ArticlePDF Available

Attributing drivers of the 2016 Kenyan drought: 2016 KENYAN DROUGHT

Authors:

Abstract

In 2016 and continuing into 2017, Kenya experienced drought conditions, with over 3 million people in need of food aid due to low rainfall during 2016. Whenever extreme events like this happen, questions are raised about the role of climate change and how natural variability such as the El Niño - Southern Oscillation influenced the likelihood and intensity of the event. Here we aim to quantify the relative contributions of different climate drivers to this drought by applying three independent methodologies of extreme event attribution. Analysing precipitation data for the South East and North West of Kenya we found no consistent signal from human-induced climate change and thus conclude that it has not greatly affected the likelihood of low rainfall such as in 2016. However, 2016 was a La Niña year and we show that this event was indeed more likely because of the specific sea surface temperatures. There is a trend in temperatures in the region due to climate change that may have exacerbated the effects of this drought. By analysing precipitation minus evaporation and soil moisture, simulated by one climate model only, we did not see a reduction in moisture in simulations in the current climate compared with simulations without climate change. However, there are expected effects of higher temperatures that our simulations do not cover, such as increased demand on water resources and stress on livestock. Although we find no significant influence of climate change on precipitation, we cannot rule out that temperature-related impacts of drought are linked to human-induced climate change.
INTERNATIONAL JOURNAL OF CLIMATOLOGY
Int. J. Climatol. (2017)
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/joc.5389
Attributing drivers of the 2016 Kenyan drought
Peter Uhe,a,b*,Sjoukje Philip,cSarah Kew,cKasturi Shah,dJoyce Kimutai,e
Emmah Mwangi,eGeert Jan van Oldenborgh,cRoop Singh,fJulie Arrighi,fEddie Jjemba,f
Heidi Cullendand Friederike Ottoa
aEnvironmental Change Institute, University of Oxford, UK
bOxford e-Research Centre, University of Oxford, UK
cRoyal Netherlands Meteorological Institute (KNMI), De Bilt, The Netherlands
dClimate Central, Princeton, NJ, USA
eKenya Meteorological Department, Nairobi, Kenya
fRed Cross Red Crescent Climate Centre, The Hague, The Netherlands
ABSTRACT: In 2016 and continuing into 2017, Kenya experienced drought conditions, with over 3 million people in need
of food aid due to low rainfall during 2016. Whenever extreme events like this happen, questions are raised about the role of
climate change and how natural variability such as the El Niño - Southern Oscillation inuenced the likelihood and intensity
of the event. Here we aim to quantify the relative contributions of different climate drivers to this drought by applying three
independent methodologies of extreme event attribution. Analysing precipitation data for the South East and North West of
Kenya we found no consistent signal from human-induced climate change and thus conclude that it has not greatly affected
the likelihood of low rainfall such as in 2016. However, 2016 was a La Niña year and we show that this event was indeed
more likely because of the specic sea surface temperatures. There is a trend in temperatures in the region due to climate
change that may have exacerbated the effects of this drought. By analysing precipitation minus evaporation and soil moisture,
simulated by one climate model only, we did not see a reduction in moisture in simulations in the current climate compared
with simulations without climate change. However, there are expected effects of higher temperatures that our simulations do
not cover, such as increased demand on water resources and stress on livestock. Although we nd no signicant inuence of
climate change on precipitation, we cannot rule out that temperature-related impacts of drought are linked to human-induced
climate change.
KEY WORDS attribution; climate change; drought; El Niño; Kenya
Received 4 April 2017; Revised 23 November 2017; Accepted 24 November 2017
1. Introduction
At the beginning of 2017, much of Kenya was suffering
the effects of the low rainfall and high temperatures that
occurred in 2016. By January, 2.6 million people in Kenya
were in need of food aid according to a report conducted by
the Kenya Food Security Steering Group (KFSSG, 2017),
rising to over 3 million in March. The Kenyan National
Drought Management Authority had declared alarm stage
in 15 counties, and alert stage in 7 counties (NDMA,
2017). In March 2017, the Kenyan government declared
the drought a National Emergency.
This is not the rst time reduced rainfall has resulted in
a crisis in Kenya. The deadly 2010– 2011 La Niña-driven
drought is still fresh in the minds of many East Africans.
The Kenyan NDMA was established following the
20102011 drought to ensure a coordinated effort to
manage drought risk in the future.
* Correspondence to: P. Uhe, School of Geographical Sciences, Univer-
sity of Bristol, UK. E-mail: peter.uhe@bristol.ac.uk
Present address: School of Geographical Sciences, University of Bristol,
UK.
Kenya is highly vulnerable to drought. Most of the coun-
try (over 80%) is characterized as arid or semi-arid lands,
with annual rainfall less than 550 or 850 mm, respectively.
Under this denition, the arid and semi-arid lands house
the majority of all livestock in Kenya (70%) and around
30% of the population (Republic of Kenya, 2012). In addi-
tion to low rainfall, there is a wide range of year-to-year
variability in rainfall (see Section 2). For example, North
West (NW) Kenya, an arid region, had 500 mm of rain-
fall in 1997 but only 150 mm in 2000. This variability in
rainfall makes droughts a common occurrence. The areas
analysed in this study, which were affected by the 2016
drought (NW and South East (SE) Kenya), are in arid or
semi-arid regions.
The food insecurity in the current 20162017
drought was rstly attributable to the low rainfall
in the OctoberDecember (OND) 2016 ‘short rains’
(Figure 1(a)), particularly in the NW and the SE of Kenya.
In addition to the lack of rain in the OND season, some
areas including the SE were still dealing with the effects of
low rainfall from the previous March– May (MAM) 2016
‘long rains’. Accordingly, there is a large precipitation
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd on behalf of the Royal Meteorological Society.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in
any medium, provided the original work is properly cited.
P. UHE et al.
> 3.0
1
0.8
0.6
0.4
0.2
–0.2
–0.4
–0.6
–0.8
–1
(a) (b)
(c) (d)
2.5 – 3.0
2.0 – 2.5
1.5 – 2.0
1.0 – 1.5
0.5 – 1.0
–0.5 – 0.5
–1.0 – –0.5
–1.5 – –1.0
–2.0 – –1.5
–2.5 – –2.0
–3.0 – –2.5
< –3.0
N/A
Figure 1. Spatial maps over Kenya for: (a) OND 2016 relative precipitation anomalies from the CHIRPS data set (anomalies with respect to
1986– 2009), (b) January December 2016 relative precipitation anomalies from the CHIRPS data set (anomalies with respect to 1986– 2009),
(c) SPI index calculated for OND 2016, and (d) SPI index calculated for MAM 2016. The SPI plots were calculated based on CHIRPS precipitation
data using the GeoCLIM toolbox (https://earlywarning.usgs.gov/fews/software-tools/20). [Colour gure can be viewed at wileyonlinelibrary.com].
decit in the SE for the whole year (Figure 1(b)).
A common way to look at the severity of a drought
is the standardized precipitation index (SPI, McKee et al.,
1993), representing the cumulative probability of a rainfall
event. The SPI for OND (Figure 1(c)) shows drought con-
ditions in the NW and SE, and corresponds well with the
precipitation anomaly in Figure 1(a). The SPI in MAM
(Figure 1(d)) shows drought conditions in that season
were limited to the SE, and were more severe than in the
OND season in some regions. Other areas of East Africa,
including Somalia and Ethiopia, were also experiencing
drought conditions and food insecurity in early 2017,
however, for this study we focus on the conditions in
Kenya.
During 2016, starting in June and peaking in Novem-
ber, there was a strong La Niña event with a relative
NINO3.4 index of 1.3 C (this denition accounts for
global warming, see Section 3 for details). La Niña is
the negative phase of the El Niño - Southern Oscillation
(ENSO). La Niña events have a negative correlation with
OND rainfall in Kenya (Mutai and Ward, 2000; Nicholson
and Selato, 2000). There was also a negative dipole mode
index (DMI) during much of 2016. The DMI represents
the Indian Ocean dipole (IOD), which is the difference
between the sea surface temperatures (SSTs) in the west-
ern and eastern Indian Ocean (Saji et al., 1999). The DMI
is also correlated with OND East African rainfall, and a
negative dipole results in reduced rainfall in the eastern
sector of East Africa (Black et al., 2003).
There is also a strong correlation between ENSO and the
DMI, so it is difcult to separate their relative inuences.
However, it has been hypothesized that the East African
rainfall is more closely connected to the IOD, and the
inuence of ENSO is manifested through its link with
the IOD (Goddard and Graham, 1999; Black, 2005). IOD
events may occur in connection with an El Niño event, but
can be triggered independently from ENSO (Ashok et al.,
2003; Fischer et al., 2005), so although in 2016 there were
both negative ENSO and IOD events, this is not necessarily
the case more generally. For instance, in 2011 there was
a positive IOD index during La Niña, and above-average
OND rainfall in Kenya.
Recent studies have also indicated that different patterns
of the ENSO SST anomalies of the same phase can lead
to signicantly different teleconnections. Hoell et al.
(2014) separated La Niña events into different phases and
found that La Niña events characterized by cool central
Pacic SSTs and warm west Pacic SSTs have a different
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
inuence on the Indian Ocean and East African rainfall
compared to the canonical east Pacic La Niña pattern
(particularly in the MAM season). In addition, Preethi
et al. (2015) compared the inuence of the canonical El
Niño, El Niño Modoki, and IOD on African rainfall. They
found that El Niño Modoki and canonical El Niño have
opposite impacts, resulting in below and above mean East
African OND rainfall, respectively. ENSO clearly plays
an important role with respect to drought risk in East
Africa. However, the literature discussed shows that dis-
entangling the role of Pacic SSTs and the inuence of the
Indian Ocean, as well as potential role of anthropogenic
climate change is not straightforward and will likely not
be uniform in time and space.
Previous studies on drought in Kenya have found dif-
fering results when analysing the effect of anthropogenic
climate change. A number of studies have pointed out a
drying trend of the long rains in East Africa in recent
decades (e.g. Lyon and DeWitt, 2012) and there have been
a number of studies discussing whether this is due to inter-
nal climate variability or forced by human-induced climate
change. The drying trend is small compared to natural vari-
ability as noted by Yang et al. (2014), and this low signal to
noise makes it difcult to attribute the cause of this trend,
especially over a period as short as a few decades. This
discussion has also been complicated by an increase in pre-
cipitation predicted by climate simulations. For example,
Shongwe et al. (2011) used multi-model ensembles from
global climate models (GCMs) to show the whole rain-
fall distribution in East Africa is positively shifted in a
future climate: mean precipitation rates increase as does
the intensity of high rainfall, and droughts become less
severe. They describe physical mechanisms explaining this
trend. However, they also stress that natural variability is
so large that this trend will not become visible until well
after the beginning of this century.
However, while GCMs show an increase in precipita-
tion, atmospheric simulations forced by the observed pat-
terns of SST variability can reproduce the recent drying
of the long rains (Yang et al., 2014; Hoell et al., 2017).
The SST patterns are a result of both natural and anthro-
pogenic forcing, and using additional simulations, Hoell
et al. (2017) indicated that the interaction of internal vari-
ability and anthropogenic forcing may have enhanced the
drying trend compared to internal variability acting alone.
Rowell et al. (2015) also examined a number of hypothe-
ses for the drying trend not being present in coupled cli-
mate models, including changes to anthropogenic aerosols
and inadequate representation of physics in climate mod-
els. They also determined that it was unlikely the drying
trend could have resulted from natural variability alone.
A mechanism of drying over East Africa was proposed
by Liebmann et al. (2017), connecting increased convec-
tion over Indonesia to increased upper atmosphere easterly
winds and lower rainfall in MAM, but did not separate the
effect of anthropogenic forcing and natural variability.
