 So I'm just doing the intro because it's very much in line with the topic of this conference Local Innovation and Global Progress, and also then Beyond Health. And we will have a session Thursday morning as a plenary, but today we will hear about two concrete examples from Mozambique and from Malawi, that is very interesting. Looking both into the issues of agriculture, but also climate impacts on agriculture. And we also can know that the topic of climate health is also a very new topic that is upon coming and also under discussion how we can support that. So I'm just doing the intro and over to Mr. Seferino from his Mozambique, no, sorry. Good afternoon, everyone. Thank you. The opportunity to share with you the experience that what we are doing as part of the adoption or the use of DHS2 for the management of agriculture and climate data. I will be presenting here with the first session. I will be doing then and then we have a demo and then after that, the colleague Tiong from Malawi will present also the experience from Malawi. And as it starts to start, so the DHS for agriculture, this project is being implemented by South Digital, which supports always from the universities and the communities, the DHS communities, which is a set of tools that are developed with the aim of integrating or enabling the integration of data coming from multiple sources, which we are one of the components of the earlier warning system, which is using some data, climate data to and other components to produce some information that can be used for specific actions related to any warning and also the management of data that can be used for planning specific interventions. So the platform is one of the objectives, the platform is to address some of the agriculture value change aspects, which includes keeping farmers and communities more updated with the information, climate information and also practices because there are several aspects that are related to the farming process, which we are somehow alerting or producing information sharing with the farmers so that they can use that information for one day to day activities also stimulate some local production and the consumption and there are also objectives on promoting trade. This is the use case that we are going to present today, the tool that we developed that can be used to communicate between the farmers and the consumers as well as raising some partnerships that there are links between the different stakeholders that can be done also through the platform. This process, we are learning it from the malaria early warning system that we implemented in Mozambique for the national malaria control program. That's the idea is to help the malaria program to use malaria data and climate data to predict malaria incident cases in the districts by identifying some permanent outbreaks. So this is an example of how we are using it. We are using climate data and there are also some statistical models that were implemented based on that information and the historical information that have been collected by the malaria program that were entered by the different districts and facilities. There is a predict that we predict what is the possibility of having an outbreak, malaria outbreak. So in this model, we used the Google engine to capture or to collect climate data and then there were some processing using some scripts and also as I mentioned the model that was developed by some researchers that we embedded it on the script and then used to predict or to generate some information that we send it back into DHS too. In terms of the agriculture, the idea as I mentioned is to collect data coming from different sources and this information is expected to be used to enable quick information access and also to be cost effective. During this process, we have worked with some farmers and in this process, some of the tools that we used were SMS best and also Android because we wanted to reach out to the different type of stakeholder. Because there are some situations that there are some lead farmers that they have mobile phones. So those are the Android phones. Those are the receiving information using the apps that were developed using the Android app in DHS to generate that information. Then there are also SMS-based information that was generated to some group of farmers that did not have these new Android phones. In this process, we also have to use the radio communities because there are some situations that some of the farmers did not have mobile phones. In that case, they were in one of the provinces where to work with these community radios and we provided tablets to them so they were able to read the data or to get data from the system. Based on that information, we are generating access to information and then using it to communicate in the specific program that they have in their radios to the different communities. Apart from this enabling this access of information, we also expect the platform to be used to strengthen the disease surveillance in relation to, as I mentioned, it's used now for malaria. We don't listen from the malaria, but the idea is to use the same approach for the agriculture, for the disease that are affecting crops and pests. Specifically, this is the architecture of the system where we are getting some of these systems that are automatic that we are still getting it using manual process. But at the moment, we are implementing or developing tools that will be able to enable or collect that information automatically to be entered in DHS2 and based on that information, they will be getting those notifications. As I mentioned, some of these models, the implementation they are in, the other ones are still in process. With regards to the implementation, this is the model that we have. There are some trucker programs that have been developed. We also have data sets where we are reporting information related to market prices. This is the