Why we need to save Africa’s historical climate data
14 Oct 2015
Climate data is the lifeblood of any early warning system and the cornerstone for any resilience building effort. It not only allows us to monitor adverse impacts across development sectors, populations and ecosystems, but also helps countries prepare for and adapt to the realities of climate change and protect national development gains and goals. Climate data generally falls into two categories: historical data and data from recent and current observations.
While most people understand the importance of current and recent climate data, fewer appreciate the equal importance of historical climate data. Historical data allows us to establish long-term trends, which in turn help us understand and better plan for future changes in climate. Historical climate data records help us develop climate models, satellite-based instruments and seasonal forecasts, as well as provide foundational data for adaptation studies at local, national and regional scales.
For example: Climate models are mathematical representations of the interactions between the atmosphere, oceans, land surface, ice and the sun. Once a climate model is developed, it has to be tested to find out if it works. And since we can’t wait for 30 years to see if a model is any good or not, the models have to be tested against the past in a process that is called “hindcasting”, which relies on historical observations of what has already happened. The simple assumption of hindcasting is that if a model can correctly predict current trends from a starting point somewhere in the past, one can expect it to predict with reasonable certainty what might happen in the future.
Meteorological data observations in most African countries date back to the early 19th Century (for example, in Tanzania the first meteorological observation was made along the coast in 1850). Once recorded on paper, the observations are kept in various formats in data archives located in meteorological agencies. But this historical data is recorded largely on paper and, depending on the age of the paper and the condition of the archives, some of data is unreadable or is wearing out at dramatic rates, while other data is recorded with handwritten ink that fades over time. This is a slow-motion tragedy: this historical information is an invaluable resource for a continent that is already being hit hard by climate change.
It is critical that we digitize this information in order to preserve it. I work with the Climate Change and Disaster Risk Reduction Cluster of the UNDP Regional Service Centre for Africa, and we have worked to carry out an assessment of climate data rescue needs in 6 African countries: Gambia, Malawi, Sierra Leone, Tanzania, Uganda and Zambia.
Some of these countries have almost 1 billion pages of historical data that is not digitized. That’s a lot of paper and a lot of work, but it also represents decades of historical information that can prove critical in long-term forecasts. These 1 billion pages are for six countries only, and the situation is similar throughout much of Africa. One thing that has stuck with me throughout this work is the anguish the men and women who working for the national meteorological agencies feel about losing this historical data.
For instance, a Chief Meteorologist at the Zambia Meteorological Department told me that one day he woke up to learn that a colleague in his office had set a large portion of climate data on fire in a bid to create more office space—literally, several years of irreplaceable data had just vanished into smoke. In other places, the poor conditions of archives have led to climate data being eaten up by termites or destroyed when offices are flooded. In Sierra Leone, the Meteorology Department points to several thousands of historical climate data sets that were lost during the civil war.
As should be clear, preserving historical data isn’t just an exercise in saving bits of old paper—it is an investment that can truly save lives and enhance climate risk preparedness by helping to create better forecasting, better projections and better early warning systems. Indeed, ensuring that all historical climate data is rescued and digitized can contribute to improving efficiency and effectiveness in the provision of climate services, as well as build societies’ resilience against the effects of climate change.