Previous attribution studies have looked at the inuence
of human-induced climate change on specic droughts
in East Africa. Lott et al. (2013) used event attribution
methodologies on the 20102011 drought in Kenya and
Somalia. They found that the 2010 short rains failed due
to the La Niña event in 2010, however, human inuence
increased the probability of the dry 2011 long rains. In
contrast, Marthews et al. (2015) found no anthropogenic
inuence on the likelihood of low rainfall in the long
rains of 2014 in northern Kenya and southern Ethiopia, but
human inuences did increase temperatures and incoming
ground surface radiation and thus the factors exacerbating
the impacts of drought. We note that the results of these two
studies for the long rains are not necessarily contradictory
as they looked at different regions in East Africa and
different years (both studies used SST-forced attribution),
which may have different drivers of the drought. This
indicates attribution results for a particular drought cannot
be easily generalized to other droughts and highlights
the importance of a case-based approach to an attribution
analysis of individual events. It should also be noted that
model dependencies are very large for drought attribution,
as exemplied by the differing trends shown in Shongwe
et al. (2011), and shown thoroughly for Europe in Hauser
et al. (2017).
Another attribution study (Funk et al., 2013) examined
the ENSO-related SST effects rather than anthropogenic
inuence on the 2012 MAM rainfall decit in eastern
Kenya and southern Somalia. By repeating the analysis for
the time periods 20032012 and 19932002 they found
that while ENSO was a key driver of the dry spells in
19932002, effects other than ENSO contributed to the
dry spells in the more recent 20032012 period. However,
this may just reect natural variability and the inuence of
the Indian Ocean independent of ENSO.
This study aims to investigate the respective contribu-
tions of anthropogenic climate change and large scale vari-
ability in SSTs for a specic case – the low rainfall over
the OND season in NW Kenya and the yearly rainfall
decit in the SE Kenya. Focusing on a single OND sea-
son for the NW region allows us to more clearly examine
possible connections with the Pacic and Indian Oceans.
Analysing the whole year’s rainfall in the SE can be linked
to the accumulated impact of low rainfall for more than one
season. The analysis is performed on observational data
sets and multiple climate models with different experimen-
tal setups. Combining different approaches in this way,
gives us a range of possible responses of this event to
anthropogenic forcing and natural variability and hence
greater condence in the results.
For this study, we do not focus on the recent drying trend
of the long rains, as we are investigating changes due to
anthropogenic forcings in the order of a century rather than
decades. The focus here is on the short rains and yearly
rainfall rather than the long rains. There is not a signicant
drying trend reported in the short rains or yearly rainfall in
the regions analysed in this study.
Section 2 describes the data products and models used;
being observational data (Section 2.1), global climate
models (Section 2.2), and large ensembles of regional cli-
mate models (Section 2.3). Section 3 then gives a brief
description of the methods of analysis used and Section
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
4 gives an evaluation of the models against observations.
We show our analysis of the low precipitation in Section
5, which is again broken up into analysis of observations
(Sections 5.1 and 5.2), a comparison with the 20102011
drought (Section 5.3), analysis of the global climate mod-
els (Section 5.4) and the large ensembles of regional cli-
mate models (Section 5.5). In Section 6, we discuss other
factors in the drought such as temperature, and look at the
large ensembles of regional climate simulations to give
an indication of the inuence of climate change to avail-
able moisture (precipitation minus evaporation and also
soil moisture).
2. Data and models
Different regions in Kenya have different rainfall charac-
teristics, so we analyse a couple of smaller regions rather
than the country as a whole. The areas of interest for
this study are restricted to the regions in Kenya that had
the most signicant dry anomalies in 2016 (Figure 1). We
label these the NW and SE (see Figure 2). The regions
are dened by county boundaries to avoid selection bias
in choosing specic regions for this analysis and to assist
in making these regions relatable to people in Kenya. The
NW region includes the counties Turkana and Marsabit,
and the SE region includes Kwale, Kili, Mombasa,
and Lamu.
Kenya has seasonal rainfall dominated by two rainy
seasons: MAM and OND. The seasonal cycle of pre-
cipitation in the NW and SE are shown in Figures 3(a)
and (b), respectively. There is high year to year variabil-
ity, shown in the time-series of precipitation anomalies in
Figures 3(c) and (d) and also the 95% range in Figures 3(a)
and (b). The NW of Kenya is very dry with average rainfall
of less than 1 mm day1for much of the year. In the year
2016 it was especially dry in OND (Figures 1(a) and 3(c)).
Considering the SE of Kenya, the dry anomaly extended
over both rainy seasons (Figures 1(b) and 3(d)).
For the NW region, we analyse the short rains only
(OND), as this region had above average rainfall in the
2016 MAM rainy season and it is the rainfall decit that
is relevant to the drought’s impact. In the SE, we consider
the whole year from January to December 2016, as this
captures the impact of below average rainfall for two
consecutive rainy seasons.
2.1. Observational data
For observations, we use both station data and gridded
data of monthly precipitation. The station data for the
years 19812016 are provided by the Kenya Meteorolog-
ical Department. These time series are extended back in
time by the corresponding data sets in the monthly Global
Historical Climatology Network station database (Peter-
son and Vose, 1997). We analyse two stations in the NW
of Kenya for the OND season. These are Lodwar (3.10N;
35.60E, 515 m above sea level, 19202016) and Marsabit
(2.00N; 37.90E, 1447 m above sea level, 1918 2016).
The only station in the SE for which we have a long time
Lodwar
Mombasa
Figure 2. Map of Kenya highlighting the counties used in the analysis:
NW includes Turkana and Marsabit and SE includes Kwale, Mombasa,
Kili, and Lamu. Approximate location of the Marsabit, Lodwar, and
Lamu stations are shown as stars. [Colour gure can be viewed at
wileyonlinelibrary.com].
series is Lamu (2.27S; 40.90E, 30 m above sea level,
19062016). We analyse the station data of Lamu for
JanuaryDecember 2016 but note that this station is not
representative for the entire SE region.
The gridded data set is a combination of two data sets:
CHIRPS (Climate Hazards Group InfraRed Precipitation
with Station data; Funk et al., 2015b) and CenTrends (Cen-
tennial Trends; Funk et al., 2015a). CHIRPS is the state
of the art observational daily dataset for East Africa for
the years 19812016. CenTrends, on the other hand, is
a monthly data set, available for 1900– 2014. CenTrends
and CHIRPS are based on a similar assimilation tech-
nique and underlying observational data for their overlap
period. They are highly correlated, with correlations over
0.95, justifying the extension of the CenTrends data set
with monthly averaged data from CHIRPS. A correction
according to the regression between the two would result
in no difference in the NW region and slightly higher pre-
cipitation values in the SE region. To avoid the introduction
of extra errors we do not adjust this value, but bear in mind
that return period in the SE region might be slightly overes-
timated. With the two data sets together, referred to as the
CenTrends-ext data set, we can provide information about
both the trend in the past and the current situation.
2.2. Global climate models
We analyse output from two global climate models
(GCMs). The rst GCM is EC-Earth (Hazeleger et al.,
2010) which is a coupled atmosphereocean model with
a resolution of T159 (about 125 km over Kenya). The ver-
sion used is EC-Earth 2.3, which is based on the European
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
(a) (b)
(c) (d)
–3
–2
–1
0
1
2
3
4
5
201101 201201 201301 201401 201501 201601
CenTrends-ext KenyaSE anomalies
–2
–1
0
1
2
3
4
5
6
201101 201201 201301 201401 201501 201601
precipitation (mm day–1) Average precipitation (mm day–1)
Average precipitation (mm day–1)
precipitation (mm day–1)
CenTrends-ext KenyaNW anomalies
Figure 3. Seasonal cycle of CHIRPS monthly precipitation in: (a) NW Kenya from 1987 to 2009 and (b) SE Kenya from 1987 to 2009. For (a) and
(b) the solid line is the mean precipitation and the dashed lines are the 95% range for each month, (c) time-series of precipitation anomalies over the
past 7 years for the NW region, and (d) time-series of precipitation anomalies for the SE region. Blue/red means more/less than average precipitation
(panels (c) and (d) show anomalies with respect to 1986– 2009). [Colour gure can be viewed at wileyonlinelibrary.com].
Centre for Medium-Range Weather Forecasts (ECMWF)
seasonal forecasting model system 3 (Stockdale et al.,
2011). For EC-Earth, continuous simulations from 1860
to 2015 are used as per the CMIP5 historical setup until
2005 and the RCP8.5 scenario from 2006 (see Taylor
et al., 2012). The ensemble includes 16 members.
The second model used is UK Met Ofce model
HadGEM3-A (Christidis et al., 2013). In the EUro-
pean CLimate Extremes Interpretation and Attribution
(EUCLEIA) project, HadGEM3-A is an atmosphere-only
model and was run with N216 horizontal resolution (about
60 km). These simulations are also part of the Climate of
the Twentieth Century (C20C) Detection and Attribution
project. For this model we have both historical simulations
for 19602015, driven by observed SST, and historicalNat
simulations (Taylor et al., 2012) representing the histori-
cal period but with only non-anthropogenic forcings. The
SST for these runs has been obtained by subtracting an
estimate of the forced change in SST obtained from the
mean CMIP5 ensemble response. As the runs for 2016
were not yet available at the time of writing, we used the
trend up to 2015 as the indicator for the effects of natural
and anthropogenic forcings.
2.3. Large ensemble regional climate modelling
To obtain an extensive sample of possible weather under
different scenarios, we make use of the large ensem-
ble distributed computing framework of weather@home
(Massey et al., 2015). This uses the Met Ofce Hadley
Centre regional atmospheric circulation model HadRM3P,
at 50-km resolution over Africa, nested in the global
atmosphere-only model HadAM3P. With this model, thou-
sands of simulations are run for two scenarios: Actual and
Natural. The Actual simulations use current greenhouse
gas (GHG) and aerosol concentrations, and observed SSTs
and sea-ice extent from the OSTIA data set (Donlon et al.,
2012). The Natural simulations use preindustrial levels
of GHGs, multiple anthropogenic warming patterns sub-
tracted from the OSTIA SSTs (as per Schaller et al., 2016)
and the maximum observed sea-ice extent in the OSTIA
data set. In addition to simulations of 2016, we have a ref-
erence data set of Actual simulations from 1987 to 2009
(referred to as Climatology).
Different simulations were produced by varying the ini-
tial conditions. Forty starting conditions from previous
simulations were used in each scenario and different ini-
tial condition perturbations were applied to the potential
temperature to obtain thousands of unique initial condi-
tions, producing different simulations of possible weather.
A larger number of simulations were computed covering
just OND 2016, than were computed for the whole year
of 2016, due to computational constraints. The number of
ensemble members for each of the scenarios are: Actual:
3474 for OND and 2536 for JanuaryDecember; Natural:
7003 for OND and 4285 for JanuaryDecember; and Cli-
matology: 3596.
3. Methods
We used statistical methods to analyse the precipitation in
the NW and SE regions of Kenya. As mentioned above,
we considered precipitation from the observational data
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
sets CHIRPS, CenTrends and observations from stations
in Marsabit, Lamu, and Lodwar. The GCMs are EC-Earth
version 2.3 and HadGEM3-A, and the regional climate
model is weather@home.