Ministry of Agriculture runs on a weekly basis, collects a weekly basis information and then they keep it in Excel. So based on that information, we are getting it in the DHS2. There are also some event programs that were developed to collect some of these information or record some of the information. So now I will hand up to Alfredo to continue with the demo of this aspect that we have during the presentation. Okay, hi to everyone. My name is Alfredo Mushanga. I am part of the Saldizia teams and I lead the development teams. So I will start this demo showing the marketplace domain at the first moment because we have here, this is the official website of the Ministry of Agriculture in Mozambique. And Wickel, they publish the prices of the products that they have. So basically they have a set of data sets like each week they publish the standard prices for the products. And as you can see here, we have all prices until May, maybe some other days are not yet published. So the first basic thing that we did is to create a simple data set on the DHS2. We're based on the product holds, the description we can have here the price per unit. So that's something very basic, but that have a very big impact because right now they deal with this data in a very complicated manner because they have a lot of Excel to deal with and they have a very complicated data set and this is one of the examples of files that they have. So changing it to DHS2 was something very basic and that have a very great impact in terms of implementation. The second part is related with the agriculture and consumer parts. What we did is a work at the field was there some community-based activities that was implemented in order to work with the local organizations and get all farmers that belong to some groups registered in the system. So we have a community-based list of all farmers that belong to these groups. So they created the DHS2 as a target instance that we want to track along the time. So basically what they can do in the system, we collect the basic information about them, the name and other personal info and then we have here some options. So these options represent service. It's in Portuguese, sorry for that. But these options represent service that they want to be part of. So if they say that they want to receive this, if they say they want to receive information about pragas that is disease like plant disease, they are going to be part of the groups that will receive notifications for this domain. But we also have the marketplace. So what we did is for each truck and entities that is a farmer, we created a program stage where it's possible to do where they can post. They can publish their own product, where they can specify what is the product, what is the amount that they have available, the unit and price of the unit and other details that are important for this selling process. We did the same for the consumers. We do have also a program stage that is called the quest. So basically the consumer is going to request something that the farmer publishes. So we linked these two programs using custom developed app. And this is one of the apps. This is the web app called Bazzara. This is a digital marketplace. And basically what we have here with this background information, we have a possibility of using the Android application because there is a specific Android app where they can publish and see all products. Our approach was to build a single application because we know that the same farmer at some moment can also be a consumer, can want to buy something with another one. So we built a single application and during this process you can find whether it's going to access as a producer or it's going to access as a consumer. And we have here this web version of the app where we have here some products and if you want to explore, we just go to this field and open the app and can see all the details. And finally, we do have this another portal. So this portal is related with the events that was mentioned here. So at this moment we have the good practice site where we have all events related to the banana planning process and we have various domains there. So basically the radio workers, they come here to this portal and also there is a small application for that and they just read what is here in terms of what is the best practice to produce whatever they want to produce. And the same thing was built also for events. So in some regions there are some events that can affect a set of farmers. One of them are the uncontrolled barns. So if there is uncontrolled barns that I need to inform all people that are near of that possible affected area because maybe it's caused by a natural situation. Like if you have a big forest, sometimes you can have an uncontrolled barn that was not caused by a human. So this is an example of a situation that we use to publish to the farmers in order to have access. So I won't be long on that. I will just stop here. Thank you. Thank you. Good afternoon. Yeah, so the use case that I'll present is on the, yeah, so the use case that I'll present is on the National Agricultural Management Information System for Malawi. In terms of system capabilities, so this is a national system. The idea is to provide integrated support for extension workers working at community level. So there are various system capabilities. So just some select capabilities is to have an early warning component. So I'll just explain a bit on the early warning component. So here is getting in data on weather, but the other aspect as well on the early warning is to be able to do food situation assessments or production estimate surveys. So after I've given the brief, a lot of us will just show you, for example, how the food situation assessment works. So that is where a household, I mean, an extension worker visits a household and then assesses whether they have enough food to last them for a two week period. And