From africa-rising-wiki
Jump to: navigation, search

Africa RISING ESA Project Review and Planning Meeting
10 - 11 September 2019
Dar es Salaam, Tanzania
[edit | edit source]


  • Review of main research and development achievements and challenges of the project.
  • Plan activities for 2019/20 season, with a special focus on farming systems research, collection of data on SI indicators and documentation of project outputs.


Day 1 [10 Sept.]
08:00 Registration
08:30 Introduction of participants
09:00 Welcome & Opening remarks
  • V. Manyong – Director, IITA Eastern Africa Hub
  • G. Mkamilo – TARI Director General/Representative
09:20 Overview of agenda for the day & housekeeping
09:30 | Project implementation updates & developments – B. Mateete {15’ pres. + 15’disc.}
10:00 Break & participants group photo
10:30 Review presentations {20’ pres. + 15’ disc. for each}
13:00 Lunch
14:00 Review presentations cont’d.
15:15 Break
15:45 Bus stop process for review of 3 posters for cross-cutting {25’ each stop X 3 rotations}
17:00 25/10 crowdsourcing to reflect & learn from what is observed during each of the review presentations
17:40 End of day 1

Day 2 [11 Sept.]
08:30 Briefing back on 25/10 crowdsourcing exercise from previous day
08:45 Communications and knowledge management – J. Odhong/E. Massam
09:15 M&E needs at IITA – S. Adebayo {20’ pres. + 10’ disc.}
09:45 Draft 2019/20 plans for the cross-cutting {10’ pres. + 10’ disc.}
10:30 Break
10:50 draft cross-cutting plans cont’d
11:30 Aligning to project systems approach - B.Mateete
11:35 Break out groups for workplan development (per listed groupings)
  • ISFM - based systems (TZ)
  • Land management (TZ)
  • Maize-grain legume intensification system (MW)
  • Land management under CA (MW)
  • Animal production – based systems (TZ/MW)
Farming Systems Specialist, Gender, GIS,Econs. M&E, Comms. to integrate & rotate within these groups
13:00 Lunch
14:00 Break out groups cont’d
15:00 Presentation & feedback on draft group work plans {10’ pres. + 10’ feedback per group}
  • Working break (tbd)
16:40 Next steps
17:00 Closing remarks - M. Bekunda
17:10 End of review & planning
19:00 Closing cocktail

Day 3 [12 Sept.]
08:30 – 11:30 Project Steering Committee Meeting
{Agenda for SC meeting sent to members separately}


Welcome & Opening remarks Victor Manyong, Africa RISING ESA PSC Chair & Director IITA Eastern Africa Hub

  • Welcomed the partners and project team as the Chair of the steering committee
  • The project has been successful
  • Partners and other stakeholders including donors are impressed with the project results and admirable team efforts.
  • Welcomed the new partners and stakeholders and encouraged them to contribute to the project with zeal.
  • There is a need to show evidence of the project achievements in phase II.
  • The team shouldn't relax even if everything is going well.
  • There is a need to start thinking about how to piston this project for what is to come next after the second phase.
  • Project partners to be prepared for the internally commissioned external review in early 2020. He explained that it was important for the team to work and prepare some important documents which will support the review team
  • Partners should prepare to showscase the evidence of achievements and some changes that has been taking place so far
  • The team needs to be conscious about the funding issues.
  • Legacy documents for the project like the handbook are vital for people who will want to later refer to the projects achievements.
  • Project partners to continue working closely with the Chief Scientist. Great collaboration will only serve to enhance project performance.
  • Colleagues from all Africa RISING program locations attending this meeting are very welcome. We hope to interact with you and gain some insights from you all as well.

Everest Makene, TARI (re-presenting TARI Director General Dr. Mkamilo)

  • The multi-disciplinary integration witness within the Africa RISING project is very commendable.
  • Having the economics, systems and gender aspects of technologies in addition to biophysical data is a feature that is unique only to Africa RISING.
  • Keep up the scientific rigor within the project even as the team also aims for other development goals.

| Project implementation updates & developments – B. Mateete
Comment:There is a challenge noted with the theory of change for phase II. It was assumed that development partners have the common vision with Africa RISING and therefore this would lead to a lot of mutually beneficial collaborations. However, this is not the case, considering the interactions with a majority of development partners so far. Some of them actually expect Africa RISING to provide them with funds for scaling.
Comment:The funding uncertainty situation affects the extent on what can be delivered, and it’s a part of challenges affecting project performance. It is important that this is noted as one of the major reasons for the inability of the team to deliver on all things that were promised in the proposals.
Question: how many chapters are still pending for the handbook to be complete?

