Tell me a story… How do climate perceptions match up to observations?

By Helen Greatrex.

“I don’t know the reasons, but I know the climate is changing,” Medhin Reda, a 45-year-old farmer eking a living for her family from two rain-fed fields in northern Ethiopia, said. “I don’t really remember drought seasons as a child… The rain was good.” [1]

The world would be nothing without stories.  They can unite or divide, transform or reassure and create vivid emotional links that transcend culture language and age.  A scientist who can tell the story of their work or include stories about its impact will be able to reach many more people than one who just quotes dry facts and figures.  One type of story that crops up  in climate science comes from the memories and perceptions of local communities.  They have great power; a non-scientist might not take in the precise science behind climate change, but the story of a farmer who can no longer feed his family due to drought lingers in the mind.  Consequently, climate perceptions are big business, cropping up everywhere from the popular press to government policy papers.

Although local perceptions about how climate has changed are often used as political spin, they are taken seriously by decision makers.  There are many development projects designed solely on the basis of local climate perceptions (rather than using climate observations), especially in data sparse regions of the world such as sub-Saharan Africa.  It might be easy to discount this as an unscientific approach, but most African livelihoods are intrinsically linked to the climate through rain-fed agriculture, so perhaps it is reasonable to assume that perceptions are valuable in a more qualitative way.

In that case, how do the perceptions of climate match up to observations?  This is difficult question to answer as it requires linking social studies with climate science, so it is only just starting to be addressed.  There have been two interesting recent studies on the subject– one by Rao on climate perceptions in Kenya [2] and one by a team from the Department of Agriculture (the one next door!) about climate perceptions in Uganda [3].

Rao went to Katumani, Kenya, to survey over 200 farmers on their perceptions of climate, and then compared these against rain-gauge observations.  He found a strong belief amongst the farmers that seasonal rainfall totals were reducing – supported by the fact that yields had declined severely in recent years.  When the research team compared rainfall and yields they found that yields had indeed reduced but that there was no change at all in rainfall!

Kenya perceptionsRainfall and yield time-series in Machakos, Kenya (5 yr moving average)

Further research seems to suggest that the reduced yields were in fact caused by reduced fertilizer subsidies. Without both pieces of information, development projects in the region would have struggled.  Those based on perceptions that addressed a lack of rainfall would not have tackled the underlying issues and those which discounted climate change would have suffered from low farmer uptake.  Taken together though, they provide a valuable start for dialogue and adaptation.

The second example is in Uganda.  There is increasing evidence to suggest that Ugandan farmers believe that rainfall is becoming more variable and that the rainfall seasonality is changing.  This has led to emotional and highly politicised articles in papers such as The Guardian that use the accounts of local farmers to make the case that ‘the outlook for Uganda is bleak’, ‘farmers will suffer’ and ‘the situation could spell disaster for subsistence farmers’ (it even mentioned polar bears!).  However, there appears to be growing evidence in the climate community that there is no trend in Ugandan rainfall amounts or distributions.

This led Sarah Cooper, Henny Osbahr and Peter Dorward from the Dept. of Agriculture and Roger Stern from the Statistical Services Centre to try and reconcile the mismatch.  They examined whether climate perceptions from 90 Ugandan farmers matched observations from local weather stations (the report can be found here [3]).  The farmers perceived changes in rainfall amount, seasonality, distribution and intensity, whilst as shown in the figure below and in the paper, very little trend can be seen in the observational data.

Uganda perceptionsTrend in monthly rainfall totals at Mbarara, Uganda

It turned out that farmers’ perceptions of an abnormal climate were very much tied into the impact of rainfall on crop production:

“The pattern [of rain] has changed since 1989 because before then I had bigger harvests”

“The weather has changed because there is a food crisis”

“There is more drought which is leading to the drying of wells which was not the case before”

Many of these changes might not have been to do with changed rainfall amounts; instead they are more likely to do with increased temperatures, reduced soil fertility and growing populations.  But regardless of this, the experience that their livelihoods are affected by a changing climate is real for farmers and it’s important for us to note that even though we might say that rainfall is not changing in a region, it does not mean that water resources there are not decreasing.

Hopefully these studies show that combining local expertise and climate observations will help adaptation much more than either one in isolation  – so hopefully one day, the ‘outlook is bleak’ climate perception articles in newspapers will get replaced with ‘the outlook is bright!’


[1] S. Regassa et al. (2010), The rain doesn’t come on time any more;  poverty, vulnerability, and climate variability in Ethiopia Rep., Oxfam International.

[2] K. P. C. Rao et al. (2011), Climate variability and change: Farmer perceptions and understanding of intra-seasonal variability in rainfall and associated risk in semi-arid Kenya, Experimental Agriculture, Volume 47, Issue 02, April 2011, pp 267-291

[3] H. Osbahr et al. (2011), Supporting agricultural innovation in Uganda to climate risk: Linking climate change and variability with farmer perceptions, Experimental Agriculture, 47(2), pp.293-316.

Leave a Reply

Your email address will not be published. Required fields are marked *