they also components like on trade on marketing, so which is, which deals with market information. And then there is agricultural extension so supporting services where an extension worker goes to the community level to document what's happening at the community level. And also, for example, follow what are called lead farmers. So these are model farmers at community level. And then there's also a component on resource, resource mapping. So we just follows, let's say what sort of support can be provided to households and also what projects are implemented where those projects are implemented and what sort of support they are providing. So I think the next thing that we're going to do is I think just show you what we have on the food situation assessment. So we'll start with the household register so the base for operations is the household register. And I think with the demo, you'll be able to see that so I think Lawrence will share the screen take you through the household register was provided in there and then we'll take a look at the food situation assessment and then we can get back and then we sort of like wrap up on that. Okay, so as you're setting up the long list of things you see there is the list of tools that we have implemented within the the platform. So it covers a different range of activities. So starting the ones am is that's for the market information. And then the ones that start with apis those are for production estimates. So there's several rounds of production estimates. So everything that reads apis is on production estimates. And then the the ones that show the dice, those are for extension services at at community level. Right, so I think now we can go to the household register. All right, so, so at the beginning of every agricultural season. And then the agricultural extension offices will go and do registration of all the households. So to do that they make use of the household register, which we did implement. So the actual adding of I mean adding data just go through what's what's what's corrected in each and every, I mean, every stage. So, you correct information on the house of demographics. Like the size of the household. So based on the size of household you correct the other information on details of the household. For example, if you have three, I mean two, it will ask you to enter details for the, the two members which you have mentioned. And then, if maybe you have a member who is above 18 years, you're supposed to get more records on the national ID because in Malawi, anyone below 15, 16 years does not have a national ID. So, we don't want to ask people to enter IDs for people who are not eligible for that. So, you have information on demographics, and then you have farmers that sometimes belong to to a cooperative or a book. So based on that, the people who try to correct information on that are the different activities that the farmer is involved, for example, irrigation, or maybe I depend on rain, rain fed aggregation. So you try, they will correct all those, all those details for that household and then after that, they will check if they get support from from NGOs. So based on the information which is corrected on this register, the household will be sampled for other other activities which will happen during the growing the aggregate season because it's not like every household which they will try to to go through as they're doing different activities they have some sampling mechanisms which they use to include other households may be in gradual production, I mean crops, others in livestock, if they have livestock or hodkacha or foods and and the like so pretty much after they have done that, we'll go through the sampling using that form and enroll those households into different different programs. Yes, so I think as a switching to the, as a switching to the next form so what we'll do next is just to show the rapid food situation assessment so this is where an extension worker goes to a household to check whether the food secure or food insecure. So, as is so you can do the switch. Yeah, so as he's doing the switch, the base so like that support by NGO and also support by government link projects can be used as a basis for determining what support should be provided to households. Alright, thank you. I'll use the same household which we've registered. So in this case, let's assume that this household is being sampled for some activities to do with the rapid food situation assessment. So that household will be enrolled into that program. So the enrollment is pretty much are you just registering that into into the program. And then you need to since this activity, the rapid situation activity happens every two weeks. So in each and every visit you have to indicate how many members you found at the household. Because the idea is in those two weeks, you want to predict if if that household is food secure or it's not, it's fitting insecure so based on the number of households and the different activities which I will show that happens at that household. The system should be able to to tell the aggregate extension work to advise maybe the household that you include food secure or food insecure or for other reporting reporting purposes. So we just assume maybe there are five members in that household. I'll not fill each and every detail. I'll just go to some of them. For example, in Malawi, stable food is maize so maybe they don't grow the other the other. So the main food crop, let's choose to be maize. And then you need to indicate what's the main source of food as in maize. How do they produce that? Is it do they buy or do they purchase? So let's take an example that they do produce on their own. And then they need to indicate if food is available in their area or where they can have access to food in that particular area. And they go into the specific household and check if they have food in that household. Then from there they