All posts by chris

Forecast for below normal Somali Gu rains based on January sea surface temperatures

Chris Funk, Amy McNally, Greg Husak, Laura Harrison, Shraddhand Shukla, and Nicholas Novella

Figure 1. July-December Noah Soil Moisture Ranks for 2010 and 2016

 

 

Here we present yet another statistical forecast for below normal East African March-May rains. As predicted, October-December East African rainfall was below normal – in fact extreme dryness was experienced in Somalia, and the Somaliland area of Ethiopia (WFP, FEWS NET). At a global scale, the world faces unprecedented emergency food assistance needs, with four countries – Nigeria, South Sudan and Yemen facing a credible risk of famine. For Somalia, the Inter-Agency Working Group on Disaster Preparedness for East and Central Africa has released an urgent call for action, and the Somalia Food Security and Nutrition Analysis Unit warns of possible famine in Somalia. The FSNAU-FEWS NET countrywide seasonal assessment conducted in December 2016 and made public on February 2nd finds that 2.9 million people will likely face Crisis and Emergency (IPC Phases 3 and 4) through June 2017.

To help identify drought conditions associated with protracted stress on livestock, the FEWS NET science team has analyzed six-month (July-December) soil moisture estimates (Figure 1). These estimates were produced at NASA using the Noah land surface model, CHIRPS precipitation, and MERRA2 surface forcings (air temperature, etc.). The outputs suggest that conditions in parts of coastal Kenya, Somalia and the Somaliland region of Ethiopia are among the driest on record (over the 1982-2016 time period), and as bad or worse than 2010. The failed 2010 season was the beginning of famine that killed more than 250,000 Somalis; half of these were young children . Looking at soil moisture, as opposed to just rainfall, sparse rainfall (a, b, c).

Figure 2. Time series of standardized July-December Somalia soil moisture for the region highlighted in Figure 1.

Extracting and plotting (Figure 2) standardized soil moisture values from region identified with blue polygon in Figure 1 (basically Somalia plus far-eastern Ethiopia), we can see that for this large area, the past six months have been the driest since 1982, due to the combined influence of low rainfall above normal temperatures. We will focus on forecasts for this region.

We now turn to forecasting the March-May 2017 Gu rains, based on January 2017 NOAA optimum interpolation sea surface temperatures. This posting will be broadly similar to the last several outlooks. The process of predicting East African long rains based on observed January sea surface temperatures has been examined specifically by FEWS NET scientists to provide support for food security situations like we currently face in the Horn of Africa. The results presented here, however, focus on Gu rains in the Somalia region shown in Figure 1.  For continuity with our previous blog posts, we also focus on seasons with cool or neutral central equatorial Pacific sea surface temperatures (1984, 1986, 1989, 1990, 1996, 1997, 1999, 2000, 2008, 2009, 2011, and 2012). These were years identified in our December 5th post, based on Niño 4 sea surface temperatures. In this study we use these years to calculate correlations and build a simple regression-based forecast.

Figure 3. Correlation between Somalia Gu rainfall and January sea surface temperatures. Correlations screened for significance at p=0.1.

Figure 3 shows the correlation between March-May Gu rains in our study region (Somalia) and January sea surface temperatures for years with cool or neutral central equatorial Pacific sea surface temperatures. Pink boxes denote regions selected as predictors: the West Pacific, Central Pacific, and South Pacific. In our previous forecast we selected the North Pacific, as opposed to South Pacific as a predictor. That forecast was based on different region and time-period. Dark yellow lines also mark the characteristic ‘Western V’ structure often associated with recent East African long rains droughts. The Western and Southern Pacific regions appear to be the strongest controls of the Somalia Gu season during years with neutral or cold central Pacific temperatures. As discussed in the Appendix below warm ocean conditions in these boxes help drive subsidence over the central Pacific and Indian Ocean. As second negative teleconnection region appears to the southeast of Madagascar. Warm ocean conditions here may slow the northward progression of the long rains. This region was not selected as a predictor, because the influence of this region has not been investigated in depth.

Figure 4. Observed January NOAA Optimum Interpolation sea surface temperatures, expressed as standardized departures from a 1981-2010 baseline.

Figure 4 shows the actual observed January 2017 sea surface temperature conditions. Very warm (+1.5 standardized anomalies) sea surface temperatures are found in the Western V regions.  This  ‘Western V’ pattern associated with the West Pacific Warming Mode produces dry conditions over East Africa. Slightly cool conditions are found in the Central Pacific. Note that very warm temperatures are also found to the southeast of Madagascar. Figures 5-7 show scatterplots for each of the individual predictor regions for years with neutral or cool central equatorial Pacific sea surface temperatures. Based on the individual Western Pacific and Southern Pacific predictors, the upcoming Gu season might be very dry (<-1Z). Based on the Central Pacific alone, Gu conditions might be near normal.

Figure 5. Scatterplot of West Pacific January sea surface temperatures and March-May Somalia rainfall. Star indicates associated prediction for 2017.
Figure 6. Scatterplot of South Pacific January sea surface temperatures and March-May Somalia rainfall. Star indicates associated prediction for 2017.
Figure 7. Scatterplot of Central Pacific January sea surface temperatures and March-May Somalia rainfall. Star indicates associated prediction for 2017.

We next combine our predictors in a multivariate cross-validated regression (Figure 8). Given the small number of degrees of freedom two predictors were used: the average of the west and south Pacific boxes, and the east Pacific sea surface temperatures. The regression slope indicated a stronger forcing for the average of the west and south Pacific boxes: -2.2Z per °C versus +0.6Z per °C. Each green circle shows one cross-validated hindcast estimate. In general the model does a good job identifying most drought years. When the model predicted below normal rains, below normal rains typically occurred (5 out of 8 times). The direct regression forecast from our model was very low (-1.3Z) – and we feel it is hard to justify such a negative outlook given the anticipated transition out of La Niña-like conditions. A more cautious approach is taken, in which we assume that all the years with below normal forecasts are reasonable analogs. We then use the mean of the observed rainfall for these years as our forecast (-0.8Z) and the standard deviation of these below normal forecast years as our standard error estimate. Our forecast and 80% confidence intervals are shown with a red circle and cross-hatch in Figure 8. The distribution associated with this forecast would indicate an 80% chance of below normal (<0Z) rainfall and a 20% of above normal (>+0Z). There would a 50% chance of poor or very poor rains (<-0.8Z).

Figure 8. Cross-validated Somalia Gu hindcasts (green circles) based on January sea surface temperatures. Red star indicates 2017 Gu forecast and 80% confidence intervals (-0.8Z±1.1Z).

Figure 9 summarizes our results. Somalia Gu rains have experienced a substantial decline. During seasons with neutral and cool Central Pacific sea surface temperatures (marked with stars), ‘Western V’-like sea surface temperatures in January (Figure 3) tend to be followed by below normal Gu rains. Not every drought fits this pattern (1991, 2005), but the evidence analyzed suggests that below normal rains for 2017 are likely. Since 1999, 12 out the last 18 Gu seasons have been below normal (<0), our model suggests that current Pacific sea surface temperatures resemble the conditions preceding many of these below normal years.