Response:The plan was to have ten chapters, but we have 1.5 that aren't complete.

| ISFM - based systems (TZ) – J. Kihara

Question: Why did you not consider averaging the productivity domain, even though there will be some variability between the systems? The data presented (with and without the ISFM-maize/ vegetables) do not seem to have much differences?

Response: It is the first time we are advancing the ISFM tool at this level. There was assessment of the system on the shifts so far, however, due to some circumstance, it appears that there was no significant shift regardless of the environment we still have the same results. This is something that we consider as we move forward.

Question: The spider diagram presented shows that the systems has no gross margin (and some of the domain have low value), does it mean that there is no gain for farmers? If yes, then should farmers practice the technologies?

Response: It probably on the issues of scaling, for example if say in one case we get $2500 of income and the other case you are making $100 or $200, once we do the scaling then value becomes low.

Question: From the radar chart, what was the bench for the yield variable? Because even with-ISFM, the values that were presented were lower than the benchmark?

Response: We consider the maximum that is observed and use the maximum as our benchmark. For the productivity domain, we considered the maximum productivity that was observed across the population in the dataset, that may be far from the mean, since there will be huge difference from the mean, then the value for the ISFM become low.
Response: It is the question of the scaling, and method that we used, we considered average value that we are averaging and then we identify the maximum from across the average group. The maximum is not the average rather than a value that is achieved by one person across the population and is considered as the achievable and can be a model.

Question: With the ISFM, data on yield reflects low value, what else is missing at least to push the yield to some reasonable level?

Response: As a team we are working to brainstorm on the kind of intervention(s) and act to reach the reasonable yield level.

Question: The difference between with and without the ISFM, nutrition outcomes will be three to four times better or higher than the origin, is that really the case, or something that shall be expected?


  • The project team should consider removing some noise (for radar diagram) and bring the meaningful and clear radar graph for example most of the variables has a lot of external factors that cannot be controlled, the team should be careful with variables like stunting , wasting and underweight because it create much noise, if are removed then there would be more meaningful radar diagram.
  • The outcomes from the presentation (on with and without the ISFM) should focus on the direct impact of the interventions.
  • The approach used as outcome measure for the nutrition needs to be changed as opposed to standard underweight wastage, the team should consider the diet quality
  • The presentation appears to have data quality issues. For-example there was a huge difference between the benchmark data without and with ISFM data as presented in radar chart, the team should have identified what needs to be addressed and do it better.
  • The expectations from audience would be to see there are more crossovers between the tradeoffs because of benefits and tradeoffs in the farm and cropping systems, the team should have considered overall the factors holistically. For example, if there were comparisons between the two systems with the fertilizers and without the fertilizer, yield will have increased on the fertilizer systems, on the other hand there should be effect on the gross margin, this effects on the cost of buying the fertilizers will increase, but such data were not presented in the radar graph.
  • The team should have regard combining the two variables (runoff avoided, and soil loss avoided) for the environmental domain as presented in technology benefits with-without ISFM. The two variables are both addressing soil water conservation.

| Maize-grain legume intensification system (MW) – R. Chikowo

Question: You’ve said that you have SI technologies that work but looking at the spider diagram presented you haven’t collected much data on some of the domains. I have seen for example on one of the technologies you said you haven’t included labour analysis. You also haven’t included/showed integration with nutrition.

Response: Maybe it wasn’t clear from the way I presented it. Somehow the spider diagram I presented didn’t do justice to the 5 domains. However, this was implied in one of the slides I presented. We are looking at the SI domains as the templates for validating our technologies. I accept however that this was an omission on our parts. When I talked about labour as a factor, this was in reference the CA where there are tradeoffs with labour going into ridging. However, those are high level systems research questions when you look at some of these technologies. The past 2 weeks I was at MSU in the USA training partners about application of SIAF…So we are working on this.

Question: What is the space between the ridges (particularly for groundnut planted in double rows)?

Response: This is a function of agroecology and what are the stressors. The architecture of the variety also plays a key role. We therefore want farmers to also continue being part of evaluating the product. While we gather data on yield, farmers can also tell us what the other effects like diseases and pests are. We also know that there is a problem with single row spacing, where you kind of limit yourself too early. For ground system we are not using boundary ridges because we plant the groundnuts on the flat surface. The spacing for the Malawi context is 37.5 row spacing and planted on the flat. By doing that we are able to double the yield, just by better plant arrangement. We are planting of the flat surface and by doing that, we can double the yield on the single row planting considering the better planting arrangements.

