Climate services for smarter farming – what’s it all about?

Climate services for smarter farming – what’s it all about?

Dr. Julian Ramirez-Villegas, a Climate Impacts Scientist, at CIAT and the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS). Photo by: Neil Palmer / CIAT

Over the last few years, CIAT, CCAFS and their partners* have been doing groundbreaking work to provide climate information to help farmers make smarter decisions. Having achieved success in Colombia and Honduras, now the team wants to take climate services to the rest of the world. Dr. Julian Ramirez-Villegas, a Climate Impacts Scientist, talked to us in-depth about what makes this a revolutionary approach that can enable farmers to thrive in a changing climate.

What is a climate service?

A climate service is basically the act of providing specific pieces of information about the climate in a systematic and sustained way to allow a user to make a decision. Climate affects crop productivity quite significantly. Globally, it’s been estimated that maybe a third of global crop production depends on climate. So you need to be able to understand what the climate effects are and be able to manage your crops.

So if you’re a farmer, accurate and reliable climate information is really important — it can help you make decisions about what crop to grow, which variety of that crop to plant, and when best to plant it.

Sustainably providing this kind of information in the right formats and means to farmers, extension agents, or other people that are helping farmers make decisions — that’s what constitutes a climate service.

Can you give a real-world example?

In parts of Casanare and Meta — departments in Colombia — where farmers grow rice or maize, there’s a single growing season, from around May to September. But during that season, you can have pretty unstable rainfall: It can rain a lot for a few days, but then it can stay dry for a few days. This uncertainty hinders crop productivity because it affects the growth of the crop significantly.

Also, the areas can experience relatively long dry spells, so if you’re a farmer, the ideal situation is to be really sure that there’s going to be rain for the few days after planting. If there isn’t, then the seed that you put in the soil won’t germinate, and all the seed that you purchased will go to waste.

So a valuable climate service in that particular case would involve providing reliable information about the most reliable time to plant.

How do you generate that information?

We’re used to forecasts on television giving us weather information for the next few days, and these help us take decisions in our daily lives. But farmers need reliable information over the course of months — seasonal climate forecasts as well as weather forecasts. Seasonal conditions are harder to predict. The new information technology tools and approaches enable us to generate reliable climatic predictions that farmers can trust.

Typically, we try to answer questions like: Are the next few months going to be wetter or dryer than normal? Or are they going to be around normal? And for that information, part of the process involves looking at weather records to try to construct possible seasonal predictions. We feed this information into crop models and use big data analytics, which allow us to calculate how a particular crop behaves under certain climatic conditions. The model will show us the likelihood of a crop performing well or poorly if planted at a particular time, or will tell us which varieties may perform best under the expected conditions. The models can be quite precise: They can tell you that if you plant your crop between the 15th and the 20th of May, for example, then you’re very likely to achieve the highest productivity.

How do the models account for the fact that climate change is also happening? Doesn’t it mean that the future won’t be like the past?

In our mathematical models, we include the long-term trend. However, in a general sense, we should always be reminded that “all models are wrong, but some are useful.” In some instances, when climate change leads to more extreme climatic conditions that have not been experienced in the past, the type of statistical models that we are using may not work well. But this is why it is very important to continuously and closely monitor local climatic conditions. This will allow us to identify where and when extreme conditions may be increasing, and make our models “learn” from these events, too.

Where do you get the historical climate information? Can you download it online or do you have to request it from the weather agency?

Many of the weather agencies we interact with are working toward having online systems where you can download directly, or make a request online. At the moment, we request it, and they’re happy to share it. Colombia has a policy for open meteorological data, which makes our work very effective and efficient there. We also recognize the hard work that IDEAM, the Colombian meteorological agency, puts into data collection, curation, and sharing.

We can typically get 15-30 years of climate information for a given location. In some cases, we can get up to 40 years. It depends on how long the meteorological agency has been recording the climate.

What kinds of recommendations do you provide to farmers?