Figure 9. Time series of standardized Somalia Gu rainfall (bars). Stars mark prior seasons with cool or neutral central Pacific sea surface temperatures. Black circles and lines indicate 2017 prediction (-0.8Z±1.1Z).

It should also be noted, however, that some dynamic forecast models predict a transition out of La Niña-like conditions before June of 2017. Sea surface temperature conditions could change dramatically over the next three months. A poor start to the rainy season, followed by improved conditions might be associated with a transition to warmer temperatures in the eastern Pacific. At present, (Fig. 10) January rainfall anomalies continue to indicate strong subsidence over East Africa and the Indian Ocean. This pattern seems likely to persist.

Fig. 10. January 2017 CHIRP rainfall anomalies (Source USGS/Climate Hazards Group/USGS).

 Appendix

This technical appendix looks briefly at current (past 30 day) climate conditions associated with current (January) sea surface temperatures. It is not intended for a general audience. On the other hand, understanding what is driving the current climate state provides support for our statistical projection. We are currently seeing a strong dipole in the amount of water vapor over East Africa and the Indian Ocean versus the Indo-Pacific Warm Pool. We also find low water vapor totals along the equator near the dateline. Our statistical model essentially predicts a continuation of this pattern for the Gu season. Associated with this dipole are strong equatorial westerly low level wind anomalies over the equatorial Indian Ocean – advecting atmospheric moisture away from East Africa and towards Indonesia. At the dateline we also see easterly zonal anomalies. This is characteristic of mild La Niña-like conditions, but the rest of the eastern equatorial Pacific looks non-La Niña-like. Presumably these  wind anomalies and water vapor conditions are associated with the observed January sea surface temperature conditions (Figure 4). The effect of warm sea surface temperatures in the equatorial West Pacific is well understood – warm tropical waters produce warm atmospheric conditions associated with low pressure, which can help draw in low level winds. More interesting might be the role played by warm south Pacific temperatures. Climatologically, upper level winds to the northeast of New Zealand move from west to east, curving north towards the equator to meet similar southerly flows from north of the equator. These converging flows produce upper level convergence and subsidence, reducing precipitation over the central Pacific. This helps to define the structure of the Walker Circulation. Over the past 30 days the upper level wind anomalies exhibit upper level ridging to the northeast of New Zealand and northwest of Hawaii that has enhanced north-south winds that converge near equator and dateline. We conjecture that this upper level riding is associated with a barotropic response to the very warm sea surface temperatures to the northeast of New Zealand and northwest of Hawaii. We can look at upper level velocity potential anomalies to see how this might associated with changes in the Walker Circulation. Purple areas in this figure are associated with upper divergence. Red regions are associated with upper level convergence. Upper level convergence will associated with subsidence and reduced precipitation, and this is what we see over Somalia and the central Pacific. We might interpret our statistical model to suggest that if temperatures in the West Pacific and South Pacific predictor regions remain substantially warmer than those over the central Pacific, atmospheric conditions will likely continue to exhibit the dry East Africa Indian Ocean/Wet Warm Pool/Dry Central Pacific pattern that has dominated the past four-to-five months.

A 2016 Climate Review and Continued Concerns for the 2017 Long Rains

Chris Funk

2016 in review – Exceptional El Niño and La Niña-like climate conditions

This posting will briefly review some aspects of 2016 climate and update our outlook for the 2017 March-June long rains (we remain pessimistic, although the outlook has improved somewhat).

2016 appears very likely to be the hottest year on record. Figure 1 shows annual NASA GISS global air temperature anomalies. The last point (2016) is based on January-November anomalies, since December data are not yet available. It seems that the massive 2015/16 El Niño helped release heat from the sub-surface of the ocean, resulting in the 2016 global air temperature anomaly (~+1°C) being substantially warmer than the previous 2001-2010 average (~+0.6°C). Ed Hawkins provides a very compelling representation o this warming spiral here. NASA also has a compelling new animation of OCO-2 satellite-based CO2 observations.

Figure 1. NASA GISS global air temperature anomalies.

Will this warming continue as a step up to a new ~+1°C ‘normal’ or will we return to more modest conditions in 2017? We will have to wait and see. What we do know, however, is that warming is already influencing food security. Somini Sengupta’s New York Times article discusses how ‘heat, hunger and war are forcing African’s onto a road on fire’, helping to fuel a dangerous and difficult increase in migration in the Sahel. Grace et al.’s 2015 paper identifies strong direct links between warming air temperatures and increased frequencies of low birthweight babies in Africa; very warm air temperatures can have direct negative impacts on mothers and infants. Warming likely increases risks of civil war in Africa and has non-linear negative impacts on economic production.

Recent FEWS NET research, however, has also highlighted another important way that global warming affects food security – by increasing the warmth of local Indo-Pacific sea surface temperatures. ‘Local’ in this sense means that we are seeing global warming manifest as localized pockets of very warm sea surface temperatures. This has important implications for climate prediction and drought early warning.

Pockets of localized very warm sea surface temperatures can enhance both El Niño-like and La Niña-like climate patterns. To illustrate this, we have prepared the following animation based on standardized January-February-March sea surface temperature from a single climate change simulation produced using the Canadian Earth System Model version 2, obtained from the Climate Explorer. The animation shows 1980-2030 sea surface temperatures, expressed as standardized anomalies from the 1961-1990 mean, with exceptionally warm (p<2.5%) regions identified with contours. The stretch in the image coloring goes from -2.5 to +2.5 standardized deviations. The point here is that as we enter the 21st century, we see pockets of very warm sea surface temperatures. Some of these anomalies look like El Niños, with warm conditions in the eastern Pacific. Some of these anomalies look like the West Pacific Warming Mode, with enhanced sea surface temperatures in the western Pacific and eastern Indian Ocean. Some look like the first half of 2016 (El Nino). Some look like conditions in October-November-December of 2016 (La  Nina plus West Pacific Warming Mode).

Our animation of a single climate change simulation can be compared to a much less extreme animation based on the average of a large multi-model ensemble, also obtained from the Climate Explorer. The stretch and contouring here is the same as in the individual model. In this view of climate change, we tend to see a fairly even warming of the global oceans, and relatively few dramatic regions of warming until the 2020s. This animation provides a valuable assessment of how climate change may influence the oceans, on average, but this multi-model ‘average’ will not be what we will experience in any giver year. Rather, what we will experience will be sea surface temperature variations that resemble a single climate change simulation (i.e. like this animation) – which seems likely to produce stronger El Niño-like and La Niña-like climate conditions.

