imagine if Ngos could confidently access high quality climate information that specifically support their decision making.

The ability to view custom climate statistics in one place 

We envision that there could be a systematic way of accessing information about Somalia's climate - where custom, NGO defined statistics are output on the same page.  This could build on existing portals or make use of new visualisation tools such as R-Shiny

This tool needs to address many or all of Concern Worldwide's requirements, especially the ability to view the situation at a national, livelihood zone, district or local scale.

This should be a place where tailored climate statistics can be showed on the same page - for example the cumulative seasonal rainfall total to-date or SPI, but also with the ability to compare against 'normal' or against historical years of note. 


understanding 'normal'

We know that a 'normal' or a 'bad' year means different things for different livelihood groups, or for people with different coping capacities.  For example, the food security curve of agronomists (well-fed after harvest, hungry when stores run out) is different to that of pastoralists (relatively OK, but crashes if the livestock die) 

We have a lot of information about these groups and about the climate disasters that have happened in the past.  It's possible to build up climate baselines for these different groups which would then be able to inform the best statistics for monitoring and any thresholds that need to be examined.


Forecast based financing

Each decision to take action costs money and time. In each case, Concern needs to justify that taking a climate action makes sense - whether that's a director asking for disaster financing, or local in-season decision making.

Concern worldwide have recently developed a forecast based financing framework - LINK TO DUSTIN'S PAPER HERE - where a series of thresholds are used to define the probability of a specific disaster happening (e.g. IPC level 4, or famine for a specific season).  From this, they can calculate the cost of responding now, vs the cost of waiting until later. 

We want to be able to use the tools above to explore the climate aspects of this system.  What are the correct climate statistics to use?  What is the uncertainty on them and how can they easily be input into the system?