Question: What is the optimum plant population when the crops are planted in double or single rows?

Response: There is the problem of the single row spacing, it has limitations particularly on sub-optimum population, plants yields responds to certain biological coefficient.

Question: Considering the land terrain where the slope is 5-7%, do you expect CA to perform well in that condition without tied ridging? You should expect run-off in that kind of terrain.

Response: It is not necessary to have the good physical structure to prevent erosions rather than when we apply good agronomy practices such as the plant arrangement. This increases infiltration and less erosion. In our program we have an area of about 15%-20% slope in a mountainous area in Malawi, we have ridge systems and have them planted , there are some vegetable barriers, during the heavy rainfalls the ridges systems always collapse on the other side we have crops planted on flat and are evenly distributed CA systems, they have higher infiltration that why they are barely to erosion and run-off. So, for us in this environment it is not necessary to have a physical barrier to prevent erosion, but when you do good agronomy it helps.There is a physical challenge in Malawi in terms of cropping. The ridges in Malawi were designed to suit maize - which is the main crop. So, we are not changing the 75 x 25 cm ridging doesn’t suit legumes. We therefore try to adapt in this way.

Question: What lessons have you learnt from the regional interventions such as SIMLESA to enhance the technologies you have presented?

Response: We are not starting from scratch, but through Christian’s activities in southern Malawi we are building on certain aspects of things that had been done in SIMLESA.

Question: Please comment on how your work is integrated with other teams for cross-site and regional integration?

Response: Of late we have been having more of these meetings (cross-country learning) and an outcome of those has been transfer of some of the technologies like doubled-up legume to Zambia and Tanzania.

Response: For example, with IFRI we have been working to collect results and analysis (research out outputs) conducted from the targeted project areas. The results have been combined with the GIS, remote sensing, and some economic modeling. This has enabled us to identify the recommendation domains and quantify the ex-ante impact of doing some research activities in some other areas, we have developed a research paper for Zambia and currently, we are working on one for Malawi. So we can now identify recommendation domains and also quantify the ex-ante effects of introducing some of these technologies to larger domains and environments and farmers beyond the areas where we work. These research outputs and the technologies are not just localized in plot level neither a specific community.

Question: How do you calculate the calories and the stability (ranks)? From what you’ve presented it looks to me like calories and maize yield are strictly related so you could just present one of them. Because this is not just calorie intake from an individual, but rather from the farm?

Response: Yes, they are very linked, but some slight difference based on how much legume component. This data analysis is not one year, but several years and sequenced planting. So, you tend to look at it more longer-term effects on a system. This is because what shows between one year and the next could be very variable therefore people can even conclude that this systems works for calorie production but not cropping etc. This can then later explain adoption.


  • I am disappointed. I told presenters to identify the gaps missing, but I didn’t see that in your presentation.
  • There was a gap on the integration of the nutrition component.

| Animal production – based systems (TZ) – B. Lukuyu

Question: In relation to the radar diagram, across the presentations there is a frequent mention of the potential to scale the intervention on the phase of Africa RISING, an important stepping stone to scaling is the engagement of agri-business in the provision of inputs and service, ensuring they are able and profitable to do so, or where farmers perhaps could sell their product profitably, however, the spider diagram presented did not indicate a mention about agri-business linkage and other opportunities to take the technologies to scale, It is my suggestions that it should be presented (on a table or spider diagram)?

Response: You are right Amos, one of the things we haven’t done very well in Africa RISING is establishing linkages with private sector. However, there are things you can approach private sector about and they will be very enthusiastic, but on others they may not be. Forage seed is an informal seed sectors, and private sectors has kept away from dealing in forage seeds that are OPV, because they are meant to be produced and circulated to the community, then there is no business opportunities for informal forage seed. This has made it difficult for the team to work with the private sector. Study recently conducted in Uganda, reveals government parastatal, research institutions and the community base organization plays the biggest role in production and distribution of forage seeds than the private sectors, besides they play the crucial role to support the accessibility of the forage seeds.

Question: Forages has clear advantage in term of productivity, to what extent you have put in place measures to enhance access to germplasm?

Response: We established bulking points for the different Napier grass varieties in different communities. These have proved to be a good source for planting materials. There are problems however with them, farmers come and pick varieties at will and select the varieties based on what they like, but the materials is accessible to farmers.

Question: About scaling of feed troughs from Ethiopia, before scaling, have you considered their fitness for the Tanzania condition given the possible social and culture differences?