2015 and 2016 may,  unfortunately, have provided us with examples of such El Niño-like and La Niña-like variations. During 2015, as successfully predicted by global climate models and NOAA, we saw the onset of an exceptionally strong El Niño that helped produce severe droughts in Ethiopia, Southern Africa, and many other regions. Our article in the Bulletin of the American Meteorological Society’s (BAMS) annual climate attribution issue makes the case for anthropogenic enhancement of the 2016/16 El Niño to increased aridity and reduced runoff in Ethiopia and Southern Africa. In Southern Africa, current FEWS NET assessments indicate that the ‘worst affected areas of Zimbabwe and southern Somalia are expected to be in Emergency (IPC Phase 4) acute food insecurity during the lean season’, while ‘areas of Southern Madagascar are in Crisis (IPC Phase 3) during the lean season as households face larger gaps in their basic food needs’. Other chapters of the BAMS special issue find increases in heat, humidity, dryness, drought, cyclones, wildfires and tidal floods, and decreases in arctic sea ice and cold in the northeastern United States.

During the summer of 2016 we experienced a rapid transition from El Niño to La Niña-like conditions. The west Pacific and eastern Indian Ocean warmed dramatically while the the eastern equatorial Pacific cooled. These conditions formed the basis of our accurate forecast for below normal East African October-November-December rains. Recent FEWS NET assessments document these very poor rains in Somalia, Eastern Kenya and Southeastern Ethiopia and describe the related high levels of food insecurity – ‘Emergency (IPC Phase 4) is likely in some areas of Somalia, and among some households in Ethiopia, while Crisis is expected for other areas of Somalia, southeastern Ethiopia, and northeastern Kenya through May 2017.’ Current FEWS NET reports identify Crisis conditions in northern Somalia, and vegetation conditions there look very poor (Figure 2), with current conditions very similar to 2010. If East Africa experiences poor 20 17 long (March-June) rains, food security in parts of Somalia, Ethiopia and Eastern Kenya are likely to deteriorate further, potentially resulting in conditions similar to 2010/11, which saw severe food emergencies in Kenya and Ethiopia and famine conditions in Somalia. The rest of this blog posting provides an outlook for the 2017 March-May long rains.

Figure 2. eMODIS NDVI for northern Somalia. Image obtained from earlywarning.usgs.gov.

A below normal long rainy season seems likely, given current sea surface temperatures

This section of our blog updates the forecast we provided in December, using November-December and just December sea surface temperatures, as opposed to the October-November December sea surface temperature data we used last month. For brevity, we do not recreate all the analysis provided last month, but merely update our results.

Before presenting our statistical forecast, we briefly review current conditions, based on the anomalies for last 30 days from NCEP/NCAR reanalysis data, provided by NOAA’s Earth System Research Laboratory. Figure 3 shows the past 30 day near-surface wind anomalies. Figure 4 shows anomalies of precipitable water (the total amount of water vapor in the atmosphere). Over the Indian Ocean and central Pacific the wind anomalies (Figure 3) continue to show a strong equatorial pattern associated with increased moisture convergence over the Indo-Pacific Warm Pool and decreased precipitable water over Eastern Africa (Figure 4). Over the eastern Indian Ocean the wind anomalies are very large (~7 ms-1). Over the western Indian Ocean and eastern East Africa we also find large (~-7 kgm-2) reductions in total precipitable water. CHIRP and NASA TMPA precipitation data (not shown) also show this strong dipole between East Africa/Western Indian Ocean and the Warm Pool region.

Figure 3. Near surface wind anomalies for the past 30 days.
Figure 4. Total precipitable water anomalies for the past 30 days.

We next examine the sea surface temperatures associated with this anomalous circulation. We use both November-December and December surface temperatures to highlight possible changes in climate conditions, since the Climate Prediction Center anticipates a transition towards ENSO-neutral conditions by early February. Figure 5 shows both November-December and December NOAA Extended Reconstruction sea surface temperature anomalies, expressed as standardized deviations from a 1981-2010 baseline. Both November-December and December exhibit very warm conditions in the warm pool and extra-tropics, with a characteristic ‘western V’ pattern combined with modestly cool central Pacific anomalies. This pattern appears to be driving the anomalous winds and moisture conditions shown in Figures 3 and 4.

Figure 5. November-December and December sea surface temperature anomalies.

This sea surface temperature pattern is similar to the West Pacific Warming Mode. Figure 6 shows a time series of West Pacific Warming Mode index values, based on November-December data. What we see is that recent El Niño events are often followed by periods of high West Pacific Warming Mode index values, with current October-November West Pacific Warming Mode conditions similar to 2010.

Figure 6. Time series of West Pacific Warm Mode index values. Vertical black lines denote recent El Nino events.

As presented last month, we next examine (Figure 7) a scatterplot of the average November-December Warm Pool and North Pacific sea surface temperatures versus central Pacific sea surface temperatures. This analysis is shown for neutral and cool ENSO seasons. 2016 is shown with a red circle. The current conditions are pretty unique. What we see is a combination of very warm Warm Pool/North Pacific sea surface temperatures combined with just moderately cool east Pacific sea surface conditions. These unique conditions make it hard to identify analog seasons. The National Multi-Model Ensemble forecasts (Figure 8) for February to April predict a continuation of precipitation dipole between the East Africa/Indian Ocean and Warm Pool regions, similar to what we are seeing in the observations. The March-April-May ensemble climate forecast (not shown) has a similar dipole, but with a weaker intensity.

Figure 7. Scatterplot of November-December Warm Pool/North Pacific and Eastern equatorial Pacific sea surface temperature anomalies.
Figure 8. The National Multi-Model Ensemble February-March-April precipitation forecast.

We next repeat the cross-validated forecast procedure used last month, but based on both November-December and December sea surface temperature fields. Warm Pool, North Pacific and central Pacific sea surface temperatures are used as predictors. We also limit ourselves to making forecasts for just ENSO-neutral and cool seasons. It is very important to note that the March-June data used to train this forecast is just for part of East Africa – southeastern Ethiopia, southern Somalia and eastern Kenya. Hence this forecast only targets a limited region of the Horn, a region strongly influenced by equatorial Indo-Pacific climate anomalies and the Somali Jet.

Figure 9. Cross-validated forecasts of March-June East African rainfall.

The green dots in Figure 9 show our cross-validated forecasts based on November-December observations. The overall cross-validated skill is modest (R2=0.36), but the ability to distinguish between wet and dry seasons is good. The 2017 forecast based on November-December sea surface temperatures is -0.8Z±1.2Z (red star), where ‘Z’ denotes a standardized departure from normal. This suggests that a below normal season is likely but not all that certain. To capture some of the uncertainty associated with the current climate state, which has seen modest cooling over the last month in the warm Warm Pool and North Pacific regions, and modest warming over the cool Central Pacific region, we also present a forecast based on just December sea surface temperatures. This forecast is also below normal (-0.5Z,  blue star), but is less pessimistic than results obtained from November-December. For the -0.8Z forecast, the probability of rainfall being less than -0.5 is 60%. For the -0.5Z forecast, this probability would be 50%.