In order to determine the return times of the low precip-
itation values we t the low tail of the observed and mod-
elled precipitation distributions to a generalized Pareto
distribution (GPD, Coles, 2001), which has a cumulative
distribution H:
H(x𝜇)=1(1𝜉(x𝜇)
𝜎)(1𝜉)
(1)
where 𝜇is the threshold, in this case chosen so that
the lowest 20% of data points are tted, 𝜎is the scale
parameter, and 𝜉is the shape parameter. To allow for a
trend in probability the threshold and shape parameters
are dependent on the 4-year running mean of global mean
temperature (T, also referred to as smoothed GMST) with
a trend 𝛼such that their ratio is constant and with an
exponential dependence that scales the whole PDF:
𝜇=𝜇0exp (𝛼T
𝜇0)
𝜎=𝜎0exp (𝛼T
𝜇0)
The trend 𝛼is tted together with the other parame-
ters in a maximum likelihood procedure. The t is con-
strained to have zero probability below zero precipitation
(𝜉<0, 𝜎<𝜇𝜉). Unphysically large shape parameters are
suppressed by a penalty term that keeps them roughly in
the range |𝜉|<0.4 (as per Schaller et al., 2014; van der
Wiel et al., 2017).
When calculating the distribution, the value of the year
of interest is not used in the t. As the GPD includes a
covariate (smoothed GMST) that varies with time, we can
evaluate the precipitation distribution for a given year (for
this study, 1920, and 2016). The precipitation series are
also shown twice, scaled with the tted trend to 2016 and
1920 GMST values. The difference in the two distributions
shows whether there is a trend, with the uncertainty range
estimated with a 1000-member nonparametric bootstrap.
This takes into account dependencies between ensemble
members of SST-forced models with a moving block
technique.
The EC-Earth and HadGEM3-A models are analysed
using the same method as for the observations, calculating
the return times and trend from the GPD t. These mod-
els have multiple long climate simulations, which reduces
the statistical uncertainty shown in the condence interval
bounds, compared to the observations. The models have
an additional structural uncertainty due to how the models
represent climate processes and this is not included in the
uncertainty estimates quoted in for each model. However,
the spread across the different models, with different rep-
resentations of the physics, gives an idea of this structural
uncertainty.
For the weather@home data, we take advantage of the
thousands of simulations to describe the distribution of
possible weather under different climate conditions, and
do not t an extreme value distribution to determine return
periods. The sampling uncertainty of return periods in
weather@home data is calculated by randomly resam-
pling the distribution 1000 times. Again we note that this
uncertainty represents the variability in the model, but not
the structural uncertainty due to model physics. Hence a
small uncertainty range does not indicate condence in
the results but indicates that the type of event is well sam-
pled using this methodology. Comparisons with the other
methods are necessary to assess condence in results. For
this study we do not bias correct the weather@home data.
Therefore, we use a threshold based on the observed return
period, instead of the observed magnitude of the event
which would rst require the data to be bias corrected.
In the gridded observational data sets we also analyse
the role of ENSO in this drought. As the teleconnection
to East African rainfall is related to the ENSO variabil-
ity and not to the trend, a detrended NINO3.4 index is
used for this analysis, see Philip et al. (2017) and van
Oldenborgh et al. (submitted). The detrended NINO3.4
index is dened in this study as SST in the NINO3.4
region (5S–5N, 120170W) minus SST averaged of
30S–30N to remove to rst-order effects of global
warming from the index. Detrending in this manner is
preferable to a linear detrending of the index itself, as
the warming signal is nonlinear. The 30S–30Nregion
was chosen to represent tropical SSTs as the difference
in NINO3.4 temperature and the wider tropical region is
important for ENSO. Correlations between monthly pre-
cipitation and the NINO3.4 index of the same month are
calculated. We then subtract a linear approximation of the
inuence of ENSO from the observational data for each
month in the 2016 event. This allows us to calculate the
return time for a hypothetical event for 2016 as if it had
happened under ENSO-neutral conditions. In order to keep
the precipitation values above zero we apply a logarithmic
transformation on the precipitation before subtracting the
inuence of NINO3.4.
We also look at correlations of the Kenyan precipitation
and Indian Ocean SSTs, represented by the DMI. The
DMI is comprised of the difference between the western
Indian Ocean (WIO) SSTs, 10S–10N, 50– 70E, and the
south eastern Indian Ocean SSTs, 100S, 90110E. The
DMI is not detrended as per NINO3.4 as it represents a
difference or gradient in the Indian Ocean SSTs. The WIO
index (detrended by subtracting the inuence of change in
GMST) is also considered as a separate index.
4. Evaluation of model precipitation
The gridded CenTrends-ext precipitation data is more
comparable to the model data than the station data so
is used to evaluate the model precipitation distributions.
For precipitation in NW Kenya in October– December
and SE Kenya in January– December, the GPD t of
the CenTrends-ext data gives a good description of the
dry tail.
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
Table 1. GPD t parameters for EC-Earth and HadGEM3-A, compared against CenTrends-ext.
𝜎/𝜇𝜉Bias correction factor
NW Kenya (October– December)
CenTrends-ext 0.25 (0.14, 0.31) 0.25 (0.33, 0.15)
EC-Earth 0.34 (0.30, 0.38) 0.41 (0.49, 0.34) 1.15
HadGEM3-A 0.19 (0.16, 0.23) 0.24 (0.38, 0.17) 1.33
SE Kenya (January– December)
CenTrends-ext 0.09 (0.05, 0.13) 0.10 (0.23, 0.06)
EC-Earth 0.11 (0.10, 0.14) 0.14 (0.26, 0.13) 0.83
HadGEM3-A 0.10 (0.08, 0.11) 0.23 (0.35, 0.09) 1.18
The parameters used for validation are the ratio of parameters 𝜎/𝜇(scale and location parameters) and 𝜉(shape parameter). Uncertainty ranges for
the parameters are shown in brackets. The bias correction factor calculated from the data is also shown.
(a) (b)
(c) (d)
Figure 4. Seasonal cycles of precipitation for 1987 –2009: (a) EC-Earth in NW Kenya, (b) EC-Earth in SE Kenya, (c) HadGEM3-A in NW Kenya, and
(d) HadGEM3-A in SE Kenya. Solid line shows the monthly mean precipitation and the dashed lines show the 95% range of monthly precipitation.
CHIRPS data shown for comparison – dots are the CHIRPS monthly mean and shaded area is the 95% range of monthly precipitation in CHIRPS.
[Colour gure can be viewed at wileyonlinelibrary.com].
The GPD ts of the monthly precipitation from the
EC-Earth and HadGEM3-A models were validated against
the ts for the CenTrends-ext precipitation distributions,
see Table 1. This table also includes the multiplicative bias
correction factor. The seasonal cycles for the models in
each region are also shown in Figure 4. The parameters
validated are the ratio of 𝜎and 𝜇(as it is assumed that the
scale parameter 𝜎scales with the position parameter 𝜇),
and also the shape parameter 𝜉. If the model parameters
are within the uncertainty bounds of the t parameters
in observations we consider this model for analysis and
employ a multiplicative bias correction for the model value
in 2016 if necessary.
The GPD t parameters, 𝜎/𝜇and 𝜉, of the EC-Earth
precipitation time series in the NW region, are respectively
just within and just outside the condence margins of the
t parameters given from observations. We include the
EC-Earth results for this region but keep in mind that this
is at the edge of the plausible range of distributions. The t
parameters for the SE region are well within the condence
margins from observations, so we trust this result, although
we have to remark that this region contains only three grid
boxes in this model. We also note that from Figure 4(a), the
short rains start earlier than the observations in NW Kenya.
The seasonal cycle is represented better in the SE, although
with an underestimation of rainfall during the long rains
(Figure 4(b)).
A statistical model evaluation of HadGEM3-A monthly
precipitation in the NW region shows that the t parame-
ters of the time series are within the uncertainty range of
the parameters tted from observations, so we include this
model in our analysis. The seasonal cycle is represented
well (Figure 4(c)), and as the historicalNat runs do not
show a trend over the whole period, this gives us greater
condence in the trend of the historical runs.
In the SE region the situation is more complicated. The
t parameters of the time series of SE Kenya precipita-
tion for HadGEM3-A in JanuaryDecember are within
the uncertainty range of the parameters of the observa-
tional distribution, and the seasonal cycle of rainfall is
represented well (Figure 4(d)). However, the historical-
Nat runs have a strong signicant trend towards drier
conditions. The 2016 low rainfall was 33 times more likely
to occur in 2016 than in 1920, based on the historicalNat
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
(a) (b)
(c)
Figure 5. (a) Map of climatological yearly precipitation bias over Kenya for weather@home. The bias is relative to CHIRPS over the 1987– 2009
period and expressed in percent. (b) and (c) are seasonal cycles of weather@home precipitation for 1987– 2009; (b) in NW Keyna and (c) in SE
Kenya. Solid line shows the monthly mean precipitation and the dashed lines show the 95% range of monthly precipitation. CHIRPS data shown
for comparison: dots are the CHIRPS monthly mean and shaded area is the 95% range of monthly precipitation in CHIRPS. [Colour gure can be
viewed at wileyonlinelibrary.com].
trend. To compare, the trend in the historical simulations
is about half of that (the 2016 event was 14 times more
likely to occur in 2016 than in 1920). Using a linear sub-
traction of the two trends indicates a wetting trend due to
anthropogenic inuences. However, as the model drift is
relatively large, we do not trust this linear subtraction and
do not consider this model further for the SE region.
Comparing the weather@home climatology to the
CHIRPS data set, the weather@home model does have
biases in rainfall, see Figure 5(a). The NW region has a
wet bias in the short rains (OND) of 0.8 mm day1.As
this region is particularly dry, this is a large percentage
bias relative to CHIRPS (75%). The SE has a dry bias
in the annual mean (1.3 mm day1) which is 51% drier
than CHIRPS. Because of this, we do not estimate return
times from the absolute precipitation values in the model.
However, we note that despite the absolute magnitude of
rainfall not being well captured, the model does replicate
the seasonal cycle reasonably well (Figures 5(b) and (c)).
During OND in the NW, the weather@home precipitation
matches the spread of precipitation in the CHIRPS data
set despite the average being too high. In addition, the year
to year variability of the weather@home OND rainfall,
forced by SSTs, shows a similar response to the observa-
tions in both regions. For the weather@home analysis, the
return periods from the CenTrends-ext data set were used
as the threshold for the 2016 event.
5. Precipitation analysis
5.1. Return times in station observations
The data for the precipitation stations at Marsabit in the
NW and Lamu in the SE are shown in Figure 6. Results
for Lodwar are not shown, as this time series did not t a
GPD properly and we were not able to calculate a return
time. Probably this is because this station is so dry that
it happens more often than not that there is hardly any
precipitation. It is very difcult to estimate the return time
of a value of almost zero. For the remaining two stations,
the data are plotted against the covariate used in the GPD
ts (smoothed GMST), showing the trends in the data
(Figures 6(a) and (b)). Return time plots of the data are
shown in Figure 6(c) and (d), with the distributions of the
data shown twice; shifted by the trend to 2016 levels and
1920 levels.
The return time of the OND 2016 event in Marsabit, in
the NW, is about 12 years (95% CI: 370 years) and the
trend is not signicant. The best estimate of the GPD t
for the return time of the 2016 event (January– December)
in Lamu, in the SE region, is about 2000 years, but the
GPD does not t the tail of the distribution very well. The
lower bound of the return time of 120 years would be a
better estimate for the occurrence of an event like this,
given the limited amount of data. This time series shows
a trend towards more precipitation (a ratio between the
return times of 2016 and 1920 of at most 0.4): in 1920 the
return period was about 50 years (95% CI: 5 150 years).
5.2. Return times in gridded observations
The return times of the CenTrends-ext data, shifted to 2016
and 1920 levels, are shown in Figure 7. The return time
of the OND 2016 event in NW region is about 3years
(95% CI: 1 –9 years) and there is no signicant trend.