will try to do some food statistics as in how much food is available in that household. For example, for maize we assume they have 20 kgs and then just checking if they have food available for the next two weeks and how much time do they eat maybe in a day. Is it once, twice or three times? If they have livestock agriculture, that is if they do some fish farming, what type of livestock do they keep? So let's assume they keep poultry and then the other details. You also have a section on coping mechanisms. Maybe if they don't have enough food, how do they deal with that? So pretty much you're correcting all the information which will guide you to determine if a household is secure or insecure. So at the end of the day, let me check for savings and expenditure. How much do they get per fortnight? So as I said, it's corrected per fortnight. How much do they spend? Maybe they spend 500 but they get 400. So I don't think in that case they have nothing to save if they do that kind of spending. And then at the end of the day, the app based on the population and what they have, it will determine the actual energy which they will need at that household based on a number of people. And then based on what they have provided, how much energy, food energy do they have? And then come up with a conclusion if the household is food secure or food insecure as it's been displayed there. So that's one of the use cases which we have in agriculture in Malawi. Thank you. So the assessment is also in one part a basis for providing food support to the households and then also for policy planning in general to look at the food situation within the country and then to determine what sort of interventions should take place. Yeah, so FAO is working on similar workflows globally. I know Afghanistan is one of them, Egypt is one of them. And I just wondering if you guys are duplicating the work or you're working hand in hand or sharing data together or what not. All right, so I think the basis for this implementation is a methodology that the Ministry of Agriculture in Malawi has in terms of assessing the food situation. So this is a direct implementation of that. But the ministry on some other aspects also does work together with FAO. So because I think FAO has also assisted in some other implementations. Thank you very much. Very fascinating presentation. I wanted to ask the presenter from Mozambique. Are the farmers receiving these notifications and are they also able to report back like if a farmer has a dead animal in his that suddenly died in his farm. Can they, do they have a way of reporting back or is just then the notifications are pushed to them. Thank you. Thank you. At the moment, just one way. They reported through the wheat farmers or the extension worker. For the extension worker, the one that checked that information and then report. They don't do it through the system. But yeah, we started with one way, whether it was the project that we had and then the challenge was how they did start because of the challenge on how to export their products. You know, if you find in Mozambique one place with the food that is not used and then other places they are starving for food and then that was one challenge. They said that we have been helping this specific farmer. They don't know how they would like to have a platform where they could export their products or everything starts from there. Then we start helping them and then start assembling the different modules. Just I can add one, one of the things that we did with it with challenges is how to get this farm as resistant in the platform. The reason why we have 3000 now is because we were working with those farmers that were in engaged on those projects. But what we are discussing with the colleagues from Malangu is to use this approach for resistance when they are resisting the outdoors to be able to capture information mobile phone for them, whether they have the just need for process there, they have to add possibility of having the phone number of their health. Then through that we can report whenever the information we can report to those individuals. The 3000 that are using that they are resisted official they do have the Android. We do it through the Android. They are the ones that can publish their products but accessing the platform is open is that we can go and then download it or even if you go to the web we will be able to see if you can do that communication. So, yeah, it is just one way at the moment, but we hope that in future we'll be able to have this the possibility of they sending the information. In fact, there are some implementation that we are having. We're in the wash program where we are using SMS based. The idea is to update the status of the different water sources. If the water source is not working, the comments should be able to report say that we have a water that is broken so that they can be solved at the moment we are doing that implementation with the SNV through UNICEF as a project of the wash project. So we would like to maybe we'll be joining this so that we can use the same approach for the farmers. Super interesting. This is a question to Tung. So how many households is registered? Do you know approximately? I think as of two days ago because we have revised the register so over the past two months they've registered 31,000 households. Okay. And you mentioned it was in 12 districts out of 28 districts. So the plan is that within the next two months, like at the community level, they should register all eligible households because the household register forms the basis for the work that they do. So in the other workflows for the system, you have sample based tools and also some census based tools. So like we have livestock dynamic forms. So those are census based so like all the households that will keep livestock you follow them. But for this, like what we're demoing this the full situation assessment