In summary, while current conditions are hard to interpret due to the unique conditions (Figure 7), we are currently seeing a strong overturning East Africa/Warm Pool circulation anomaly (Figures 3 and 4) that climate models predict lasting into February-March-April (Figure 8). Our statistical forecasts anticipate below normal long rains, and this outlook appears substantiated by current circulation anomalies. We will continue to update our analysis on a monthly basis.  As we transition out of La Niña-like conditions, we may see Warm Pool precipitation decrease, leading to a more favorable outlook for East Africa. For now, however, a below normal outlook seems warranted. One plausible scenario might be a poor start to the season, with normal rains later in the season.

 

Below normal forecast for the 2017 East African long rains

Chris Funk, Greg Husak, Diriba Korecha, Gideon Galu and Shraddhand Shukla

Here we present an empirical forecast of below normal rainfall for the East African 2017 March-June (MAMJ) long rains. This forecast is based on statistical relationships between March-June rainfall in eastern East Africa (38-50°E, 5°S-8°N) and October-November Sea Surface Temperature (SST) observations in the Indo-Pacific. FEWS NET research has shown that dynamic models in this region and season lack skill, we therefore advocate using an outlook based on FEWS NET’s numerous statistical and diagnostic analyses focused on understanding and predicting the long rains (a, b, c, d, e, f).

As predicted based on September SSTs and October Kenya-Somalia rainfall, much of eastern East Africa received very poor October-November 2016  short rains and in many regions vegetation conditions are extremely poor. For the southeastern highlands region of Ethiopia, current dry conditions follow four poor rainy seasons during the 2015 and 2016 Belg and Kiremt rainy  seasons. For southern Somalia and eastern Kenyan crop growing areas, poor short rain harvests in 2016 follow a poor 2016 March-June cropping season. Current vegetation conditions in our focus region are very degraded (Figure 1). A poor 2017 March-June long rains could lead to yet another poor harvest in eastern Kenya, south-eastern Ethiopia and southern Somalia while delaying any substantial respite for pastoralists.

Figure 1. eMODIS NDVI anomalies and focus region.
Figure 1. eMODIS NDVI anomalies and focus region.

Eastern East Africa has experienced more frequent MAMJ droughts since 1995 (Figure 2) due to warm conditions in the Indo-Pacific warm pool, cool La Nina-like conditions in the eastern Pacific, and Pacific Decadal Oscillation-like SST variations in the northwest Pacific. FEWS NET research has suggested that warming in the Eastern Indian and Western Pacific Ocean has led to stronger March-June droughts during La Nina-like seasons (a, b).

In this study we focus on recent years with cool or neutral October-November eastern equatorial Pacific SSTs. During these years strong predictive relationships allow us to anticipate below normal rains for the following March-June rainy season. During El Niño-like years, predictability for the long rains tends to be low. While El Niño can bring above normal rains, some El Niño events, like 1992, were associated with drought and famine conditions. Excluding these El Niño seasons produces more coherent prediction results.

Beginning in 1980, we identify 16 cool-neutral October-November seasons, based on Niño 4 SSTs: 1980, 1981, 1983, 1984, 1985, 1988, 1989, 1995, 1996, 1998, 1999, 2000, 2007, 2008, 2010, and 2011. However, as shown below (Table 1), the selection of years has relatively little impact on our prediction of below normal rains. Selecting cool seasons or using all the years after 1980/81 produce similar pessimistic (below normal) forecasts for 2017.

Figure 2. Eastern East Africa standardized rainfall anomalies [Z scores]. Based on a 1981-2014 baseline.
Figure 2. Eastern East Africa standardized rainfall anomalies [Z scores]. Based on a 1981-2014 baseline.
Figure 3 shows a map of the correlation between October-November SSTs during neutral-Nina years and March-June East African rainfall the following spring. This is a familiar pattern we have seen associated with recent long rains droughts. A curve of negative correlations (~-0.7) begins in the north-central Pacific, curves through the Indo-Pacific Warm Pool and continues into the southeastern Pacific. A region of positive correlation in the north- central Pacific provides a counterpoint, and the combination of these warm and cool SSTs creates a pattern of low and high pressure that intensifies moisture convergence and rainfall over the eastern Indian Ocean and the Western Pacific – an area called the ‘Warm Pool’ because of its high average ocean temperatures. Increased Warm Pool precipitation leads to below normal East African rainfall.

Figure 3. Correlation between EA long rains and October-November SSTs.
Figure 3. Correlation between EA long rains and October-November SSTs.

Motivated by a desire to understand, and better anticipate, back-to-back droughts such as those associated with the 2010/2011 East African food security crisis, FEWS NET scientists have published numerous papers describing how warm Warm Pool and cool central Pacific SSTs combine to produce dry long rain conditions (a, b, c, d, e, f), while other scientists have emphasized the important role played by the north Pacific (g, h). All three regions (Warm Pool, North Pacific, East Pacific) are currently warm or cool in ways that are conducive to poor East African long rains. In 2012 and 2014, FEWS NET scientists used these relationships to make successful forecasts of below normal rains based on February or March SSTs.  Here, we extend this work to examine levels of predictability based on October-November SSTs, finding reasonable levels of skill and a forecast for poor March-June 2017 rainfall.

Figure 4. January-November 2016 Standardized Precipitation Index values. Based on CHIRPS data.
Figure 4. January-November 2016 Standardized Precipitation Index values. Based on CHIRPS data.

This forecast activity has been prompted by two factors – the very poor performance of 2016 January-November rainfall (Figure 4), and the emergence of a very strong West Pacific SST gradient (WPG) (Figure 5). What we see in Figure 4 is an area of extreme sustained dryness in the eastern Horn that appears associated with both the current poor vegetation conditions (Figure 1 below) and also the sustained low Normalized Difference Vegetation Index values throughout the year. Figure 5 suggests that this dryness has been associated with the development of an extremely steep West Pacific Gradient, caused by a combination of very warm West Pacific SSTs and moderately cool La Niña SSTs. What is interesting about Figure 5 is that some (but not all) recent strong El Niños rapidly transitioned into strong West Pacific Gradient events: the 1987/88, 1997/98 and 2008/09 Niños were followed by strong gradient conditions when cool La Niña SSTs combined with warm conditions in the Western Pacific. Following the 1997/1998 and 2008/09 Niño events, strong gradient conditions also tended to persist for several years while long rains performance tended to be below normal (Figure 2).

Figure 5. Time series of standardized 6-month running average Nino 3.4 and West Pacific SSTs, and their difference (WPG = Nino3.4 minus West Pacific).
Figure 5. Time series of standardized 6-month running average Nino 3.4 and West Pacific SSTs, and their difference (WPG = Nino3.4 minus West Pacific).