The area-averaged drought is somewhat less extreme
than the drought measured by station data. For the SE
region, the return time of the 2016 JanuaryDecember
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
0
500
1000
1500
2000
2 5 10 100 1000 10000
(mm year–1)
Return period (year)
January-December summed precipitation Lamu-ext
1906:2015 (95% CI)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
–0.6 –0.4 –0.2 00.2 0.4 0.6 0.8 1
(mm year–1)
Global mean surface temperature (smoothed)
January-December summed precipitation p Lamu-ext
1906:2015 (95% CI)
0
100
200
300
400
500
600
700
800
2 5 10 100 1000 10000
(mm season–1)
Return period (year)
October-December summed precipitation Marsabit-ext
1918:2015 (95% CI)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
0
100
200
300
400
500
600
700
800
900
1000
(a) (b)
(c) (d)
–0.4 –0.2 0 0.2 0.4 0.6 0.8 1
(mm season–1)
Global mean surface temperature (smoothed)
October-December summed precipitation p Marsabit-ext
1918:2015 (95% CI)
Figure 6. Precipitation station data averaged over OND (Marsabit) or January–December (Lamu). (a) and (b) are precipitation anomalies plotted
against the change in global mean temperature (smoothed) for Marsabit and Lamu, respectively. The thick red line denotes the time-varying mean
of the data used in the GPD t (lowest 20%) and the thin lines are 1𝜎and 2𝜎below, respectively. The purple square shows the 2016 value, which
was not used in the t, and the two vertical red lines show the 95% condence interval of 𝜇for the climates of 1920 and 2016. (c) and (d) show the
return periods for the data for Marsabit and Lamu, respectively. The data are shown twice, shifted to the climate of 2016 with the tted trend (red
signs) and shifted to 1920 (blue signs). Lines representing the GPD and 95% condence interval of the GPD are also shown for the 2016 climate
(red lines) and the 1920 climate (blue lines). The observed value in 2016, not used in the t, is shown as a horizontal purple line. [Colour gure can
be viewed at wileyonlinelibrary.com].
precipitation was about 5 years (95% CI: 221 years), with
no signicant trend. This is much less extreme than in the
station data of Lamu. From Figures 1(a) and (b), we see
that the Lamu station is in the driest part of the SE region,
so we expected the area average to be less extreme than in
the station data. Lamu is one of the worst-affected areas,
although this station might not be representative for the
whole region due to coastal inuence or the patchy nature
of this drought.
In general, precipitation in Kenya in OND is positively
correlated with El Niño, which means that in an El Niño
year we expect more rain. As in 2016 we experienced La
Niña conditions, we expect this season to be drier than
under normal conditions. There is also a strong correlation
between precipitation and Indian Ocean SSTs (WIO and
DMI), especially in OND. The MAM precipitation is not
correlated signicantly with either NINO3.4 or Indian
Ocean SSTs in the NW or SE of Kenya.
For the NW region, the variance explained by NINO3.4
in OND is 16%. When we subtract the inuence of
NINO3.4 linearly from the logarithm of monthly pre-
cipitation, the OND average precipitation in 2016 in
an ENSO-neutral year would have been 0.89 mm day1
instead of 0.62 mm day1. In OND, the WIO explains
45% of the variance in NW Kenya precipitation, and
subtracting the inuence of WIO (detrended with GMST)
the average precipitation would have been 0.95 mm day1.
Here we have not disentangled the inuence of WIO and
NINO3.4 but point out that ENSO may be acting through
its teleconnection with the Indian Ocean. We calculated the
return time of the 2016 event as if it had happened under
these neutral ENSO conditions. This would have been a
relatively normal year, with a return period of 2 years (95%
CI: 13). We conclude that indeed, La Niña and the Indian
Ocean SSTs caused the difference between a normal year
and a dry year.
In the SE region the correlation with NINO3.4 is espe-
cially high in SON, where it explains about 17% of the
variance. Because the correlation during the rest of the
year is lower or not signicant, the variance explained by
NINO3.4 over the whole year is slightly lower, about 14%.
When we subtract the inuence of NINO3.4 from precip-
itation, the yearly average precipitation is 2.0 mm day1
instead of 1.9 mm day1(also 2.0 mm day1with WIO
inuence subtracted, which explains 3% of the variance
in precipitation). We also calculate the return time of the
2016 event in this region if it had happened under normal
ENSO conditions. It would have been a relatively dry year
even under normal conditions, with a return period of about
4 years (95% CI: 2 –11).
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
0
2
4
6
8
10
12
14
1 2 5 10 100 1000 10000
(mm season–1)
Return period (year)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
5
10
15
20
25
30
35
40
45
50
1 2 5 10 100 1000
10000
(mm year–1)
Return period (year)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
(a)
(b)
Figure 7. Same as for Figures 6(c) and (d) but for CenTrends-ext data
averaged over (a) NW Kenya for OND averages and (b) SE Kenya
for January– December averages. [Colour gure can be viewed at
wileyonlinelibrary.com].
5.3. Comparison with 20102011 drought
As an indication of the relative severity, we compare the
precipitation in the 2016 drought with the OctoberJune
2010 drought, which caused widespread famine. In
Marsabit, the return period of OND 2010 was 11 years
(95% CI: 340), compared to 12 years in 2016. So the
2016 drought is comparable to the OND part of the
20102011 drought, at this specic station.
In Lamu, the dry season October 2010June 2011, had
a return period of 80 years (95% CI: 5 –400 years), com-
pared to greater than 120 years in January– December
2016. However, in this t the year 2016 is used as well,
which makes the trend towards more rainfall not signif-
icant, highlighting the sensitivity of this analysis as the
t only includes the driest 20% years. Also, consider-
ing that we are comparing two different periods (October
2010June 2011 and JanuaryDecember 2016) and that
the estimate of the 2016 drought is not very well-dened
for this specic station, we are not able to rank these
droughts with condence.
For the CenTrends-ext data in the NW, the return period
for OND 2010 was about 60 years (95% CI: 4 –180 years),
compared to 3 years in 2016. So over this area, the
2016 drought was not unusual and the OND part of the
20102011 drought was much worse. In the SE, October
2010June 2011 had a return period of at least 50 years.
So this 20102011 drought was much worse than the
JanuaryDecember 2016 drought with a return period of
0
0.5
1
1.5
2
2.5
3
3.5
1 2 5 10 100 1000 10000
(mm day–1)
Return period (year)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
0.5
1
1.5
2
2.5
3
3.5
4
1 2 5 10 100 1000 10000
(mm day–1)
Return period (year)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
(a)
(b)
Figure 8. Same as for Figures 6(c) and (d) but for EC-Earth model
data averaged over (a) NW Kenya for OND averages and (b) SE
Kenya for January– December averages. [Colour gure can be viewed
at wileyonlinelibrary.com].
5 years, although we have to keep in mind that these con-
sider different seasons.
From this analysis, we conclude that the 20102011
drought was more severe than the 2016 drought on
the larger scale (in the gridded CenTrends-ext data
set), however, this does vary from place to place. For
example, Marsabit experienced conditions in 2016 similar
to 20102011, and Lamu was especially dry in 2016,
although the return period was not easily comparable to
20102011.
5.4. Global climate model analysis
Return periods for the GCMs EC-Earth and HadGEM3-A
were calculated using GPD ts for the OND season in
the NW and JanuaryDecember in the SE, as per the
observations. Values for the multiplicative bias correction
are given in Table 1.
In EC-Earth, the 2016 OND event in the NW had a
return period of about 6 years (95% CI: 49 years) (see
Figure 8(a)). There is no signicant trend in the model.
The correlation of OND precipitation in NW Kenya with
the detrended NINO3.4 index in EC-Earth is very low so
we are not able to calculate the inuence of El Niño in this
region.
Results for the SE in EC-Earth, indicate a return period
of the JanuaryDecember 2016 event of about 5 years
(95% CI: 45 years) (see Figure 8(b)). In this region we
see a small trend towards less precipitation: the ratio
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
0
0.5
1
1.5
2
2.5
3
3.5
4
1 2 5 10 100 1000 10000
(mm day–1)
Return period (year)
GPD >80% scale fit 1920
GPD >80% scale fit 2016
observed 2016
Figure 9. Same as for Figure 6(c) but for HadGEM3-A model data
averaged over NW Kenya for OND averages. [Colour gure can be
viewed at wileyonlinelibrary.com].
between 2016 and 1920 is 1.4 (95% CI: 1.12.1). This
means that in 1920 such an event would have happened
every 7 years instead of roughly every 5 years now.
The variance of SE Kenya precipitation explained
by the detrended NINO3.4 index in EC-Earth is about
10%, which is comparable to that in CenTrends-ext
data. We therefore calculate the return time of the
JanuaryDecember 2016 precipitation value in which the
inuence of NINO3.4 is subtracted, similar to the analysis
done in observations, with a multiplicative bias correction.
If the inuence of La Niña is subtracted, this increases the
rainfall, resulting in an event with return period of 3 years
(95% CI: 3 –4 years) instead of 5 years.
In the HadGEM3-A model, our results for the NW indi-
cate a return period of the 2016 OND event of about
33 years (95% CI: 17– 44 years) (see Figure 9). In this
region we see a small trend towards more precipitation: the
ratio between 2016 and 1920 is 0.23 (95% CI: 0.120.72).
This means that in 1920 such an event would have hap-
pened every 8 years (95% CI: 5 15 years) instead of once
every 33 years now.
The variance explained by the detrended NINO3.4 index
in OND is about 10%, which is slightly lower than in
CenTrends data. We calculate the return time of the OND
2016 precipitation value in which the inuence of the
detrended NINO3.4 is subtracted, similar to the analysis
done in observations, with a multiplicative bias correction.
Without La Niña, the event would have a magnitude that
occurs once in 5 years (95% CI: 3– 5 years), so would not
be particularly exceptional.
5.5. weather@home analysis
Figure 10 shows the weather@home distributions of pre-
cipitation for the two regions. These consistently show
that the 2016 simulations are drier than the climatology,
which can mainly be attributed to the SSTs (e.g. La Niña)
that occurred in 2016 compared to the average year. In
addition, the Actual 2016 simulations are slightly wetter
than the Natural 2016 in the NW for OND, indicating that
this event may have been made less likely due to climate
change.
Precipitation (mm day–1)Precipitation (mm day–1)
(a)
(b)
Figure 10. Return periods for precipitation in the weather@home model:
(a) NW Kenya in OND, (b) SE Kenya in January–December. For each
panel, three distributions are shown: Actual 2016 (red), Natural 2016
(blue), and Climatology 1987– 2009 (orange). Five to 95% condence
intervals from sampling error are shown by shading. [Colour gure can
be viewed at wileyonlinelibrary.com].
To quantify the changes in return time in the
weather@home model, we use the return time from
the CenTrends-ext data set to dene the event. From
Section 5.2, the CenTrends-ext 2016 precipitation had
return periods of 5 and 3 years, respectively, in the SE for
JanuaryDecember and NW for OND.
To determine the change in likelihood due to anthro-
pogenic inuence, we take the distribution from the Actual
2016 simulations as a reference and see how they changed
from the Natural 2016 simulations, for the 1 in 5-year
event in the SE and 1 in 3-year event in the NW. The 1
in 5-year occurrence of yearly rainfall in the SE is not
changed between the 2016 Actual and 2016 Natural sim-
ulations. The event in the NW has become less likely due
to climate change in the OND season, with a ratio between
Actual and Natural simulations of 0.76 (CI: 0.710.81).
We also note that the event has become also less likely due
to climate change in the SE for the OND season, compen-
sated by an increase of probability of drought in the MAM
season (not shown). This indicates that human-induced
warming is inuencing the seasons differently, as found
by Lott et al. (2013).
