Tuesday, October 25, 2016

High resolution land cover maps for West Africa

Finding high resolution land cover maps for East Africa is relatively easy, as there a quite some numbers of product freely accessible online. However, for me West Africa remained quite a challenge. But only until today!



I came across a new set of high resolution land cover maps for selected West African countries done by OSS (Observatoire du Sahara et du Sahel). What a wonderful product!

They have created fantastic atlases for Senegal, Mauritania, Niger, Burkina Faso Chad and Mali. The one for Nigeria is on the way. Sad that the data is not directly accessible on their website, but the beautiful atlases are!


Have a look at the Burkina Faso one here!

Friday, October 21, 2016

Identifying dynamics of natural resources in Somaliland through participatory mapping

If i tell you Somaliland, what is your first thought? War? terrorists? Al-Shebab? But in reality Somaliland is all but this. It is a place, which GDP depends for more than 80% on livestock production, exiting for a livestock scientist. It is a young democracy of an country that has not been recognized by the international community, yet has its own currency, own visa procedure and own government. Much of the international aid focuses on emergency relief and food aid. Little money flows for development, but some does, for example with FAO-SWALIM, who create a lot of good maps.

Hargeisa livestock market
Also Terra Nuova is one of the NGOs that does development work, and works mainly in the livestock sector. ILRI was contracted by Terra Nuova to implement the research component of one of their livestock project. Next to collecting analyzing data and create documentation for a country that has lost it all in a war, we also build capacity of ISTVS the vet school in Sheik


This week, i have hosted a participatory mapping workshop with decision maker to discuss the current state of natural resource that are needed to support the livestock sector. Participatory mapping asks participants to map, i.e. tell us where a given natural resource is and tell us more about that resource, for example the current state of the resource, who the user is, if there are conflicts and if people somehow are changing behavior. 


This allows to identify dynamics that define changes of natural resource, so that we can predict changes that might occur in future and support policy maker to take actions now that could address future issues.  


In a first step we have asked participants to rank the resource needed for the livestock sector. All three groups came up with the following order : 1. land, 2.water, 3. livestock feed. Number four was different from the groups, some would say livestock, other soil.

Want to know more about what we learned from the workshop? We are still processing the reports! So keep posted! 

Friday, October 14, 2016

What is big data?

What do you think big data is? Please fill this pool before reading the rest of the post!





Really, have you filled the pool?


I had recently a discussion about Big Data with my colleagues, and realized that most of them believe that big data is about a lot of data... what about you? It is true that until today there is no accepted definition for the concept of Big Data, the only agreement across definitions one can find is : Big Data is about so much more than the big amount of data.


http://www.big-data-book.com/on-the-book

I have recently finished reading this book, "Big Data : A revolution that will transform how we, work and think." A read for anyone who is interested in this emerging hot topic.

This book basically claims : 
It ushers in three big shifts: more, messy and correlations (the book’s chapters 2, 3 and 4). First, more. We can finally harness a vast quantity of information, and in some cases, we can analyze all the data about a phenomenon. This lets us drill down into the details we could never see before. Second, messy. When we harness more data, we can shed our preference for data that’s only of the best calibre, and let in some imperfections. The benefits of using more data outweighs cleaner but less data. Third, correlations. Instead of trying to uncover causality, the reasons behind things, it is often sufficient to simply uncover practical answers. So if some combinations of aspirin and orange juice puts a deadly disease into remission, it is less important to know what the biological mechanism is than to just drink the potion. For many things, with big data it is faster, cheaper and good enough to learn “what,” not “why.”



And why is it a revolution? 
A reason that we can do these things is that we have so much more data, and one reason for that is because we are taking more aspects of society and rendering it into a data form (discussed in chapter 5). With so much data around, and the ability to process it, big data is the bedrock of new companies.
The value of data is in its secondary uses, not simply in the primary purpose for which it was initially collected, which is the way we tended to value it in the past (noted in chapter 6). Hence, a big delivery company can reuse data on who sends packages to whom to make economic forecasts. A travel site crunches billions of old flight-price records from airlines, to predict whether a given airfare is a good one, or if the price is likely to increase or decrease. These extraordinary data services require three things: the data, the skills, and a big data mindset (examined in chapter 7). Today, the skills are lacking, few have the mindset even though the data seems abundant. But over time, the skills and creativity will become commonplace — and the most prized part will be the data itself.




What are the threats (or how societies will have to re-invent privacy)? 
Big data also has a dark side (chapter 8). Privacy is harder to protect because the traditional legal and technical mechanisms don’t work well with big data. And a new problem emerges: propensity — penalizing people based on what they are predicted to do, not what the have done. At the same time, there will be an increasing need to stay vigilant so that we don’t fall victim to the “dictatorship of data,” the idea that we shut off our reasoned judgment and endow in the data-driven decisions more than they deserve.
Solutions to these thorny problems (raised in chapter 9) include a fundamental rethink of privacy law and the technology to protect personal information. Also, a new class of professional called the “algorithmist” that will do for the big data age what accountants and auditors did for an era 100 years ago, when the cornucopia of information swamping society was in the form of financial data.
What role is left for humanity? For intuition, experience and acting in defiance of what the data suggests? Big data is set to change not only how we interact with the world, but ourselves.

So look at the pool again? Which answers would you choose now? If you are not ticking all of them, then get the book!

Thursday, October 6, 2016

Reaching the last mile : how WeFarm is about to bring the revolution into communicating to smallholder farmers

Mondays morning can be though, but this week i had a fascinating one. I hosted a short presentation of WeFarm, a social entreprise that offers sms services for peer-to-peer farmer advice. Though many young people are tech oriented and many have access to internet and social media in Kenya, the vast majority of farmers still own feature phones, i.e. cannot access the internet. This is the gap that WeFarm adresses.

They have a very interesting business model :  all the sms are free to the farmers (at least on Safaricom), the services are funded by the organization that want to use the data WeFarm are collecting.  This is the first business model i have seen that is very realistic about farmer capability and willingness to pay!



At the back of WeFarm there are data analyst who develop machine learning approaches to improve on how feedback is sent to farmer but also to make their data valuable to potential customers. WeFarm is also hoping to introduce a peer-rating system where farmers will be able to 'upvote' good pieces of information, ensuring that other farmers receive quality advice and answers.
most farmers still have feature phones

It is a promising start up, clearly addressing the gap of bringing the information to the last mile namely the smallholder farmer. However, this morning we also raised many questions and challenges such as:
  • How can WeFarm work in pastoral areas where farmers are likely to be moving and what to know about conditions on the other side of the country and not per se from their neighbors?
  • How can expert knowlege be introduced in WeFarm business models?
  • How can accuracy of advice be improved?


We had an interesting debate from which both sides, WeFarm and scientists could learn from each other! Wanna know more, check up WeFarm's site (and if you are a farmer, register to their services! it is free!)