Focusing more closely on the Warm Pool region shown in Figure 3, which we selected because of its high correlation with March-June long rains in East Africa, we see very warm conditions (Figure 6, top), suggestive of below normal March-June rains. Several previous post-El Niño years (1998, 2008, and 2010) exhibited similar levels of warmth. Focusing just on years with neutral or cool El Niño-Southern Oscillation conditions, we find a strong negative relationship between October-November Warm Pool SSTs and East African (EA) long rains in the following year (Figure 6, bottom), with a +1°C increase in Warm Pool SSTs associated with a ~-3 standardized anomaly decrease in EA precipitation. Note, however, that a +1°C anomaly is unprecedented in this region (Figure 6, top). The high slope coefficient for this warm pool region is likely to be related to the strong relationship between increases in Warm Pool SSTs and rainfall over the Warm Pool. This region is very warm, and a +0.5°C increase in SST can lead to cyclones and very heavy Warm Pool convection. In turn, this increased precipitation can drive an overturning circulation associated with subsidence and reduced rainfall over East Africa.  Atmospheric model simulations support this relationship, showing a strong (~-0.8) anti-correlation between ensemble average Warm Pool and East African precipitation. A simple bivariate regression with Warm Pool SSTs suggests that we might expect a -1.5Z standardized March-June rainfall anomaly given the current observed SST conditions (-1.5Z = +0.5°C * -3 Z/°C + 0.05Z).

Figure 6. Top - October-November 2016 Warm Pool SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November Warm Pool SST anomalies and standardized East African March-June rainfall anomalies.
Figure 6. Top – October-November 2016 Warm Pool SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November Warm Pool SST anomalies and standardized East African March-June rainfall anomalies.

Figure 7 shows similar results but for the North Pacific region from Figure 3. We identify a strong negative relationship between North Pacific SST and East African long rains.  October-November North Pacific SSTs are very warm (~+0.9°C), similar to 1998, 2007, 2010, and 2011, seasons preceding below normal long rains (Figure 2). A simple bivariate regression prediction based on North Pacific SSTs would be: -1.4Z = +0.9°C * -1.8Z/°C + 0.2Z.

Figure 7. Top - October-November 2016 North Pacific SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November North Pacific SST anomalies and standardized East African March-June rainfall anomalies.
Figure 7. Top – October-November 2016 North Pacific SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November North Pacific SST anomalies and standardized East African March-June rainfall anomalies.

Figure 8 shows a similar scatterplot and time series for the eastern Pacific region identified in Figure 3. Here we find a modest predictive relationship (R2=0.34). East Pacific SSTs do a fairly good job of stratifying dry events. Just using East Pacific SSTs to predict MAMJ rains we would arrive at below normal conditions -0.7Z = 1.8Z/°C * -0.5°C *+ 0.2Z.

Figure 8. Top - October-November 2016 East Pacific SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November East Pacific SST anomalies and standardized East African March-June rainfall anomalies.
Figure 8. Top – October-November 2016 East Pacific SST anomalies, derived using a 1981-2010 baseline. Bottom – Scatterplot showing October-November East Pacific SST anomalies and standardized East African March-June rainfall anomalies.

A difficult question to answer, because we have so few samples, is whether a very warm Western and Northern Pacific, acting alone, can produce substantial East African drying. Figure 9 chracterizes the current situation and identifies two modestly cool East Pacific analogs (2000 and 2008). This scatterplot shows the unique conditions we face at present. The x-axis represents the average Warm Pool/North Pacific SSTs. The y-axis shows our East Pacific SST anomalies. The average of our Western and Northern Pacific time series indicate the warmest conditions on record. The East Pacific shows modest La Niña-like cooling. Because of the non-stationarity of the climate system it is hard to find a large number of analogs. The two closest seasons with ~-0.5°C eastern Pacific SST appears to be October-November of 2000 and 2008, although 2008 is much closer. The right panels of Figure 9 show the longs rains for the following years (2001, 2009). These results could indicate a range of outcomes from slightly below normal to the worst on record (see Figure 2). We do not have to have a strong La Nina to have a severe long rains drought.

Additional evidence supporting the important role of Western and Northern Pacific SSTs comes from our statistical analyses (Figures 6-7), published forecast models (references cited above) and the dynamic response of climate models to Western and Northern Pacific forcing (i), West Pacific forcing (c), and the West Pacific Gradient (d), which is driven by both the western and eastern Pacific, and is currently very strong (Figure 5).

Figure 9. Assessing risks during modestly cool East Pacific seasons. Left - scatterplot of the average WP/NP SST anomalies and East Pacific Anomalies. Right - CHIRPS MAMJ anomalies for modestly cool East Pacific analog seasons.
Figure 9. Assessing risks during modestly cool East Pacific seasons. Left – scatterplot of the average WP/NP SST anomalies and East Pacific Anomalies. Right – CHIRPS MAMJ anomalies for modestly cool East Pacific analog seasons.

We conclude by combining all three predictors using a standard statistical technique – take-one-away cross-validated regression – to evaluate the robustness of our prediction scheme. Cross-validation is carried out by sequentially removing a year’s data, recalculating regression coefficients without this data, and then comparing the forecasted values with the real observations. These values are shown in the scatterplot in Figure 10. The cross-validated R2 is modest (0.36), but the model appears to do a good job of capturing most of the dry events in 1999, 2000, 2008, 2009, 2011 and 2012 but not 1984. Two false alarms appear in 1989 and 2001, although 2001 rainfall performance was dry in some current crisis areas (Figure 9).

Figure 10. Forecasts of March-June Eastern East African rains based on October-November SSTs. Green circles show cross-validated hindcasts. Red star denotes the 2017 forecast (a -1 standardized anomaly ± 1.2Z).
Figure 10. Forecasts of March-June Eastern East African rains based on October-November SSTs. Green circles show cross-validated hindcasts. Red star denotes the 2017 forecast (a -1 standardized anomaly ± 1.2Z).

While our regression-based forecast for 2017 MAMJ rains is very low (-1.5Z), we advocate a more modest outlook this far ahead of the long rains. We assume that MAMJ 2017 will be similar to the other 8 years for which our forecast was below normal. These analog years correspond to the dots to the left of the y-axis in Fig. 10.

Table 1. Forecast statistics for various sets of model years. Years
Table 1. Forecast statistics for various sets of model years.

This approach seems robust to different selections of years (Table 1).  In all cases we find a prediction for below normal MAMJ rainfall. Please also note that the strong negative western and northern slope coefficients remain fairly robust across all the year selections. Knowing the true value of these coefficients is difficult, but the data converges on a pessimistic outlook.  Table 1 also suggests that during neutral-Nina seasons East Africa’s sensitivity to Warm Pool and North Pacific SSTs increases.

Assuming 1989, 1999, 2000, 2001, 2008, 2009, 2011 and 2012 as analog years leads to a forecast of -1Z±1.2Z, with the range depicting 80% confidence intervals. This model gives a 75% chance of below normal rains and has an historic hit rate of predicting 6 out of 7 recent droughts (86% success rate). The red star in Figure 10 shows our 2017 forecast. Different selections of years were tested (Table 1), and all selections produced similar results: below normal longs rains are the most likely outcome, primarily due to negative teleconnections with western and northern Pacific SST. Our forecast range (-1Z±1.2Z) also fits with the outcomes spanned by our two modest-Nina analogs (2000/2001, 2008/2009, Figure 9).