Response: The first activity will be study to understand what local farmers feels about, sharing feed traps, and have their inputs. It will be adapting the feed tracks through participatory process with the farmers and allowing their inputs into it and adjust it if there is a need. We will propose a system of introducing them and scaling out with partners.


  • For scaling out the technologies, we will need to go beyond just writing the MOU’s. There is a need to think about the environment that the MOU could work and succeed for certain specific technologies. For-example for technologies that do not necessary give quick return on investments and yet they have other benefits, do not necessary attract the private sector partners.
  • I don’t think that it is appropriate to always a must that we test a technology that has been validated elsewhere (like the feed troughs from Ethiopia). I therefore suggest that we take it to the farmers and then monitor how they are adapting it.
  • We need to revisit our theory of change and identify where there are gaps to making it a reality and fill up the missing data. We may be doing great in scientific rigor, but how well are we doing when we review what we are achieving to get us to the end point of the theory of change.
  • From your presentation its quite clear that you do not have a lot of data in some themes, while not so much on others. So, what are we/you going to do about that so that next year for those that are missing you can have them? This kind of approach makes more sense that continuing to gather data on areas that you clearly already have enough. This is the most logical approach if you are going to be able to gather data on all the SI dimensions.
  • One of the points of the radar graphs is to compare two or more systems. In yours you don’t have these comparisons. You just only showed the data on the system you have on crop residues and livestock feeds etc. – what are you comparing with?
  • The presentations were missing data that link back to the log frame. There should be a balance to ensure that SIAF and the logframe is balanced.
  • If the scientists were not trying to address systems within their presentations, then you would see presentations that refer to data for the logframe. So it is because of this focus on systems that you aren’t able to see the data that directly responds to the logframe.

| Farm level systems case studies (TZ) – L. Claessens

Question: What do you mean by data collection tools design for the three farms, why are the data collection tools not generalized to all the three farmers (from the selected case studies) and what was the rational for selecting the three specific case studies?

Response: The farmers were selected because different disciplines were conducting activities on the selected farms. It is sort of validation on what is happening (if different disciplines are conducted in one farm, the question to address will be how these different disciplines are contributing to the totality of those farms). They provide a good opportunity for looking at the tradeoffs.

The tool that was developed was to get information from the disciplinary scientist working with individual farmers. This tool allows the scientist to provide data, these data will be used to perform the analysis.

Question: Why do you use random maximum number, while there are potential attainable, for example from the data we collected e.g. the most efficiency guy, we could benchmark to the most efficient guy instead of many picking numbers

Response: For productivity (on data received form the one farm) I would not like to use the random maximum number, I would also not know how to get it from the survey? For example, how far do we take agroecology, do we take the maximum of whole base line survey as the benchmark for these individual farms?

Response: Maybe you could consider conducting a frontier analysis?

Question: Why are you not using data from the Tanzania baseline survey conducted in 2015 (for Tanzania)? It has a lot of information for all the five SI domains. Because what you presented is focusing on the farm scale outcomes or variables, and we do have farm scale data form the baseline survey, and it has richer information and had included at least hundreds of farmers for Africa RISING and control farmers. The data is available and free, and it is Africa RISING free public good and it will be much more solid because it included other farmers instead of only three farms or one. These information’s are available, there are social, human. At the farm level, we have diversity of data

Response: If there is a baseline data from selected three farmers, it is fine. however, if the disciplinary scientist execute the actual data collection from these three farmers, we feel that they would have more accurate data compare to the available baseline data. That is why we chose the three farms, this approach will allow us to conduct the complete validation on what may be happening.

Question: We cannot treat all farms with this level of intensity! so is this realistic if we scale-up? Are these 3 farmers representative?

Response: The case studies (three individual studies) is not for scaling. The question could be what made the farmers (Maile, Monica and Lukumai) willing to embark on a substantial process of change on their farms, what made them ready? However, the data cannot be generalized, but it is an exploration into that as well. Not all studies have to be representative, as for social science you can make a small case study then we move to the other level.