By comparing the return times between the climatol-
ogy and Actual 2016 simulations, we see the effect of
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
P–E (mm day–1)
P–E (mm day–1)
Soil moisture (kg. m–2)Soil moisture (kg. m–2)
(a) (b)
(c) (d)
Figure 11. Return time plots of weather@home simulations as per Figure 10 but for P– E in mmday1(a and c) and soil moisture in the top 10 cm
in kg m2(b and d). Upper panels are for the NW region in OND and lower panels are for the SE region in January– December. [Colour gure can
be viewed at wileyonlinelibrary.com].
the SST patterns of 2016 compared to a normal year. For
the SE, we take the 1 in 5-year threshold in the clima-
tology simulations. This event occurs most years with the
SSTs in Actual 2016, at least four times more likely (CI:
4.064.64). In OND in the NW, an event that occurs 1 in
3 years in the climatology simulations, occurs most years
in the Actual 2016 simulations and is at least 2 times as
likely (CI: 2.502.75). The event was thus mainly caused
by the SST patterns observed during OND 2016 which
include La Niña and IOD.
The correlation of precipitation in Kenya with SST
indices (DMI for Indian Ocean dipole and detrended
NINO3.4 for ENSO) is strong in the OND season, with
the DMI better correlated than NINO3.4 or WIO. These
are based on 3-month means against OND precipitation,
for 19872009 weather@home climatology simulations.
For precipitation in the SE, NINO3.4 explains 29% of the
variance in OND whereas the DMI explains 40% of the
variance. The correlations are not so strong in the NW,
with NINO3.4 explaining 4% of the variance or the DMI
explaining 21% of the variance. Teleconnections between
the different regions mean the indices are not independent,
but this shows the Indian Ocean SSTs are more closely
linked to Kenyan OND precipitation than Pacic SSTs in
this model.
We note that for OND 2016, the DMI was 0.5 (from the
OSTIA SSTs forcing the weather@home model), whereas
the NINO3.4 was 1.1. We also looked at lags in the cor-
relation between the SST indices and the weather@home
precipitation and found that the 12-month lagged
NINO3.4 has a slightly greater correlation for SE region
precipitation but for all other cases (DMI and NINO3.4
for NW Kenya) the current SST indices correlated higher
than the index from the previous months.
6. Other factors in the 2016 drought
Although there is no clear evidence as to the anthro-
pogenic effects on the Kenyan precipitation in this drought,
other factors such as temperature can play an important
role. For example, higher temperatures may result in an
increase in evapotranspiration and reduce moisture avail-
ability, impacting negatively on agriculture and worsening
the effects of the drought. In the OND season of 2016,
the NE was 0.8 CandtheSEwas0.7
C above normal
(using the MERRA data set, compared to 19812010).
These higher temperatures have been reported in connec-
tion with drier conditions and lower expected crop yields
(e.g. FEWSNet, 2017a).
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
To investigate a connection between higher temperatures
and moisture availability, we conducted an examination
of additional variables in the weather@home simulations.
Specically, we looked at the net precipitation minus evap-
oration (P E) and soil moisture in the top 10 cm of soil.
Firstly, we note that the simulations indicate a shift of the
temperature distribution in the Actual 2016 simulations,
compared to the Natural 2016 simulations, as expected due
to climate change. The Actual 2016 simulations were also
warmer than in the Climatology simulations, relating to the
particular conditions of 2016 (not shown). However, return
time plots of P E and soil moisture (Figure 11) show the
same trend as the precipitation with there being slightly
more moisture in the Actual 2016 simulations compared
to the Natural 2016 simulation, and the Climatology simu-
lations being wetter. This can be explained by the precipi-
tation limiting the moisture supply if there is no precipi-
tation, there cannot be much additional evaporation. We do
additionally note that due to the biases in the precipitation
we do not quantify the change in return periods for these
variables compared to the climatology.
We see that in the weather@home model, the mois-
ture availability is dominated by the precipitation and not
decreased signicantly due to temperature effects for the
2016 OND season in the NW and JanuaryDecember
2016 in the SE. However, as we do not have long time
series of measurements of variables such as soil mois-
ture or evaporation, we cannot easily evaluate the model
performance for these quantities. The model does have
biases in precipitation, and there may be inuences of
temperature on water availability that the model does
not capture well, such as the effect on water storages or
irrigated plots.
This analysis is a probe into possible factors impacting
the drought other than precipitation; however, without
corroborating evidence from other methods, we do not
make any strong conclusions based on these results. On the
other hand, we consider that higher temperatures would
also be expected to increase water demand and stress on
livestock (Rojas-Downing et al., 2017). So even without a
reduction in water availability, the effects of a drought in a
warming world may be exacerbated for specic industries.
This is a complicated system though, and there are also
possible opposing effects such as reductions in stomatal
conductance due to increased CO2levels acting to reduce
evapotranspiration. Only an attribution study integrated
with a realistic impact model will be able to draw rm
conclusions.
7. Conclusions
The low precipitation of the 2016 Kenyan drought in
the gridded observational data sets were not particularly
extreme in the regions analysed, with return periods of
5 years (SE, January– December) and 3years (NW, OND).
However, the 2016 rainfall decit at the specic stations
we looked at was more extreme. In Lamu, an event as
observed in 2016 is expected to occur less than once in
0.001 0.01 0.1 1 10 100
Kenya NW, October-December
CenTrends-ext
EC-Earth
HadGEM3-A
Weather@home
Average
0.001 0.01 0.1 1 10 100
Kenya SE, January-December
CenTrends-ext
EC-Earth
Weather@home
Average
0.001 0.01 0.1 1 10 100
Kenya stations
Ratio of return times
Marsabit, OND
Lamu, January
-December
Figure 12. Synthesis showing the range of possible values for the ratio
of return times between climate of 2016 and 1920 (or Natural climate
for weather@home). The upper plots are gridded data sets for OND in
NE Kenya, middle plots are gridded data sets for January– December in
SE Kenya, and the lower plots are the station observations. The average
ratios across the methods are also shown for each region, these are simple
averages weighting each method equally. Observations are shown in
blue, models in red and the average in purple. [Colour gure can be
viewed at wileyonlinelibrary.com].
120 years and in Marsabit around once in 12 years. This
indicates that although the 2016 event may not appear to be
an extremely dry year over the spatial average of rainfall,
individual locations may have had little to no rain.
There is no detectable change in the likelihood of low
rainfall like the drought event observed during the ‘short
rain’ period in 2016 in the NW of Kenya and over the
annual mean of 2016 in the SE region in Kenya. Models
conrm this, suggesting that anthropogenic climate change
did not result in a signicant trend in such a drought event.
We show the spread of the possible ratios of return times
between 2016 and pre-industrial in Figure 12. In these
plots, a ratio of 1 represents no change due to climate
change, less than 1 means the event has become less
likely due to climate change (a wetting trend), and greater
than 1 means the event has become more likely due to
climate change (a drying trend). The trends due to climate
change were not signicant for the majority of methods
considered. The averages (calculated from the gridded
observations and climate models) are consistent with 1 (no
change).
In comparison to anthropogenic climate change, the
year to year variability of rainfall due to factors such
as ENSO has a large inuence on a drought like this.
The year 2016 was a La Niña year, which is correlated
with lower rainfall in Kenya in the OND season. In terms
of variance of rainfall explained by the NINO3.4 index,
this is 16% in the NW (OND) and 14% in the SE for
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
P. UHE et al.
the CenTrends-ext data set (January– December). If the
inuence of ENSO on the rainfall decit is subtracted
from the 2016 event, this results in estimated rainfall
of 0.89 mm day1instead of the observed 0.62 mm day1
in the NE and 2.0 mm day1instead of 1.9 mm day1in
the SE. So the hypothetical event with neutral ENSO
conditions would have occurred slightly more frequently,
every 2 years in the NW and every 4 years in the SE. The
models that correctly reproduce the effects of La Niña
show comparable results. The weather@home ensemble
studies the effect of global SSTs on the probability of
drought, and nds that it is made two (NW) to four (SE)
times more likely by these SSTs.
The SST patterns, including the pattern associated with
La Niña, did increase the likelihood of this drought. Sea-
sonal forecasts now have some skill in predicting ENSO
months ahead, and hence can give advance warning of
the likelihood of low rainfall. This information is already
available in the form of seasonal outlooks provided by the
Kenya Meteorological Department for each rainy season;
this study highlights the benets of using this information
to determine the likelihood of drought conditions. Increas-
ing the quality and informed use of forecast data can be
very useful in reducing impacts of droughts. If govern-
ments, aid organizations, and the general population can
make decisions knowing the likelihood of drought condi-
tions, they may be able to prevent or at least reduce such
high levels of food insecurity.
Precipitation is not the only relevant variable though,
and higher than normal temperatures in 2016 might have
accelerated forage and water depletion across most of
the pastoral and marginal agricultural areas (FEWSNet,
2017b). This additional stressor has increased due to cli-
mate change and will have implications during future
extreme heat and drought events, as well as for livelihood
activities such as crop production. The lack of signicant
climate change signal for precipitation in the areas stud-
ied highlights the importance of planning for the large
year-to-year variations in climate. In addition, the dom-
inant inuence of predictable climate patterns, such as
La Niña, during the 20162017 drought implies a high
potential for using forecasts to trigger preventative actions,
before the onset of the meteorological event.
To conclude, there is no detectable change in the likeli-
hood of low rainfall like the drought in 2016, due to anthro-
pogenic climate change, as observed in the short rains in
NW Kenya and yearly rainfall in SE Kenya. We do note
that the timing of rainfall events is also important for agri-
culture. This study investigated change in seasonal aver-
aged rainfall, but it is also possible that there are changes
in timing, frequency, and duration of rainfall events within
a season due to climate change. Further investigation into
changes to rainfall over shorter time scales, and using
impact models will be benecial in future studies.
Acknowledgements
This study was conducted as part of the Raising Risk
Awareness project and the World Weather Attribution
activity coordinated by Climate Central. We would like to
thank the Kenya Meteorological Department for supply-
ing their observational data. For their technical expertise,
we would like to thank our colleagues at the Oxford eRe-
search Centre: A. Bowery, M. Rashid, S. Sparrow and D.
Wallom, and the Met Ofce Hadley Centre PRECIS team
for their technical and scientic support for the develop-
ment and application of weather@home. weather@home
simulations were simulated using computing resources
from the AWS Cloud Credits for Research program. This
work was supported by the EUCLEIA project, funded
by the European Union’s Seventh Framework Programme
[FP7/2007-2013] under grant agreement no. 607085.
References
Ashok K, Guan Z, Yamagata T. 2003. A look at the relationship between
the ENSO and the Indian Ocean dipole. J. Meteorol. Soc. Jpn. 81(1):
41– 56. https://doi.org/10.2151/jmsj.81.41.
Black E. 2005. The relationship between Indian Ocean sea-surface
temperature and East African rainfall. Philos. Trans. R. Soc. Lond. A
363(1826): 43– 47. https://doi.org/10.1098/rsta.2004.1474.
Black E, Slingo J, Sperber K. 2003. An observational study
of the relationship between excessively strong short rains in
coastal East Africa and Indian Ocean SST. Mon. Weather
Rev. 131: 74–94. https://doi.org/10.1175/ 1520-0493(2003)131
0074:AOSOTR2.0.CO;2.
Christidis N, Stott PA, Scaife AA, Arribas A, Jones GS, Copsey D,
Knight JR, Tennant WJ. 2013. A new HadGEM3-A-based system for
attribution of weather- and climate-related extreme events. J. Clim.
26(9): 2756– 2783. https://doi.org/10.1175/JCLI-D-12- 00169.1.