Note that the closest analog (2008/2009) was a very poor rainy season, and that our WPG time series (Figure 5) and regression estimates all suggest SST conditions similar to those preceding some very poor rainy seasons (1999, 2000 and 2011).  In Figures 11 and 12 we highlight this similarity by showing 2016 and 2010 October-November SST and rainfall anomalies. 2016 SSTs are warmer in the western Pacific and northern Pacific than in 2010, but the eastern Pacific is warmer as well. Milder La Niña conditions in 2016 are accompanied by exceptionally warm conditions in the western and northern Pacific. The corresponding ‘Western V’ structure contrasting the western/northern Pacific with cool eastern Pacific SSTs tends to produce dry conditions in the eastern Horn (Figure 3). The observed 2016 SSTs (Figure 11) look a lot like the inverse of our correlation map (Figure 3), indicating a good chance for below normal rains.

Figure 11. Standardized October-November SSTs. A 1981-2010 baseline was used to derive anomalies.
Figure 11. Standardized October-November SSTs. A 1981-2010 baseline was used to derive anomalies.

October-November rainfall anomalies (Figure 12) also suggest conditions similar to, but not identical with, 2010. During October-November of 2010 La Niña-like conditions were better developed, with deeper rainfall reductions near the dateline and more enhanced precipitation over the warm pool. On the other hand, in 2016, rainfall deficits over the Indian Ocean, indicative of the strength of the overurning circulation, are actually larger than those present in 2010. 2016 conditions over the Warm Pool and eastern Pacific appear La Niña-like, with a vigorous overturning circulation. This circulation, however, seems weaker than in 2010, consistent with weaker 2016 La Nina conditions. Going forward, monitoring maps like Figure 11 and 12 will help us assess whether the climate is maintaining a dangerous state or transitioning to a more normal conditions.

Figure 12. October-November CHIRP anomalies.
Figure 12. October-November CHIRP anomalies.

Increased confidence in our outlook is provided by building a MAMJ rainfall composite using atmospheric global circulation model simulations for our analog years. When our statistical model predicted below normal rains based on October-November SSTs, the climate model simulations, driven with observed MAMJ SSTs the following year, indicated a strong overturning circulation with drying over the Horn of Africa (Figure 13). These results suggest that the observed overturning circulation (Figure 12) may persist into 2017. These simulations provide convergent support for our statistical analyses.

Figure 13. Standardized March-June precipitation from ECHAM5 simulations for analog years.
Figure 13. Standardized March-June precipitation from ECHAM5 simulations for analog years.

We can produce a spatial map of the expected drying by using the composite function in the GeoCLIM tool (Figure 14). This map is based on 1989, 1999, 2000, 2001, 2008, 2009, 2011 and 2012. Spatially, these results are broadly similar to rainfall estimates produced by the ECHAM5 model (Figure 13), which also shows below normal rainfall across much of eastern Africa for our analog years. Southern Ethiopia, Somalia and Central and Eastern Kenya appear likely to receive below normal rains. While anticipated conditions in northern Somalia are normal, this could be due to deficiencies in our CHIRPS data – this region has very few stations, and satellite-based rainfall estimates can have trouble seeing low stratus-related rainfall over the easternmost areas of East Africa. Low confidence in our data in this region tranlates into low confidence in our forecast for northern Somalia. On the other hand, this region can also be influenced by extra-tropical disturbances (see below). For Ethiopia, Kenya and Southern Somalia, we have good station coverage as well as high confidence in our satellite estimates.

Figure 14. March-June CHIRPS rainfall anomalies for selected analog seasons - 1989, 1999, 2000, 2001, 2008, 2009, 2011 and 2012.
Figure 14. March-June CHIRPS rainfall anomalies for selected analog seasons – 1989, 1999, 2000, 2001, 2008, 2009, 2011 and 2012.

This analog analysis indicates that the belg growing season in the eastern highlands of Ethiopia may be below normal, resulting in the fifth poor season in parts of this region, which had below normal belg and kiremt rains in 2015 and 2016. Crop growing conditions in central/eastern Kenya and southern Somalia may also be below normal. For many Eastern Kenya and Southern Somalia maize growing regions this would mean a third poor season, since MAMJ and October-November 2016 rains were low. Pastoral areas in these areas may receive little respite.

Summary: Ocean conditions appear similar to 2010/11 and 2008/09, and for our target region below normal rains in MAMJ appears to be the most likely outcome given October-November SST conditions. We are certainly facing an elevated chance of very poor rainfall in at least parts of our target region, though we find normal years within our span of analogs, and SST conditions may change between now and 2017. Spatially, many of the areas expected to receive poor 2017 March-June rains are already very dry (Figures 1 and 4), and the hydrologic impact of these consecutive droughts could be similar in magnitude to 2010/11. We will be able to provide more information as the season progresses. We may see the ‘Western V’ SST pattern cool while eastern Pacific SSTs warm. If not, a continued below normal outlook seems warranted. Continued monitoring of large scale Indo-Pacific SST and rainfall anomalies will help us guage the persistence of the current overturning circulation that produced an exceptionally dry short rains season.

Spatial and Seasonal Forecast Limitations

In closing, we would like to emphasize the spatial and seasonal limitations of this forecast, which has been targeted on Eastern Equatorial East Africa (Figure 1), where prior FEWS NET research has noted strong negative links between increased precipitation over the eastern Indian and Western Pacific Ocean during the March-June season. Deep convection over the Indo-Pacific warm pool almost always produces subsidence and drying over at least part of this focus region. This response, however, tends to be strongest near the equator, and our outlook should not be considered as a forecast for all or even most of the Greater Horn of Africa. Many different climate drivers play different roles in different locations, creating a complex climatic tableau.

In October-May, in the northern parts of the Horn, we know that Ethiopia, Sudan, Eritrea, Djibouti, Yemen, Saudi Arabia and other middle east countries, and northern Somali (Somali Land) are all significantly affected by the southward penetration of north-south or back-hanged mid-latitude troughs. These troughs often originate from polar low pressure cells and penetrate into the tropical regions, where they are associated with short-lived (few days to a week) rainfall events across northern Africa, the Mediterranean, the Middle East and other countries bordering or confined within the Great East Africa Rift Valley (north of the Equator). These wet anomalies are very common between October and May, and current ECMWF forecasts for December indicate substantial trough activity (Figure 15). Whenever these troughs approach or lie along the Rift Valley, the Sahara and Arabian High pressure system disintegrates and is partly pushed over the Arabian Sea, which usually leads to the formation of easterly moisture flows from the Arabian Sea and north Indian Ocean.  These troughs are typically well-established in the mid-troposphere (500 hPa) but sometimes deepens up to 200 hPa, depending on the westerly frontal systems from which they emerge. Some years with anomalously wet October-February conditions were aligned with times when more westerly fronts and the associated trough systems penetrated in to the tropics. The wet anomaly over Ethiopia between 23-30 November 2016 appears to be related to such a trough. The number of troughs so far developed and that will be developing in the coming days seem likely to be more frequent (Figure 15). This anomaly is expected to enhance unseasonal rains over northern sectors of GHA and the Middle East in months of December 2016 as well as January 2017.

