Reported Rainfall timing in Village of Genete (Ethiopia)

In the village of Genete, we can perform this crosscheck.

We will use the timing of the season in decads, a common form for agronomists. A decad reflects approximately 10 days from the new year, so decad 4 starts Feb 1, Decad 7 starts March 1, Decad 19 July 1, Decad 25 is September 1.

The quantitative focus group in Genete said that the key times they were vulnerable to rainfall were during flowering (late July and September) and preparation and planting, beginning in March and continuing into early July.

Below you can see the Rainfall climatology for the village, that is, the average rainfall during the year over their reported cropping cycle. You can see bumps in the climatology plot during the two seasons common in this part of Ethiopia, the earlier Belg, and the later Kiremt seasons. In some places these happen a long time apart, but on others they touch, or even merge, and the farmers make cropping choices for two seasons, or one long season depending on what seasonal timing they experience.

The most vulnerable times of year initially reported are higlighted in blue for the beginning of the year, and orange for the flowering, later in the year.

To provide an additional verification source, green bar reflects the timing that the vegetation would change color to reflect seasonal end, after the rainfall in the time highlighted in orange had passed, and the landscape had time to respond.

There was intense discussion about if the most important start of the season was decad 7 or 18.

Evaluating information and indexes

The chart below compares the bad years that the village reported with the years that had low rainfall in the satellite record, during the highlighted times just discussed.

• Years are listed on the x axis
• Taller bars are worse years,
• Grey bars are the farmer reported droughts
• The Tallest grey bar is the worst year the village remembered

The satellite rainfall record is filtered by the farmer reported vulnerability timing

• The early season sowing (blue), late season flowering (orange), and vegetative de-greening (green) periods are the same colors as before

• the colored bars represent how bad the satellite rainfall was during those decads in each year

• There is an additional red bar that represents if any of the satellite metrics (the combined severity) reflected a bad year, which was used to design the insurance index.

• A red o marks any year where none of the satellite indicators were bad enough to be considered a meaningful deficit.

You can explore the agreement in the information

• Which years did the farmers say were bad that were also represented in the satelite data for the reported seasonal timing?

• Which farmer bad years were not reflected in the satellite data with that timing?

• What is the ratio of years with hits vs misses?

The answers to those questions are summarized automatically if you scroll to the matching table at the bottom of the page.

In the matching table, you can see how much any data source matched another by looking at the rows and colums. For example, you can see that the combined severity agreed with the farmer bad years 50% of the time.

But what if we wanted to know the matching of the timing for rainfall vulnerability earlier in the season?