Coles S. 2001. An Introduction to Statistical Modeling of Extreme Values.
Springer Series in Statistics. Springer-Verlag: London, 208 pp.
Donlon CJ, Martin M, Stark J, Roberts-Jones J, Fiedler E, Wimmer
W. 2012. The operational sea surface temperature and sea ice anal-
ysis (OSTIA) system (Advanced Along Track Scanning Radiome-
ter(AATSR) Special Issue). Remote Sens. Environ. 116: 140– 158.
https://doi.org/10.1016/j.rse.2010.10.017.
FEWSNet. 2017a. East Africa seasonal monitor. http://www.fews
.net/sites/default/les/documents/reports/EA_Seasonal%20Monitor_
2017_01_06.pdf (accessed 30 March 2017).
FEWSNet. 2017b. Pastoral areas to experience worst food insecurity
outcomes as the dry season progresses. http://www.fews.net/east-
africa/kenya/key-message-update/january-2017 (accessed 30 March
2017).
Fischer AS, Terray P, Guilyardi E, Gualdi S, Delecluse P. 2005. Two
independent triggers for the Indian Ocean dipole/zonal mode in a
coupled GCM. J. Clim. 18(17): 3428– 3449. https://doi.org/10.1175/
JCLI3478.1.
Funk C, Husak G, Michaelsen J, Shukla S, Hoell A, Lyon B, Hoer-
ling MP, Liebmann B, Zhang T, Verdin J, Galu G, Eilerts G, Row-
land J. 2013. Attribution of 2012 and 2013– 12 rainfall decits in
eastern Kenya and southern Somalia [in “explaining extremes of
2012 from a climate perspective”]. Bull. Amer. Met. Soc. 94(9):
S45– S48.
Funk C, Nicholson S, Landsfeld M, Klotter D, Peterson P, Harrison
L. 2015a. The centennial trends greater horn of Africa precipitation
dataset. Dryad. https://doi.org/10.5061/dryad.nk78.
Funk C, Peterson P, Landsfeld M, Pedreros D, Verdin J, Shukla S, Husak
G, Rowland J, Harrison L, Hoell A, Michaelsen J. 2015b. The climate
hazards infrared precipitation with stations – a new environmental
record for monitoring extremes. Sci. Data 2(150): 066. https://doi.org/
10.1038/sdata.2015.66.
Goddard L, Graham NE. 1999. Importance of the Indian Ocean for
simulating rainfall anomalies over eastern and southern Africa. J.
Geophys. Res. 104(D16): 19099–19116.
Hauser M, Gudmundsson L, Orth R, Jézéquel A, Haustein K, Vautard
R, van Oldenborgh GJ, Wilcox L, Seneviratne SI. 2017. Methods
and model dependency of extreme event attribution: the 2015 Euro-
pean drought. Earth’s Future 5: 1034–1043. https://doi.org/10.1002/
2017EF000612.
Hazeleger W, Severijns C, Semmler T, ¸Stef ˘
anescu S, Yang S, Wang X,
Wyser K, Dutra E, Baldasano JM, Bintanja R, Bougeault P, Caballero
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
2016 KENYAN DROUGHT
R, Ekman AML, Christensen JH, van den Hurk B, Jimenez P, Jones C,
Kållberg P, Koenigk T, McGrath R, Miranda P, van Noije T, Palmer
T, Parodi JA, Schmith T, Selten F, Storelvmo T, Sterl A, Tapamo H,
Vancoppenolle M, Viterbo P, Willén U. 2010. EC-Earth: a seamless
earth-system prediction approach in action. Bull. Amer. Meteor. Soc.
91(10): 1357– 1363. https://doi.org/10.1175/2010BAMS2877.1.
Hoell A, Funk C, Barlow M. 2014. La Niña diversity and northwest
Indian Ocean rim teleconnections. Clim. Dyn. 43(9): 2707–2724.
https://doi.org/10.1007/s00382-014- 2083-y.
Hoell A, Hoerling M, Eischeid J, Quan X-W, Liebmann B. 2017.
Reconciling theories for human and natural attribution of recent east
Africa drying. J. Clim. 30(6): 1939– 1957. https://doi.org/10.1175/
JCLI-D- 16-0558.1.
KFSSG. 2017. The 2016 short rains season assessment report. http://
reliefweb.int/sites/reliefweb.int/les/resources/Kenya_2016_SRA_
NationalReport.pdf (accessed 30 March 2017).
Liebmann B, Bladé I, Funk C, Allured D, Quan XW, Hoerling M,
Hoell A, Peterson P, Thiaw WM. 2017. Climatology and interannual
variability of boreal spring wet season precipitation in the eastern
horn of Africa and implications for its recent decline. J. Clim. 30(10):
3867– 3886. https://doi.org/10.1175/JCLI-D-16- 0452.1.
Lott FC, Christdis N, Stott PA. 2013. Can the 2011 East African
drought be attributed to climate change? Geophys. Res. Lett. 40(6):
1177– 1181. https://doi.org/10.1002/grl.50235.
Lyon B, DeWitt DG. 2012. A recent and abrupt decline in the east African
long rains. Geophys. Res. Lett. 39(2): l02702. https://doi.org/10.1029/
2011GL050337.
Marthews TR, Otto FEL, Mitchell D, Dadson SJ, Jones RG. 2015, 2015.
The 2014 drought in the horn of Africa: attribution of meteorological
studies. Bull. Amer. Meteor. Soc. 17: S83–S88. https://doi.org/10
.1175/BAMS-D-15-00115.1.
Massey N, Jones R, Otto FEL, Aina T, Wilson S, Murphy JM, Hassell D,
Yamazaki YH, Allen MR. 2015. weather@home – development and
validation of a very large ensemble modelling system for probabilistic
event attribution. Q. J. Roy. Meteor. Soc. 141: 1528 1545. https://doi
.org/10.1002/qj.2455.
McKee TB, Doesken NJ, Kleist J. 1993. The relationship of drought
frequency and duration to time scales. In Proceedings of the 8th Con-
ference on Applied Climatology, Vol. 17, American Meteorological
Society, Boston, MA, 179–183.
Mutai CC, Ward MN. 2000. East African rainfall and the trop-
ical circulation/convection on intraseasonal to interannual
timescales. J. Clim. 13: 3915– 3939. https://doi.org/10.1175/1520-
0442(2000)013<3915:EARATT>2.0.CO;2.
NDMA. 2017. National drought early warning bulletin. http://
www.ndma.go.ke/resource-center/send/39- drought-updates/4116-
national-drought- early-warning- bulletin-march-2017 (accessed 30
March 2017).
Nicholson SE, Selato JC. 2000. The inuence of La Nina on African rain-
fall. Int. J. Climatol. 20: 1761– 1776. https://doi.org/10.1002/1097-
0088(20001130)20:14<1761::AID-JOC580>3.0.CO;2-W.
Peterson TC, Vose RS. 1997. An overview of the global histori-
cal climatology network temperature database. Bull. Am. Meteorol.
Soc. 78(12): 2837– 2849. https://doi.org/10.1175/1520-0477(1997)
0782837:AOOTGH2.0.CO;2.
Philip S, Kew SF, van Oldenborgh GJ, Otto F, O’Keefe S, Haustein K,
King A, Zegeye A, Eshetu Z, Hailemariam K, Singh R, Jjemba E,
Funk C, Cullen H. 2017. Attribution analysis of the Ethiopian drought
of 2015. J. Clim. https://doi.org/10.1175/JCLI-D-17-0274.1.
Preethi B, Sabin TP, Adedoyin JA, Ashok K. 2015. Impacts of the ENSO
Modoki and other tropical Indo-Pacic climate-drivers on African
rainfall. Sci. Rep. 5: 16653. https://doi.org/10.1038/srep16653.
Republic of Kenya. 2012. Releasing our full potential. Sessional paper
no. 8 of 2012 on national policy for the sustainable development of
northern Kenya and other arid lands, Ministry of State for Develop-
ment of Northern Kenya and other Arid Lands.
Rojas-Downing MM, Nejadhashemi AP, Harrigan T, Woznicki SA.
2017. Climate change and livestock: impacts, adaptation, and miti-
gation. Clim Risk Manag 16 (suppl C): 145– 163. https://doi.org/10
.1016/j.crm.2017.02.001.
Rowell DP, Booth BBB, Nicholson SE, Good P. 2015. Reconciling past
and future rainfall trends over east Africa. J. Clim. 28(24): 9768– 9788.
https://doi.org/10.1175/JCLI-D- 15-0140.1.
Saji N, Goswami B, Vinayachandran P, Yamagata T. 1999. A dipole
mode in the tropical Indian Ocean. Nature 401(6751): 360– 363.
Schaller N, Otto FEL, van Oldenborgh GJ, Massey NR, Sparrow S,
Allen MR. 2014. The heavy precipitation event of May– June 2013
in the upper Danube and Elbe basins [in “explaining extremes of
2013 from a climate perspective”]. Bull. Amer. Meteor. Soc. 95(9):
S69– S72.
Schaller N, Kay AL, Lamb R, Massey NR, van Oldenborgh GJ, Otto
FEL, Sparrow SN, Vautard R, Yiou P, Ashpole I, Bowery A, Crooks
SM, Haustein K, Huntingford C, Ingram WJ, Jones RG, Legg T, Miller
J, Skeggs J, Wallom D, Weisheimer A, Wilson S, Stott PA, Allen
MR. 2016. Human inuence on climate in the 2014 southern England
winter oods and their impacts. Nat. Clim. Chang. 6: 627– 634.
Shongwe ME, van Oldenborgh GJ, van den Hurk BJJM, van Aalst
MK. 2011. Projected changes in mean and extreme precipitation in
Africa under global warming. Part II: East Africa. J. Clim. 24(14):
3718– 3733. https://doi.org/10.1175/2010JCLI2883.1.
Stockdale TN, Anderson DLT, Balmaseda MA, Doblas-Reyes F, Fer-
ranti L, Mogensen K, Palmer TN, Molteni F, Vitart F. 2011. ECMWF
seasonal forecast system 3 and its prediction of sea surface temper-
ature. Clim. Dyn. 37(3): 455– 471. https://doi.org/10.1007/s00382-
010-0947- 3.
Taylor K, Stouffer RL, Meehl GA. 2012. An overview of CMIP5
and the experiment design. Bull. Amer. Meteor. Soc. 93: 485–498.
https://doi.org/10.1175/ BAMS-D-11-00094.1.
van der Wiel K, Kapnick SB, van Oldenborgh GJ, Whan K, Philip S,
Vecchi GA, Singh RK, Arrighi J, Cullen H. 2017. Rapid attribution
of the August 2016 ood-inducing extreme precipitation in south
Louisiana to climate change. Hydrol. Earth Syst. Sci. 21(2): 897– 921.
https://doi.org/10.5194/hess-21- 897-2017.
Yang W, Seager R, Cane MA, Lyon B. 2014. The east African long rains
in observations and models. J. Clim. 27(19): 7185– 7202. https://doi
.org/10.1175/JCLI-D- 13-00447.1.
© 2017 The Authors. International Journal of Climatology published by John Wiley & Sons Ltd Int. J. Climatol. (2017)
on behalf of the Royal Meteorological Society.
... Since the beginning of the COVID-19 pandemic in March 2020, the Alto Turiaçu Indigenous Territory (Maranhão state) has suffered from 'conflicts caused by invasions of loggers and traffickers'. 209 Kwaxipuhu, an Indigenous member of the Ka'apor community, was beaten to death on 3 July 2020 as a result of the situation. 210 186. ...