We would like to emphasize that in addition to large scale global teleconnection systems associated with the El Niño/Southern Oscillation, West/North Pacific and Indian Ocean Dipole there are a number of other key regional factors that are just as important as ENSO in predicting rains during the October-May semi-dry seasons over the northern sectors of the GHA and the Middle East. These include the pressure gradient and sea surface temperature gradient between the Indian Ocean and Arabian Sea (including the land surface), the oscillatory nature of the North Atlantic Oscillation, the west-east pressure/sea surface temperature gradient over the subtropical Atlantic Ocean, the number and frequency of circumpolar troughs, the position and intensity of the semi-permanent high/low pressure systems (the Azores, Siberia, Sahara, Arabia highs) and middle and upper-tropospheric dynamics and the Madden-Julien Oscillation. Here, we do not take these complexities into account, so our forecast is only appropriate for the March-June over eastern equatorial areas of East Africa.

Figure 15. ECMWF forecast of 850 hPa temperature and 500 hPa geopotential heights. Trough analysis by Diriba Korecha.
Figure 15. ECMWF forecast of 850 hPa temperature and 500 hPa geopotential heights. Trough analysis by Diriba Korecha.

Acknowledgements:  We would like to thank James Verdin and Andrew Hoell for constructive and insightful discussion and analysis involving the current climate situation. We would also like to thank FEWS NET and the USGS for supporting the research that supports this assessment.

Very poor pasture conditions likely to persist through early spring

Chris Funk

This is yet another post about the low short/Deyr rains in East Africa, but today’s focus is on the very poor pastoral conditions in southeastern Ethiopia, northeastern Kenya, and Somalia. This blog is motivated by the extremely low NDVI values, discussed in a recent alert by FEWS NET as well as an email from my friend John McGrath at OXFAM who forwarded me an email from his colleague James Firebrace, who is reporting some severe outbreaks of acute malnutrition and water shortages in eastern Somaliland. After analyzing the available remote sensing data, I find such reports plausible, given the cumulative January-November NDVI record, and would suggest we pay very close attention to food security conditions in this area.

I was already planning to write this post when I arrived at work this morning, before I read John’s email. My concerns were motivated by the very large observed NDVI anomalies combined with the fact that the full impact of the October/November dryness may not be fully registered by the current vegetation.  There is a large degree of persistence and predictability in the NDVI record, and we wrote a paper on this back in 2006 that included a target application on exactly the region under stress now. NDVI tends to lag behind rainfall, so the current poor rainfall could lead to further drying. Even without more drying, it is unlikely (but possible) that the region will see much relief until spring 2017.

Figure 1 shows the most recent MODIS land surface temperature and NDVI anomalies. Many pastoral areas are very warm, which may indicate low levels of soil moisture, and satellite-observed vegetation levels are near or below historic lows in many areas.

 

Figure 1. MODIS Land Surface Temperature and NDVI anomalies for the 1st dekad of November 2016.
Figure 1. MODIS Land Surface Temperature and NDVI anomalies for the 1st dekad of November 2016.

These conditions appear related to long term persistent dryness in the region, as indicated by Standardized Precipitation Index values for the last dekad, month, 2 months, 3 months and 6 months (Figure 2).

Figure 2. Standardized Precipitation Index values calculated from the 1st dekad of November 2016. From earlywarning.usgs.gov.
Figure 2. Standardized Precipitation Index values calculated from the 1st dekad of November 2016. From earlywarning.usgs.gov.

Convergent evidence for water stress and poor water availability for cattle, goats and camels is provided by the FEWS NET waterpoint viewer. Figure 3 shows the current assessment – much of northeastern Kenya and southeastern Ethiopia is under alert or near-dry conditions.

Figure 3. USGS Waterpoint conditions as of early November 2016.

Figure 3. USGS Waterpoint conditions as of early November 2016.

We can look at stressed regions using time series of MODIS Normalized Difference Vegetation Index (NDVI) data. MODIS NDVI measures vegetative health at very high resolution (250m). Figure 4 shows time series of NDVI from far eastern Ethiopia. One year, 2003, was less green. Two recent very poor years 2009/2010 and 2010/2011 were actually greener. By early November, most years have already reached or passed the maximum NDVI value. Typical behavior is for the vegetation to senesce and dry from this point forward.

Figure 4. Somaliland NDVI time series.
Figure 4. Somaliland NDVI time series.

Figures 4, 3, 2 and 1 provide convergent evidence support John McGrath’s concern about very dry conditions in far eastern Ethiopia. This year looks way drier than most recent years.  In terms of long term moisture stress and poor pastoral conditions, it is very concerning that both January-to-early November NOAA CPC RFE2 rainfall totals and January-to-early November NDVI averages indicate very poor conditions (Figure 5). I am not an expert on pastoral early warning, but I believe this kind of repeated aridity can weaken herds and reduce their milk production.

Figure 5. Scatterplot showing 2003-2016 Jan-Nov RFE2 rainfall totals and average eMODIS NDVI for Somaliland.
Figure 5. Scatterplot showing 2003-2016 Jan-Nov RFE2 rainfall totals and average eMODIS NDVI for Somaliland.

Focusing on the same type of target – the anticipated average NDVI conditions between the second dekad of November and the last dekad of March, we find that simple regressions between the Jan-Nov1 NDVI (as shown in Figure 5) and Nov2-March NDVI have a reasonable R2 (~0.4). We can use this simple regression to project a plausible ‘what-if’ scenario for the rest of the dry season. This result is shown in Figure 6.

Figure 6. Somaliland NDVI projections.
Figure 6. Somaliland NDVI projections.

What this figure is telling us is that its quite likely that unless an unseasonal storm drops a lot of water in Somaliland, further browning of the vegetation is likely, as we saw in 2009/2010 and 2010/11. These seasons also appear to be the closest analogs, based on our projected Nov 2016-March 2017 ‘forecast’.

We next carry out a similar analysis for the Mandera region of NE Kenya (Figure 7). Once again, we see historically low levels of NDVI, drier than in 2010. Here, we do seem to be earlier in the season, and there seem to be more years that had a late December recovery.

Figure 7. NDVI time series for the Mandera region of NE Kenya.
Figure 7. NDVI time series for the Mandera region of NE Kenya.

Applying our same ‘forecast’ strategy, i.e. predicting Nov 2016-March 2017 NDVI based on January-November 2016 NDVI (R2=0.4), gives us the results shown in Figure 8. Here the forecast is more modest – a simple continuance of current historic low levels. The analog years based on this analysis might be 2005/2006 and 2010/2011. The USGS waterpoint analysis (Figure 3) provides a strong level of convergence on dryness in this area, and barring an unseasonal upturn in the rains, which has happened before, poor pastoral conditions seem likely to continue, based on the data analyzed here.

Figure 8. Projection of Nov 2016-March 2017 NDVI for Mandera.
Figure 8. Projection of Nov 2016-March 2017 NDVI for Mandera.