For rice and maize, which are the two crops that we work on most in Colombia, the analysis tells you basically three things: firstly, whether a farmer in a particular locality should plant or not plant — because there might be a risk of crop failure; secondly, if the farmer should plant, then when they should do it; and thirdly, which crop variety they should plant, based on the likely seasonal climate.

There’s no standard set of recommendations. They are tailored depending on the climate predictions for that season, and on the local conditions and knowledge of technicians and farmers. Agronomy in a way is a kind of recipe, but you need to ensure you get the ingredients right for each situation.

How do you make sure this information gets to farmers?

We have a series of different delivery mechanisms, and when I say “we,” it’s actually not only CIAT, but also our partners in these countries.

We have been creating new tools and knowledge, but at the same time, we have been building the capacity of farmers’ organizations, to empower them and help them embrace this knowledge. We’re also working with teams of people who not only run models, but who also look at the climate conditions, and to interpret the outputs of models and convert that into advice for farmers.

We have also set up and been working through platforms called technical agroclimatic committees. These are roundtables of people from different institutions, including those within the meteorological service and different farmers’ organizations for different crops, so you have climate experts and farmers sitting together. The committees are able to issue the forecast as a joint output, along with recommendations for farmers in a given region. That comes typically in the form of a bulletin. Because these agroclimatic committees are local, they provide very specific information, and as such, they have been quite effective: Whoever comes to the meeting leaves with a clear set of recommendations and a clear idea of what might be coming weather-wise in the next few months. They then are able to share these with the farmers they work with.

What are some crucial elements to providing successful climate services?

First, providing a climate service is a two-way communication process with users. You need to talk to the users of the service, and you need to make sure that the information is tailored to the needs of those users.

Building the capacity of farmers organizations was crucial to our work in Colombia. The project we have there provided them with funding, which allowed them to hire people to develop tools with us. However, now they are themselves funding their teams, thus preserving the analytical capacity in-house, and being less dependent on external funding. Of course, external funding always helps to explore new topics and expand work, but the core capacity is now there.

With experts from IDEAM, farmer organizations, and other institutions, we also developed an online platform that automatically provides forecasts. With the help of farmers and technicians, we were able to make this much more tailored to users. It was a lot of work, but it increased the sense that they belong to the process. It empowered them and helped make the tool much more locally relevant and useful.

So you can see that providing climate services is the work of many people. Even inside CIAT, there are more than 30 people working on it. And outside CIAT, there are farmers’ organizations, secretaries or ministries of agriculture, meteorological service providers, and climate experts from the International Research Institute for Climate and Society (IRI) at Columbia University. It’s in no small part thanks to IRI that we and our partners know what climate prediction tools exist and how to use them.

This work dates back a number of years, and it’s had many, many players without whom we could not have not pursued this.

So what makes climate services a unique proposition to farmers and for CIAT?

Before we started our first project in Colombia, under an agreement with the Ministry of Agriculture, a lot of people were aware of the importance of the climate but didn’t know what to do about it. After going to the field, we realized that Colombian farmers were planting based on what happened last year. So if I planted on the first of May last year and I got a good crop, then I’ll do the same this year. With the amount of climate variability that we have here in Colombia, particularly rainfall, that’s a recipe for disaster. You cannot expect that climate conditions are going to be exactly the same from one year to the other: At the extreme, you might have a La Niña cycle this year, so it’s very wet, and an El Niño cycle next year, meaning it’s hotter and drier. It just wouldn’t make sense to apply exactly the same strategy across time. There was clearly a gap for a service that would systematically provide information about what to do when certain climate conditions are coming.

CIAT, of course, has been working on climate change for a long time, and leads the CGIAR Research Program of Climate Change, Agriculture, and Food Security (CCAFS). We’re collaborating with, for example, IRI at Columbia University, to be able to build tools that can connect what climate scientists are producing to insights that are relevant to farmers. We’re also able to use our network of partners to reach farmers. And this is where we see our role and comparative advantage: in building that bridge to connect hardcore climate scientists with farmers in the field.

What are the challenges to providing climate services? How do you address them?