... 257 258 A day earlier, armed men had taken a boat which had been seized from prospectors by ICMBio inspectors and police officers during Operation Maracá . b) Regional impacts 209. The Climate Experts' Report explains in detail how intense deforestation results in changes to the local hydrological cycle, causing decreasing rainfall for surrounding regions. ...
... Meanwhile, some of the most catastrophic droughts in the world continue to occur in East Africa 208 . Though no single drought there has been linked directly to climate change, this is likely due in part to a relatively short observational record and high natural variability, especially for precipitation [209][210][211] . More generally, the drying of the major rainy season in the region, the 'long rains' 212 , is likely connected to climate change 213,214 . ...
... In particular, rainfall in Kenya during short rainy season is highly correlated with El Nino, which implies more rain is expected during an El Nino year. It is expected that this season (OND) should be drier than usual, as we witnessed La Nina conditions in 2016 (Uhe et al., 2018). ...
... Between 1975 and 2018 alone, five drought events occurred in the latter seven years (2005/2006, 2008/2009, 2010/2011, 2015/2016, 2016/2017) compared to only seven events in former 30 years (1975 and 2004). The 2015 and 2016 drought over most parts of Ethiopia related to El Nino conditions , while 2016 and 2017 drought over the Greater Horn related to La Nina (Uhe et al. 2018). Further work by Preethi et al. (2015) indicates that the canonical El Nino and the El Nino Modoki are associated with depressed rainfall southern hemispheric regions and enhances the rainfall in northern areas of East Africa during March to May (MAM) season. ...
Article
Full-text available
The Turkana low-level jet stream (TJ) is important to climatic conditions over northern Kenya and East Africa. The representation of the TJ in climate models varies due to the TJ interaction with Turkana channel that is influenced by model resolution and influences the model representation of the regional climate. This study compares features of the TJ in CMIP6 AMIP model simulations with ERA5. Models reveal climatological wind speeds that match those of the reanalysis from the ERA5 at the jet entrance (13 m/s) but lower magnitudes of wind speed and vertical shears compared to ERA5 within the Turkana channel. The models with slowest wind speeds, have a flattened Turkana channel and fail to exhibit the terrain constriction at 37° E which otherwise aids in accelerating winds to form a jet core. Furthermore, they fail to represent the narrowing of the channel as in ERA5, thereby forming blocking walls in the channel, forcing vertical ascent and mixing, and weakening shear. This boosting of ascent motion promotes rainfall formation and enhances wet anomalies at the exit of the TJ when the jet stream is weaker. By applying a new narrowing index, we demonstrate the need to improve topography details in the CMIP6 models, particularly those with resolution coarser than 1.5°, in order to properly simulate the TJ and the observed rainfall over the northwestern areas of eastern Africa.
... The abrupt transitions from drought to flood experienced in Ethiopia and Kenya in the last 5 years underscore the need to understand societal and physical processes characterizing these drought-to-flood events. Accordingly, we analyzed the years 2017-2018, in which a severe drought (Funk et al., 2019;Philip et al., 2018;Uhe et al., 2018) was followed by widespread floods (Kilavi et al., 2018;Njogu, 2021) . The time series of the monthly anomaly standardized precipitation index (dark green lines at the bottom of the graphs) were computed from Climate Hazards group Infrared Precipitation with Stations data. ...
Article
Full-text available
Disaster risks are the results of complex spatiotemporal interactions between risk components, impacts and societal response. The complexities of these interactions increase when multi-risk events occur in vulnerable contexts characterized by ethnic conflicts, unstable governments, and high levels of poverty, resulting in impacts that are larger than anticipated. Yet, only few multi-risk studies explore human-environment interactions, as most studies are hazard-focused, consider only a single-type of multi-risk interaction, and rarely account for spatiotemporal dynamics of risk components. Here, we developed a step-wise, bottom-up approach, in which a range of qualitative and semi-quantitative methods was used iteratively to reconstruct interactions and feedback loops between risk components and impacts of consecutive drought-to-flood events, and explore their spatiotemporal variations. Within this approach, we conceptualize disaster risk as a set of multiple (societal and physical) events interacting and evolving across space and time. The approach was applied to the 2017–2018 humanitarian crises in Kenya and Ethiopia, where extensive flooding followed a severe drought lasting 18–24 months. The events were also accompanied by government elections, crop pest outbreaks and ethnic conflicts. Results show that (a) the highly vulnerable Kenyan and Ethiopian contexts further aggravated drought and flood impacts; (b) heavy rainfall after drought led to both an increase and decrease of the drought impacts dependent on topographic and socio-economic conditions; (c) societal response to one hazard may influence risk components of opposite hazards. A better understanding of the human-water interactions that characterize multi-risk events can support the development of effective monitoring systems and response strategies. © 2022 The Authors. Earth's Future published by Wiley Periodicals LLC on behalf of American Geophysical Union.
... Meanwhile, some of the most catastrophic droughts in the world continue to occur in East Africa (Gebremeskel et al 2019). No single drought there has been linked directly to climate change, partly due to a relatively short observational record, high uncertainties and high natural variability, especially for precipitation (Uhe et al 2018, Philip et al 2018b, Kew et al 2021. There is limited evidence that anthropogenic warming of Western Pacific sea surface temperatures may contribute to more frequent drought (Funk 2012, ...
Article
Full-text available
Extreme event attribution aims to elucidate the link between global climate change, extreme weather events, and the harms experienced on the ground by people, property, and nature. It therefore allows the disentangling of different drivers of extreme weather from human-induced climate change and hence provides valuable information to adapt to climate change and to assess loss and damage. However, providing such assessments systematically is currently out of reach. This is due to limitations in attribution science, including the capacity for studying different types of events, as well as the geographical heterogeneity of both climate and impact data availability. Here, we review current knowledge of the influences of climate change on five different extreme weather hazards (extreme temperatures, heavy rainfall, drought, wildfire, tropical cyclones), the impacts of recent extreme weather events of each type, and thus the degree to which various impacts are attributable to climate change. For instance, heat extremes have increased in likelihood and intensity worldwide due to climate change, with tens of thousands of deaths directly attributable. This is likely a significant underestimate due to the limited availability of impact information in lower- and middle-income countries. Meanwhile, tropical cyclone rainfall and storm surge height have increased for individual events and across all basins. In the North Atlantic basin, climate change amplified the rainfall of events that, combined, caused half a trillion USD in damages. At the same time, severe droughts in many parts of the world are not attributable to climate change. To advance our understanding of present-day extreme weather impacts due to climate change developments on several levels are required. These include improving the recording of extreme weather impacts around the world, improving the coverage of attribution studies across different events and regions, and using attribution studies to explore the contributions of both climate and non-climate drivers of impacts.
... The results obtained from the historical analysis of scP-DSI based on EOF approach captured the major historical spatial-temporal characteristics of drought over Isiolo County, Kenya, of 1983Kenya, of -1984Kenya, of , 1992Kenya, of -1993Kenya, of , 1999Kenya, of -2000Kenya, of , and 2009Kenya, of -2011. The results are coherent with the previous studies done by Balint et al. (2013), Shilenje et al. (2019), Uhe et al. (2018), and Wambua et al. (2018) and, therefore, can be used as an early warning tool to trigger drought emergency response and contingency planning based on the results of the RCP 4.5/8.5 projected drought scenarios. The projected drought events are anticipated to cause adverse effect on the livestock production. ...
Article
Full-text available
This study was to determine the spatiotemporal characteristics of historical and projected drought events throughout Isiolo County, Kenya, through using self-calibrating Palmer Drought Severity Index (scPDSI). The historical scPDSI was computed at monthly timescale using a 39-year long monthly mean precipitation, monthly average temperature, and climatological (1950-1996) soil available water holding capacity. The projected scPDSI under representative concentration pathways (RCPs) was computed using bias corrected monthly temperature and precipitation model output data from Coordinated Regional Climate Downscaling Experiment (CORDEX). ScPDSI runs have shown that more severe drought events dominated the period between was identified under RCP4.5 to be the most severe drought lasting for 25 months, while under RCP8.5 run, Nov 2042-Nov 2046 was identified as the most severe drought with the duration of water stress anticipated to last for 49 months. The empirical orthogonal function (EOF) analysis results indicated that the two-leading eigenvectors accounted for over 85% of the spatial variability for both historical and projected droughts under the RCPs. Subsequently, the Mann-Kendall (MK) test was applied to the projected scPDSI, temperature, and precipitation time series in order to determine the local expected drought trends. The MK test of the identified significant increase in trend for temperatures under RCP8.5 and precipitation under RCP4.5 towards the end of the last decade under the study period is considered. Both scenarios showed a decline in trends of drought events in Isiolo County from 2020 to 2050.
Article
The changing probabilities of extreme climate and weather events, in terms of frequency, intensity, spatial extent, duration, and timing is one of the most noticeable and damaging manifestations of human-induced climate change. During the March-April-May (MAM) rainfall season of 2012, 2016 and 2018, Kenya experienced high rainfall that caused both widespread and localised flooding, resulting in human and livestock deaths, destruction of infrastructure and property, bursting of riverbanks, submerging of farmlands and emergence of isolated cases of water-borne diseases. Here, we aim to quantify how the magnitude of heavy rainfall during these seasons may have been altered by human-induced climate change. We undertake a probabilistic attribution analysis using three different approaches utilising two observational datasets and two independent climate model experiment set-ups. We analyse three different seasonal heavy rainfall indices, maximum consecutive 5-day, 10-day, and 20-day rainfall, to compare the magnitude of maxima recorded in MAM 2012, 2016 and 2018 with the magnitude of maxima in a pre-industrial climate (with little or no anthropogenic influence). We find a shift towards intensification of extreme rainfall in today's climate, although these increases are not in all cases statistically distinguishable from our estimates of magnitudes in the preindustrial climate. Although we find no significant anthropogenic climate change influence, the intensification of extreme rainfall amid the observed drying trend and the projected increases in rainfall in the MAM season in Kenya, leave the already vulnerable societies with uncertainties about how to prepare for a changing climate. This study, therefore, provides a basis for an in-depth assessment of current and future trends of extreme rainfall in East Africa in adapting to changing climate risks for sustainable development in the already vulnerable and less resilient society.
Article
Climate change is leading to changing patterns of precipitation and increasingly extreme global weather. There is an urgent need to synthesize our current knowledge on climate risks to water security, which in turn is fundamental for achieving sustainable water management. Climate Risk and Sustainable Water Management discusses hydrological extremes, climate variability, climate impact assessment, risk analysis, and hydrological modelling. It provides a comprehensive interdisciplinary exploration of climate risks to water security, helping to guide sustainable water management in a changing and uncertain future. The relevant theory is accessibly explained using examples throughout, helping readers to apply the knowledge learned to their own situations and challenges. This textbook is especially valuable to students of hydrology, resource management, climate change, and geography, as well as a reference textbook for researchers, civil and environmental engineers, and water management professionals concerned with water-related hazards, water cycles, and climate change.
Chapter
This study sought to establish the occurrence and magnitude of droughts in semiarid Eastern Kenya. The study utilized gridded and in situ rainfall data sets for the period 1973–2013. Majority of droughts in Eastern Kenya are moderate. Stations close to Mt Kenya region experienced more droughts during the MAM season, while those in the lowlands had the highest number of droughts during the OND season. The occurrence rate of moderate drought is higher during OND (18%) than during MAM (10%). The lowlands of Tharaka-Nithi and Kitui counties had the highest occurrence rate of severe droughts during MAM and OND, respectively. There was no discernible pattern in occurrence of drought between gridded and in situ data during MAM. However, there was a variation in the occurrence rate of OND droughts between in situ and gridded data. There is a need to tap into the potential of MAM given the season's lower vulnerability to drought.