it's sample based. And also production estimates it's sample based. So what we do is that when the households have been registered we have like an app that does the sampling. So then the sample households are then enrolled to this. Yeah, so over the next two months is for them to cover the rest of the households. And then we can use the sampling based on that. Is it every household or is it only household at risk? So you have to like for the household register, all households are supposed to be registered. And then for this then you're taking a sample from that. Yeah, broader list. There's a question there. Thank you. I just wanted to know what has been your, the production of your accuracy estimates or the accuracy of your production estimates. I'm sorry. And what did you have before you before you rolled out a DHS to instance in agriculture. All right, so for the production estimates, the full season that we're going to run that is this forthcoming season. So the previous one we run a pilot just to test the tools. So because the production estimates is just during the rainy season. So the next to be run will be like from September until and too much. But before getting the tools in here, it's sort of like a paper based exercise. You have the extension workers going into the community, doing the survey on paper, and then consolidating that, getting the data into Excel, and then doing the calculations for the estimates to the district level and then you'd have, you know, a meeting convened at national level to do the, to do the reviews. But then at least now with this, because we have that process embedded, as long as it's entered there, then you can sort of calculate and cascade going. You have like a measurable, you have some metrics that could show what your accuracy levels have been before and now so while you're using paper based compared to How would they do a chance to instance? All right, so not at the moment, because as I was saying, the full scale running the prediction part will be this forthcoming season. Okay, sure. Thanks. All right, so I think we've done no other questions we can reflect. Yeah, so we also reflect on some experiences. I mean, that I've been there in terms of implementing GHIs to the machine switch. All right, so for some reflections in terms of the work. So I think with agriculture, the level of digitalization and digital capacity, I think like in Malawian, I think the experiences are also quite similar wasn't big. The level of digital capacity is lower compared to health, for example, will have done a lot of implementations over over the years so it's pretty green. So I think one thing that we've been trying to concentrate and work on is on capacity development. And I think there will still be need for further capacity development at different levels to so across things like general digital literacy, and then also capacity for configuration systems management and so on and so forth. But similar at the same time also trying to reuse some of the existing capacity that has been developed within health. And besides that, not all workflows are currently adequately supported. So there are a few ways need to have work around. So one thing, for example, I've talked about a sampling. So, because out of the box at the moment, we can't generate the samples. So what we've done, for example, in the, for the Malawian use cases to develop a sampling app that would get into the household register and then run the sampling. And then besides that, as well, also when you're handling like that many forms as I showed, it also is cumbersome to do the manual enrollment. So then the app also handles enrollment of the eligible households due to the various survey tools that are sample, sample based. And also in terms of reporting formats, there is need for some adjustments. So we do have, for example, a custom app that assists with presenting the data in terms of the tables and formats that the ministry does use. And, but now I think we've also seen like the development within the core DHS of the custom reporting app. So we intend to test that and leverage that. And I think another interesting bit is on the workflow at the lowest level. For example, having double tracked entities. Because the general tracker model is to track an entity. So when you get to the community level, you have a household. And then you want to track the household, but you may also want to track individuals within within that household. So but in this case, to also listen on the workload, we've mainly just focused on tracking the household, but having the members of the household just within that household, but I think going forward, supporting that dual tracking would be essential because we've encountered that in our other implementation because we having also an implementation, it's community level for a community health information system. So within that, then we have a household actually like three levels you have a community and then you have a household within the community and then you have people that are supposed to be linked to the to the household. Yeah, so I think that's some other points that I felt we should possibly reflect on. So, any additional comments or questions there, the welcome. Yes. So, first, there was a there was a question earlier about if if there is a dead dead cow or some kind of animal disease. I was just thinking in terms of connections to health. Definitely climate health is a big and new topic but but just specifically on animal health one this concept of one health is that also because you seem to be focused a lot on crops, which is great, but you're also thinking to expand into animals and livestock. And I guess the other question is is related to your last comment about household registries. So did you foresee some more direct connection with the civil registration and and if you have those households you could also use that for health, or or education or