Right now, we’re working with farmers’ organizations, who are empowered with the tools; they provide information to thousands of farmers. But there are many farmers that don’t belong to any farmer organization. They are typically small-scale farmers who are difficult to reach by typical extension services or communication channels. Also, particularly in Latin America, there are many regions that don’t have government-sponsored extension services at all. That makes it more difficult to reach these farmers with climate information. Plus, they’re often the most vulnerable farmers.

In these cases, other communication channels should be used. For example, radio would be much more effective because it is particularly good at reaching those in remote areas. But then it’s not only about the mechanism; it’s also about what you are communicating. In many localities that we’ve worked in, people would say, “Yeah on the radio I get the forecast, I get the climate predictions, but if I live in the Cauca Valley and the forecast is for the Andean region, how is that going to be useful for me?” We need to make the forecasts locally relevant.

Radio is just one example. Text messages, or even TV, could also work. There’s one very interesting example in Rwanda. There, CIAT is in the process of establishing a system whereby you have a TV screen located in district agricultural offices, which constantly provides climate predictions that are tailored to agriculture in that locality. I thought that was quite a neat idea.

Another key challenge is improving the accuracy of the prediction models. You say, “OK, you’re making climate predictions, so you’re telling me what might happen in the next six months. Is that really accurate?” The analysis that we’ve done suggests that the predictions are accurate about 80 percent of the time. This is actually a really high success rate, but we need to find ways of making the predictions yet more accurate, if we want to reduce the climate risks associated with farming to a minimum.

Do you think 100 percent accuracy is possible?

No natural phenomenon is 100 percent predictable. But we can reach greater levels of accuracy with better models. This would require significant investment in research on climate prediction.

Not only that; in some cases, we’ve realized there’s also an issue of data quality. So in regions where data quality is poor or where you have very few weather stations, the climate is more challenging to predict.

What’s next for CIAT’s climate service work?

Some of the prediction methods and mathematical models that we’re using are of a lot of interest to CIAT and partners in Africa and in Asia. So we really want to improve knowledge exchange across regions. To do this, we need to ensure that everyone knows there is a climate service framework and that we’re able to fit projects into that framework. Next, we want to take full advantage of the tools that we have or that we’re developing, to reduce duplication and enhance integration.

We see of course the area of climate services growing as we go into the future. Right now, we’re working with the U.S. Agency for International Development (USAID) to create a climate service suitability map. This map would take into account factors such as level of climate predictability, the difference between potential and actual yields in different regions, and the levels of food insecurity and malnutrition. It would show hot spots, where if you invest in climate services, you might be very effective at getting development outcomes. Once we get this work done, it should help USAID reorient its investment in its different priority countries.

So we expect a lot of growth, and I think so far we’re getting a lot of traction.

Would you say climate services is like the missing recipe to farming success?

Yes, though that’s not to say that it all works perfectly, but we’ve made enormous progress, and right now, we estimate that 300,000 farmers are receiving climate information as a result of our work. It’s a great start, but there’s a lot more to do.

CIAT wants to make this work truly global. We’ve proven that it works, that farmer organizations and farmers like and embrace it, and that it can save them money and boost productivity. Imagine if we could implement similar systems in sub-Saharan Africa or South East Asia — there are potentially millions and millions that could benefit.

* CIAT’s partners in providing climate services include the Colombian National Federation of Rice Growers (FEDEARROZ), Colombian Association for Fruits and Vegetables (ASOHOFRUCOL), the National Federation of Cereal and Grain Legume Growers (FENALCE), the National Institute of Hydrology, Meteorology, and Environmental Studies (IDEAM), National Directorate of Science and Technology, Honduras (DICTA),  Agronet, Local Technical Agro-climatic Committees (LTACs), the Permanent Committee for Contingencies (COPECO), and the Secretariat of Agriculture and Livestock (SAG). Funders include the Colombian Ministry of Agriculture and Rural Development (MADR), the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS), the Climate Services for Resilience Development (CSRD) Program of the U.S. Agency for International Development (USAID), and The Nature Conservancy (TNC).