  • It could be good to have a table to document the extracted information for all efficiency yields data, economics and get efficiency cases. The table can be used as the benchmark towards what needs to be worked upon. Other information’s to be included (in the table), could be the least efficiency cases. The table will help to clarify the range in which the technologies should be, also easy to identify the gaps in terms of efficacy of the yields. For example, where there are reductions and recommends, we could be able to apply inputs hence attain the highest efficiency case.
  • Instead of generating new set of data, it would be better to get data from the published literature (s) over the past seven years, by doing so It would save the costs (time and finance).
  • The approach selected (case studies) should have consider effectiveness in term of scaling the technologies.
  • Regarding the comment on the (systems, SIAF) data should now be part of the workplans, consideration should be on profiling each of the farmers who hosts the technologies at the farm scale, this influence more of realistic data instead of extrapolating from the plot level.
  • It was recommended that the team should look on the trials and find the potentials within the agroecology and find out what can be potentially produced, realistically and proceed to the farms scale level (consider the farms proportion relating to the technologies). The economics out of this approach will be more realistic while the obtained information(s) could narrow the questions of why farmers are not adopting the technologies. When it is at the farm scale level, the results might have been different
  • This is exactly on what we need to address we are tracking the missing information from the three case studies at individual farms scale selected and profiling these farmers. For now, we just have data from the plot level. We are yet to bring that to the farm level along with the system level (that Job is working on) with the individual farmer.

| Farm level systems case studies (MW) – M. Mutenje

Question: From the analytical framework and results presented, particularly for maize crop, the indicator for nutrition shows that your target is to conduct only the grain analysis, is that adequate or you need to complement it with consumption?

Response: We are also performing the full dietary diversity study and are looking at the SI interventions aiming at reducing Aflatoxin contamination for maize and groundnuts.

Question: What Regis presented were data from the same case study? How does this (study) inform/fill the data gaps that are missing?

Response: There are two things that we are looking at, first we need to better understand farmers who are the adopters of these kind of interventions. We will work to understand what their approach for success is, however, this a separate study from the one Regis will be conducting

Second, we need to conduct study on profiling different types of mother traial farms (those who implements the interventions in their own field), baby trials farms( who are learning from the mothers) and understand what are their results based on, what influences their decision making, household livelihood strategies, yield gaps, and maybe we can go beyond and look the farming systems and gaps emerging from these systems.

Response: After we worked with farmers (baby and mother trails for 6 to 7 years) we also want to answer at least two questions: first is the factor of the intensity of the engagement, and we also need to understand how they transmit the technologies at farm scale, also understand between the mother and baby trials based on farmer’s intensity of exposure?

This study also focuses to understand different typologies of farmers, for example, some farmers will pick the technology, and some will still need time hence profiling becomes key. The study will help us to understand the characteristics of farmers who within the three years would adapt the technologies to their farms. On the other side, the study will be rich is we can do profiling of at least 20-30 farmers and categorize them and see the kind of the effort we need to arrive at better results.

Question: The studies combined mother and baby trials and other components including rainfall variation and different agroecologies, however, the results obtained regarding the chains was missing!

Response: The missing data were for the mother trails only, but there are data on on-farm trials
Response: There is a major social economic survey, this survey captures the missing data for social and human indicators. Data has been collected but are not full analyzed. What was presented was an outlined of what the team are expected to do in the next season rather than the results (data) from last year. Within the next few months we shall have completed the full data set for the missing data, the data will be from the results of the social and economic survey.


  • Withdrawing from some of the farmers (who has been receiving inputs for beyond three years) would be a good treatment to study.
  • What we need to be thinking about in most cases, we should consider the original classification that was done by Jeron (for Tanzania and Malawi) when planning. Also, it would be good to understand farmers’ typologies and characterize them according to some criteria for example why some farmers are referred as the mother, innovators and so on. Also, it would be good to think about someway of picking these people (mother and baby trail farmers) so that the influence made should inform the scaling out, instead of using the outliers.
  • There should be an attempt to have a joint synergetic approach to collect data on different systems promoted using the analysis frame work, also look at the different types of farmers/adopter of the technologies and generate enough bio-physical social economic data to better understand the drivers for their adaptation
  • It would be nicer to consider interviewing 1000 farmers who are adopters, profile them and try to classify and understand what their decision-making is, however the resources are limited.
  • For baby trials, we can conduct a study and see what they are doing at farm scale on their own field, we can engage students to visit 36 baby farmers to collect data and use the GPS to record and collect data from the maize crops trial are fertilized and record and collect data on the total production, same as for soybean crops then at the end the proportional of farms that is under legumes will be identified same applies for the amount of the nitrogen, we can estimate and balance on that.

| M&E Changes and achievements – C. Azzarri/B. Haile/A. Sambala

| Quantifying trends of rainfall and temperature extremes over Central Tanzania to guide targeting of climate smart technologies - F. Muthoni

| Spatial assessment of land degradation in semi-arid zone of central Tanzania - F. Muthoni

| What fits where? FarmMATCH: Matching Agricultural Technologies to Context and Household - J. Groot

| Digital transformation of agriculture: A game changer for sub-Saharan Africa - H. Sseguya