Relationship between climate projections and trends

Following on from lasts weeks post that looked at the projections for water availability across the whole of the African continent, this post is going to look at the relationship between the projections and the current trends. A key case study that we'll look at today is the 'East African paradox' (Rowell & Booth, 2015). This will highlight how projections are just estimations of the future will the help of modelling and substantial quantitative data. So with any estimations, there will be discrepancies between the projections and the actual real world trends. So let's get into it!

Source: Andy Hall - A young girl stands amid the graves of children who died after a drought hit Dadaab refugee camp in Kenya in 2011

In East Africa what are the climatic projections then? From figure 1.1, you can see the long term projections for precipitation all the way too 2100 (Rowell & Booth, 2015). There is a steady increase above the average, to around 10-15% above anomaly by 2100, from the 39 CMIP5 model. This projection is supported by a range of other sources, which all highlight that there is projected to be significant increases in MAM precipitation (March-May) (Ongoma et al. 2017). 

                                                      Figure 1.1 - Lowpass-filtered time series (50% amplitude cutoff at 10 yr) of observed (left-hand line) and projected (right-hand line) MAM rainfall anomalies averaged over the greater Horn of Africa (area shown by inset). (Rowell & Booth, 2015).

What figure 1.1 does beyond illustrating the projections for precipitation, it also highlights the current trends seen from 1980 to 2010. It can be seen that the current trends do not currently fit the 39 CMIP5 projections. Whilst in general it is argued that "projections of climate change over Africa are highly uncertain, with wide disparity amongst models in their magnitude of local rainfall and temperature change..." (Rowell et al. 2015), this East African paradox case study shows no sign of diverging from this highly uncertain projection trend. Precipitation has significantly decreased in East Africa, with a series of largely devastating droughts hitting countries such as Kenya, as seen in the picture above. 

Hypothesis for divergence between trend and projection in East Africa:

Many papers have argued for various hypothesis of trend and projection variability and inaccuracy. Rowell and Booth (2015) highlight three main hypothesis types:

1. The contrasting rainfall trends may arise from data errors.

2. The past or projected trends may arise simply from random sampling of a statistically stable climate.

3. There could be genuine physical reasons for the contrasts.

So lets look at these hypothesis individually now. The first hypothesis, that the divergence in trends and projections arises from errors in data has been promptly rejected. The significant and thorough data sets that are used in these models are accurate and represent real world data collected, thus, it is not the raw data itself that is error prone. The second hypothesis highlights that natural variability may be contributing to this divergence in trends, with "within a red noise process, the timing of a strong natural multidecadal coupled ocean–atmosphere mode may produce random trends for a period of a decade or more" (Rowell and Booth, 2015). This theory is supported by sea surface temperature (SST) trends in the Pacific, with significant decadal variability, which is a major driver of climate in East Africa (Lyon & Vigaud, 2017). The CMIP5 model does not seem to illustrate this natural SST variability in its precipitation projections for East Africa, which is argued in these papers as to part of the reason for the difference between current trends and projections. The third hypothesis for this African paradox is that there could be physical reasons for the contrasts in trends. Rowell and Booth highlight "that the balance  competing anthropogenic forcings could be changing. In particular, the past trend may have been driven more by aerosol emissions, or by land-use changes, whereas the projected trend is driven more by carbon emissions". 

All these three hypothesis highlight a few key conclusions. Our anthropogenic forcing may be changing, with new forcing factors now have more effect then they historically did. Further more, our current projections may simply be ignoring natural variability possibly. Overall, the academic papers in this post all highlight one agreed and unified position, that climatic projection models exhibit major errors and whilst becoming increasingly sophisticated and accurate, will also be prone to discrepancies between real world trends and projections. This brief post therefore highlights the East African paradox case study, and provides some short detail on the potential reasons for the differences that can occur between trends and projections. If you want to learn more about this topic, make sure to read the key paper in this post:  'Reconciling past and future rainfall trends over East Africa. (Rowell & Booth, 2015).