Article
Pastoralists in sub-Saharan Africa are particularly vulnerable to changing climatic conditions and heavily affected by an increasing frequency of severe droughts. To increase drought resilience, livestock index insurance constitutes a promising tool that is aligned to mitigate covariate risks. Uptake rates for these products are, however, relatively low. This article uses choice-experimental data collected in Northern Kenya to analyse preferences for hypothetical livestock index insurance contracts among pastoralists. Results show that pastoralists prefer lower strike levels as well as transparency in the form of regular radio announcements and text messages informing the policyholder about the location-specific index readings. Highly valued are trust enhancing features such as an index certification. We do not find any indication that the option to split the premium payment into two installments as a way to relax potential liquidity constraints could lead to higher demand.
Article
Full-text available
In northern and central Ethiopia, 2015 was a very dry year. Rainfall was only from one-half to three-quarters of the usual amount, with both the "belg" (February-May) and "kiremt" rains (June-September) affected. The timing of the rains that did fall was also erratic. Many crops failed, causing food shortages for many millions of people. The role of climate change in the probability of a drought like this is investigated, focusing on the large-scale precipitation deficit in February-September 2015 in northern and central Ethiopia. Using a gridded analysis that combines station data with satellite observations, it is estimated that the return period of this drought was more than 60 years (lower bound 95% confidence interval), with a most likely value of several hundred years. No trend is detected in the observations, but the large natural variability and short time series means large trends could go undetected in the observations. Two out of three large climate model ensembles that simulated rainfall reasonably well show no trend while the third shows an increased probability of drought. Taking the model spread into account the drought still cannot be clearly attributed to anthropogenic climate change, with the 95% confidence interval ranging from a probability decrease between preindustrial and today of a factor of 0.3 and an increase of a factor of 5 for a drought like this one or worse. A soil moisture dataset also shows a nonsignificant drying trend. According to ENSO correlations in the observations, the strong 2015 El Niño did increase the severity of the drought.
Article
Full-text available
A stationary low pressure system and elevated levels of precipitable water provided a nearly continuous source of precipitation over Louisiana, United States (US), starting around 10 August 2016. Precipitation was heaviest in the region broadly encompassing the city of Baton Rouge, with a 3-day maximum found at a station in Livingston, LA (east of Baton Rouge), from 12 to 14 August 2016 (648.3 mm, 25.5 inches). The intense precipitation was followed by inland flash flooding and river flooding and in subsequent days produced additional backwater flooding. On 16 August, Louisiana officials reported that 30 000 people had been rescued, nearly 10 600 people had slept in shelters on the night of 14 August and at least 60 600 homes had been impacted to varying degrees. As of 17 August, the floods were reported to have killed at least 13 people. As the disaster was unfolding, the Red Cross called the flooding the worst natural disaster in the US since Super Storm Sandy made landfall in New Jersey on 24 October 2012. Before the floodwaters had receded, the media began questioning whether this extreme event was caused by anthropogenic climate change. To provide the necessary analysis to understand the potential role of anthropogenic climate change, a rapid attribution analysis was launched in real time using the best readily available observational data and high-resolution global climate model simulations. The objective of this study is to show the possibility of performing rapid attribution studies when both observational and model data and analysis methods are readily available upon the start. It is the authors' aspiration that the results be used to guide further studies of the devastating precipitation and flooding event. Here, we present a first estimate of how anthropogenic climate change has affected the likelihood of a comparable extreme precipitation event in the central US Gulf Coast. While the flooding event of interest triggering this study occurred in south Louisiana, for the purposes of our analysis, we have defined an extreme precipitation event by taking the spatial maximum of annual 3-day inland maximum precipitation over the region of 29–31° N, 85–95° W, which we refer to as the central US Gulf Coast. Using observational data, we find that the observed local return time of the 12–14 August precipitation event in 2016 is about 550 years (95 % confidence interval (CI): 450–1450). The probability for an event like this to happen anywhere in the region is presently 1 in 30 years (CI 11–110). We estimate that these probabilities and the intensity of extreme precipitation events of this return time have increased since 1900. A central US Gulf Coast extreme precipitation event has effectively become more likely in 2016 than it was in 1900. The global climate models tell a similar story; in the most accurate analyses, the regional probability of 3-day extreme precipitation increases by more than a factor of 1.4 due to anthropogenic climate change. The magnitude of the shift in probabilities is greater in the 25 km (higher-resolution) climate model than in the 50 km model. The evidence for a relation to El Niño half a year earlier is equivocal, with some analyses showing a positive connection and others none.
Article
Full-text available
An observation-based analysis and large simulation ensembles show no evidence that climate change made heavy precipitation in the upper Danube and Elbe basins in May-June, such as observed in 2013, More likely.
Article
Science on the role of anthropogenic influence on extreme weather events, such as heatwaves or droughts, has evolved rapidly in the past years. The approach of "event attribution" compares the occurrence-probability of an event in the present, factual climate with its probability in a hypothetical, counterfactual climate without human-induced climate change. Several methods can be used for event attribution, based on climate model simulations and observations, and usually researchers only assess a subset of methods and data sources. Here, we explore the role of methodological choices for the attribution of the 2015 meteorological summer drought in Europe. We present contradicting conclusions on the relevance of human influence as a function of the chosen data source and event attribution methodology. Assessments using the maximum number of models and counterfactual climates with pre-industrial greenhouse gas concentrations point to an enhanced drought risk in Europe. However, other evaluations show contradictory evidence. These results highlight the need for a multi-model and multi-method framework in event attribution research, especially for events with a low signal-to-noise ratio and high model dependency such as regional droughts.
Article
The 1981-2014 climatology and variability of the March-May eastern Horn of Africa boreal spring wet season are examined using precipitation, upper- and lower-level winds, low-level specific humidity, and convective available potential energy (CAPE), with the aim of better understanding the establishment of the wet season and the cause of the recent observed decline. At 850 mb, the development of the wet season is characterized by increasing specific humidity and winds that veer from northeasterly in February to southerly in June and advect moisture into the region, in agreement with an earlier study. Equally important, however, is a substantial weakening of the 200-mb climatological easterly winds in March. Likewise, the shutdown of the wet season coincides with the return of strong easterly winds in June. Similar changes are seen in the daily evolution of specific humidity and 200-mb wind when composited relative to the interannual wet season onset and end, with the easterlies decreasing (increasing) several days prior to the start (end) of the wet season. The 1981-2014 decrease in March-May precipitation has also coincided with an increase in 200-mb easterly winds, with no attendant change in specific humidity, leading to the conclusion that, while high values of specific humidity are an important ingredient of the wet season, the recent observed precipitation decline has resulted mostly from a strengthening of the 200-mb easterlies. This change in the easterly winds appears to be related to an increase in convection over the Indonesian region and in the associated outflow from that enhanced heat source.
Article
Global demand for livestock products is expected to double by 2050, mainly due to improvement in the worldwide standard of living. Meanwhile, climate change is a threat to livestock production because of the impact on quality of feed crop and forage, water availability, animal and milk production, livestock diseases, animal reproduction, and biodiversity. This study reviews the global impacts of climate change on livestock production, the contribution of livestock production to climate change, and specific climate change adaptation and mitigation strategies in the livestock sector. Livestock production will be limited by climate variability as animal water consumption is expected to increase by a factor of three, demand for agricultural lands increase due to need for 70% growth in production, and food security concern since about one-third of the global cereal harvest is used for livestock feed. Meanwhile, the livestock sector contributes 14.5% of global greenhouse gas (GHG) emissions, driving further climate change. Consequently, the livestock sector will be a key player in the mitigation of GHG emissions and improving global food security. Therefore, in the transition to sustainable livestock production, there is a need for: a) assessments related to the use of adaptation and mitigation measures tailored to the location and livestock production system in use, and b) policies that support and facilitate the implementation of climate change adaptation and mitigation measures.
Article
Two theories for observed East Africa drying trends during March-May 1979-2013 are reconciled. Both hypothesize that variations in tropical sea surface temperatures (SSTs) caused East Africa drying. The first invokes a mainly human cause resulting from sensitivity to secular warming of Indo-west Pacific SSTs. The second invokes a mainly natural cause resulting from sensitivity to a strong articulation of ENSO-like Pacific decadal variability involving warming of the west Pacific and cooling of the central Pacific. Historical atmospheric model simulations indicate that observed SST variations contributed significantly to the East Africa drying trend during March-May 1979-2013. By contrast, historical coupled model simulations suggest that external radiative forcing alone, including the ocean’s response to that forcing, did not contribute significantly to East Africa drying. Recognizing that the observed SST variations involved a co-mingling of natural and anthropogenic effects, we diagnosed how East African rainfall sensitivity was conditionally dependent on the interplay of those factors. East African rainfall trends in historical coupled models were inter-compared between two composites of ENSO-like decadal variability, one operating in the early 20th century before appreciable global warming and the other in the early 21st century of strong global warming. We find the co-action of global warming with ENSO-like decadal variability can significantly enhance 35-yr East Africa drying trends relative to when the natural mode of ocean variability acts alone. A human-induced change via its interplay with an extreme articulation of natural variability may thus have been key to Africa drying; however, our results are speculative owing to differences among two independent suites of coupled model ensembles.
Article
A stationary low pressure system and elevated levels of precipitable water provided a nearly continuous source of precipitation over Louisiana, United States (U.S.) starting around 10 August, 2016. Precipitation was heaviest in the region broadly encompassing the city of Baton Rouge, with a three-day maximum found at a station in Livingston, LA (east of Baton Rouge) from 12–14 August, 2016 (648.3 mm, 25.5 inches). The intense precipitation was followed by inland flash flooding and river flooding and in subsequent days produced additional backwater flooding. On 16 August, Louisiana officials reported that 30,000 people had been rescued, nearly 10,600 people had slept in shelters on the night of 14 August, and at least 60,600 homes had been impacted to varying degrees. As of 17 August, the floods were reported to have killed at least thirteen people. As the disaster was unfolding, the Red Cross called the flooding the worst natural disaster in the U.S. since Super Storm Sandy made landfall in New Jersey on 24 October, 2012. Before the floodwaters had receded, the media began questioning whether this extreme event was caused by anthropogenic climate change. To provide the necessary analysis to understand the potential role of anthropogenic climate change, a rapid attribution analysis was launched in real-time using the best readily available observational data and high-resolution global climate model simulations. The objective of this study is to show the possibility of performing rapid attribution studies when both observational and model data, and analysis methods are readily available upon the start. It is the authors aspiration that the results be used to guide further studies of the devastating precipitation and flooding event. Here we present a first estimate of how anthropogenic climate change has affected the likelihood of a comparable extreme precipitation event in the Central U.S. Gulf Coast. While the flooding event of interest triggering this study occurred in south Louisiana, for the purposes of our analysis, we have defined an extreme precipitation event by taking the spatial maximum of annual 3-day inland maximum precipitation over the region: 29–31º N, 85–95º W, which we refer to as the Central U.S. Gulf Coast. Using observational data, we find that the observed local return time of the 12–14 August precipitation event in 2016 is about 550 years (95 % confidence interval (C.I.): 450–1450). The probability for an event like this to happen anywhere in the region is presently 1 in 30 years (C.I. 11–110). We estimate that these probabilities and the intensity of extreme precipitation events of this return time have increased since 1900. A Central U.S. Gulf Coast extreme precipitation event has effectively become more likely in 2016 than it was in 1900. The global climate models tell a similar story, with the regional probability of 3-day extreme precipitation increasing due to anthropogenic climate change by a factor of more than a factor 1.4 in the most accurate analyses. The magnitude of the shift in probabilities is greater in the 25 km (higher resolution) climate model than in the 50 km model. The evidence for a relation to El Niño half a year earlier is equivocal, with some analyses showing a positive connection and others none.