other. All right. Thank you. So I'll start and then if the in addition self can can chip in. Yeah, so in terms of the health aspect. So maybe because of time we didn't run through everything, but we do have like tools within this for for the livestock component. Yeah, so tracking the various animal dynamics and so on and so forth. So those are available. And then there's also an ongoing discussion in terms of how this can be linked to the one health surveillance platform within within the country because mainly it's in as much as it's supposed to be, you know, human animal environment. So far, the focus has been on the on the human component. Yeah, so but then there's that discussion to see how this can be linked within that. So I think another discussion that we've had is there's the, like as we have this exchange program with you, I owe the rest of a lot of money. And also invest of Jerusalem. So, yeah, I think over the past few weeks we're working on some some courses as well and one thing that also did come up from there was how to look at the system configurations where you addressing multiple use cases. And one thing that's been discussed is that when you look at health, for example, the community health worker also has a household register. Yeah, and if, and then they're within this they also components of nutrition. And if someone is sick, then it's going to affect the labor availability within agriculture and stuff. So I think that's possible linkages is is is critical. And also just on your other question. Also, when you look at this that you're supposed to provide certain integrated services to individuals but at the moment you'd see like agriculture, the, the register health, the household register. And then you're supposed to provide some support to cushion certain households against shocks, but you know there's sort of like configure configuration silos across ministries. So obviously I think going forward, there's need to also have that configuration of the services to then align these sort of systems that you you're not duplicating the effort. I think here the extension worker will visit all the households register them. The surveillance. I mean, the health surveillance assistant will visit all the households and register them. Yeah, thank you. For taking this example, we are we are supporting the implementation of community information system in Angola. And we have been discussing because the Angola model is different from the others. In Angola, they have one, which they call ADECO, which is gender development community development agent. That is not only looking at the health but is also looking for education, financial, all these aspects economic. So we will be having a meeting in July, I think, SHIPO. So and then the Angola team that will be joining. So we hope that we in this meeting will be discussing about this, this process and then see how this can be adopted in other countries and then suggested, for example, see how we can use it in Mozambique and Angola because Angola is already using that. So that is one way of looking at the integrated services that can be provided at the community. And with regard to the disease surveillance, we have been looking at that. Actually, there is, they do have a process within the agriculture, there is a notification. And then after that, these extension workers are notified and then they communicate with these at the district level. The district agriculture, often they send people to the community to contain or even to do the treatment. So after that, they do have that flow, which is similar to the health. So we just need to see how can we link the two and then provide that. Fortunately, you know, in our, we do have these vertical systems in the country. You see the agriculture, education, water and sanitation, health, they're all working in the different separate ways. And we hope that this implementation can help them. In Mozambique, we are also supporting the WASH project. In the WASH project, we do have schools, we have communities. And one thing that the WASH project that they did to present the results, they invited also the Minister of Education and Health to participate in and then to show the platform that have been developed. This platform is based also on DHS. So the idea is, okay, we have this. How can we use the same data instead of doing, because when the WASH, the work with the community, they go to the schools to do the assessment of the situation of the school, they go to the community. But this information is only available under the Minister of the Infrastructure. It's not available to other ministers. So that's the discussion that you need to be there in place and then make sure that this data can be shared across the different penises. Thanks. So, yep, I think there's just three minutes to go. So unless if there's anything, we could, all right. Sorry, maybe I might be ignorant. I'm just wondering, we are talking about the same DHS too. Is this a different interface or is it the same DHS too that is being used in the medical field? And if there's an alert that comes through the system, can it be escalated to surveillance team? Are we talking about two separate interface or it's the same? Thank you. All right, so like for the Malawi case, the implementation is a separate instance. So this one is like implemented within the Ministry of Agriculture and then the Ministry of Health has its own. But I think the limitations at the moment are just mainly because of the existing service configuration. But with the platform itself, it is possible to push the notifications across should that be enabled and demanded by the two entities. And that's why I was saying that there's need to have that sort of discussion because a lot of processes that are similar across and interdependent. Right. So I think we can end there. So thanks. Thanks for your time. Yeah, thanks for your time. And for the contributions.