transmission mode (Buregyeya et al., 2010) and while farmers had imagined that they could control BXW by restricting specific tools to specific areas of the plantation, they discovered that this was not practical and in some way they only caused further disease spread within field thus was not effective as they eventually rightly observed.

The low farmer rating of the effectiveness of avoiding dumping of rubbish in farm is expected especially that this is only a means of preventing disease introduction into the field but would not in any way prevent disease spread within the field from the already infected plants. Also, the low farmer rating on the practice of removal of only symptomatic part is expected especially that Xcm infection is systemic (Ssekiwoko et al., 2010; Ocimati et al., 2013a; Ocimati et al., 2013b). By simply removing only symptomatic  part  would  not  render  other  deeper parts where Xcm has migrated free of the pathogen and it would only be a matter of time when symptoms appear in another adjacent part.

The observed reduction in the number of infected plants per farm and the total number of infected plants over the landscape over the years was attributed to active destruction of infected plants. Further, the persistent presence of infected plants on their farms even by 2016, despite all efforts, was attributed to the laxity on the part of farmer who had actually reduced on destruction efforts as shown in Figure 2 and this further justifies a need for a continuous and more sustainable engagement of farmers.

Nevertheless, application of LEAFF caused a reduction in prevalence of BXW from 57% to 30% and subsequent eradication of the same in up to 82% of the farms that once had the disease compared to 64% where such approach was not applied. The relatively high level of disease eradiacation among non LEAFF farmers was attributed to spillover effects. In addition, it should be noted that other previously promoted approaches were still active in these communities and this result only illustrates the superiority of LEAFF over such other approaches such as Community sensitization for action which is largely universally applied in all banana growing areas of Uganda. LEAFF performance is however comparable to Participatory Development Communication approach (PDC) that caused a drop in BXW prevalence from 88 to 18% (Kubiriba and Tushemereirwe, 2014), Community action which caused a drop in BXW prevalence to 68% (Kubiriba et al., 2012) and FFS approach which caused a drop in BXW prevalence by 43% compared to 15%  with  other  approaches  before  it  (Kubiriba and Tushemereirwe, 2014).

Changes in banana production and productivity

Observed initial values in area under banana and the yields suggested that banana production was possibly equally important among the LEAFF and non LEAFF participating farmers. This is further supported by the observed lack of significant variability in Matooke acreage among LEAFF and non LEAFF participating farmers. Further, maintenance of Matooke acreages over the years seems to suggest that regardless of presence of BXW and implementation of LEAFF approach, households were already operating at optimal acreages and that there had been sufficient efforts to manage effects of BXW in Matooke fields. In addition, field devastation by BXW to cause reduction in acreage is usually less pronounced in Matooke fields. This is clearly evident in our results that showed reduction in Kayinja acreage among the Non LEAFF participating farmers. The observed increase in Kayinja acreage among LEAFF participating farmers further emphasizes the significance of LEAFF approach. Further, the observed higher Matooke yields also emphasize the significance of LEAFF approach. It is however not very clear why there was a greater banana yield recovery among the non LEAFF participating farmers than among LEAFF participating farmers, but since there had been an increase in banana acreage among the LEAFF participating farmers; it is most likely that most plants at the time of assessment had not yet bore fruits to contribute to yield recovery as they are likely to have  just been planted. Note that it is more common to continuously plant matooke unlike for Kayinja. In addition, the banana acreage was generally lower among the LEAFF compared to the non LEAFF participating farmers.

The observation that reduction in BXW prevalence on farms did not necessarily correspond with increase in yields together with  a general drop in the number of bunches harvested and sold (Figure 3) among the  LEAFF participating farmers is attributed to the time requirement for ecosystem recovery which this approach also emphasizes. Disease infection together with roguing reduces the plant cover and yields directly but they also expose the soil to degradation resulting in reduction in bunch weights. But towards 2016, it was noted that bunch weights and general number of fingers were beginning to recover and improve (Figure 4).

LEAFF has been found to be a powerfull approach in mobilizing for community action against BXW where when applied in the study areas it caused a reduction in disease  prevalence  from  57  to  30%  and   subsequent eradication of the same in up to 82% of the farms that once had the disease compared to 64% where such approach was not applied. Given this approach’s superiority over others previously used in East and central Africa, it is recommended for  scalling out to other areas within this region.

While LEAFF has been proved effective in increasing farmers adoption of options for BXW management and consequently lead to increased yields, it lacks provisions for handling the increased produce which would negatively affect their sustainable use. More over all the participatory approaches so far used for BXW management, focused on the need to control this diaese as the main entry point. Indeed the main challenge was BXW but for the implementers, it was assumed that BXW challenge and the need to controll it per se would its self  act as an incentive for farmers to invest time to learn, share and inovate for sustainable control. There is need to consider a new perspective where farming households today are nolonger focusing on food sufficiency only but use their farms as a source of income and will invest time if there is a threat to their income or there is a potential to improve their income if they engage. Otherwise they will make BXW control secondary and rather focus on other income generating activities only returning to BXW control activities during their free time. Each farming household is a trading entity  of some sort and trade strengthening to maximise profits from their agricultural activities should be the driver/ incentive for action. Research therefore needs to explore the incentives and benefits/success that would come along when the above participatory aproaches are organised around whole value chains, intergrating marketing and strengthening agricultural trading for better incomes. It is envisaged that through such, other practices such as environmental management, ecosystem sustainability and management of epidemics such as BXW should be an intergral part to protect and improve their incomes and in this way, it shall be more sustainable.

The authors have not declared any conflict of interests.