We conclude by presenting a new experimental Potential EvapoTranspiration (PET)  forecast, produced by Daniel McEvoy (DRI) and Shrad Shukla (CHG). This forecast (Figure 9) shows the median CFSv2 ensemble forecast over East Africa. The CFSv2 is a sophisticated coupled ocean-atmosphere model used for seasonal forecasts. The total number of ensemble members is 28. They were initialized at an interval of 5 days starting October 8th, 2016 through November 7, 2016.  According to this figure the PET outlook appears to be mostly within normal range (standardized anomaly of -0.44 to 0.44) over the drought regions (Kenya and Somalia).

Figure 9. PET forecasts for Eastern Africa. Kindly provided by Daniel McEvoy (DRI) and Shrad Shukla (CHG).
Figure 9. PET forecasts for Eastern Africa. Kindly provided by Daniel McEvoy (DRI) and Shrad Shukla (CHG).

Putting this all together, we can say that there is lots of evidence that pastoral areas in eastern Ethiopia, NE Kenya, and Somalia appear to have experienced very low rainfall, and that this has resulted in very low NDVI values, both at present and over the January-November time frame (Fig. 6, 8). The poor pasture conditions associated with these low NDVI values appears associated with very warm air temperatures (Figure 1) and water point estimates that appear to be almost dry in many places (Figure 3). The cumulative low NDVI values of the current Jan-Nov season appear to be very low, likely to persist, and similar to seasons like 2009/2010 and 2010/2011.  PET forecasts appear normal, but normal behavior for this region is associated with senescence and NDVI declines in most cases.

Setting aside the outlook for spring 2017 rains, we can look at FEWS NET assessments from January of 2010 and 2011 as possible indicators of where we might be in a few months (a, b, c). We at the CHG are not qualified to do food security assessments, but it does seem safe to say that the current pastoral conditions are very poor in many places, and unlikely to improve until spring. This may mean continued reductions in herd health and pastoral livelihoods. Close monitoring and assessment seems warranted.

 

 

A very poor East African short rains seems almost certain

Chris Funk

 

Last month on this blog we predicted a poor short rainy season for Eastern Kenya and Southern Somalia. Unfortunately that forecast verified. Strong cool-warm-cool sea surface temperature contrasts across the cool Western Indian Ocean, warm Eastern Indian/Western Pacific, and cool Eastern Indian oceans conspire to produce a very strong precipitation dipole (Figure 1) contrasting very heavy precipitation over the Indo-Pacific Warm Pool to the east and the Western Indian Ocean and East Africa to the west. USGS crop and rangeland models indicate ‘no start’ to the growing season almost everywhere.

Figure 1. October 2016 CHIRP rainfall anomalies
Figure 1. October 2016 CHIRP rainfall anomalies

This is very concerning given that the short rains typically end in November for most places in Eastern Kenya and Southern Somalia in November. There is also a great deal of persistence in the climate system between October and November-December. Thus the correlation between October and October-November-December (OND) E. Kenya/S. Somalia rains (please see last post for the specific region) is extremely high: 0.91. Despite that, the mean of the OND rains (224 mm) is much greater than the mean for October (64 mm) (Figure 2).

Figure 2. Scatterplot showing the strong relationship between October and OND E. Kenya/S. Somalia rains.
Figure 2. Scatterplot showing the strong relationship between October and OND E. Kenya/S. Somalia rains.

This is really a quite stunning predictive relationship. November-December E. Kenya/S. Somalia rains also have a strong correlation with October sea surface temperatures (Figure 3). These correlations are really high, with a negative relationship to the Indo-Pacific Warm Pool of -0.9, and positive relationships to the Western Indian and Eastern Pacific of +0.5 and +0.7.

Figure 3. Correlation between November-December E E Kenya/S Somalia rainfall and November-December sea surface temperatures.
Figure 3. Correlation between November-December E E Kenya/S Somalia rainfall and November-December sea surface temperatures. Based on data from 1996-2015.

Unfortunately, the current (October) Ocean looks very much like the opposite of Figure 3 – warm in the Indo-Pacific Warm Pool and cool in the Eastern Pacific (Figure 4). The warmth in the West Pacific is quite exceptional >+2.5 standardized anomalies, which is equivalent to a very warm 30.3 Celsius.

Figure 4. Observed October 2016 sea surface temperatures expressed as standardized anomalies. Based on a 1981-2010 baseline and NOAA Extended Reconstruction v4 data.
Figure 4. Observed October 2016 sea surface temperatures expressed as standardized anomalies. Based on a 1981-2010 baseline and NOAA Extended Reconstruction v4 data.

These strong SST gradients and warm Western Pacific sea surface temperatures contributed to the strong Precipitation dipole (Figure 1) and dry East African conditions in October. At this point, as shown in a recent Africa Hazards Briefing by Nick Novella at NOAA’s Climate Prediction Center (Figure 5), rainfall is very low and recovery is very unlikely. Nick’s awesome Seasonal Performance Probability Tool uses the long historical record of the ARC2 dataset to explore the potential future outcome of a season at a given time and place. It is not looking good for Eastern Africa.

Figure 5. Past rainfall performance and potential future rainfall performance at Baidoa, Somalia. Each grey line represents a projection based on a previous year of the ARC2 data set.
Figure 5. Past rainfall performance and potential future rainfall performance at Baidoa, Somalia. Each grey line represents a projection based on a previous year of the ARC2 data set.

We next explore this issue using two separate approaches: analog seasons and formal (but simple) cross-validated forecasts of OND rainfall. Our composites are based on similar dry Octobers (Figure 6)  1996, 1998, 2005, and 2010 exhibited similar very low October rains.

Figure 6. 1981-2016 CHIRPS E Kenya/S. Somalia rainfall and analog years.
Figure 6. 1981-2016 CHIRPS E Kenya/S. Somalia rainfall and analog years.

What is interesting about these years is that they also tended to exhibit a strong gradient between the Indo-Pacific Warm Pool and Eastern Pacific Ocean (Figure 7). The Western Indian Ocean and the associated Indian Ocean dipole likely played a role as well, but is not explored here, since recent correlations with E Kenya/S Somali rains seems quite low (Figure 3).

Fig. 7. Indo-Pacific warm pool and Eastern Pacific sea surface temperatures from the boxes identified in Figure 5. Dry October analogs are shown with filled black circles.
Figure 7. Indo-Pacific warm pool and Eastern Pacific sea surface temperatures from the boxes identified in Figure 5. Dry October analogs are shown with filled black circles.

Note that the selection of analog years was based on just dry Octobers, so finding our analogs clustered in Figure 7 helps reinforce the strong sea surface temperature forcing indicated in Figure 3, which in turn helps explain the strong Oct-OND relationship indicated in Figure 2. When I used the excellent GeoCLIM tool to plot rainfall anomalies for these analog seasons (Figure 8) I found that these analogs would indicate very poor rainfall performance in Kenya and Somalia, and a poor performance of the Deyr rains in Ethiopia. The 2016 March-April-May rains were very poor in Kenya and Ethiopia. Southeastern-Central Ethiopia is extremely food insecure, having experienced substantial droughts or dryness in the past two Belg and past two Kiremt rainy seasons.

Figure 8. Analog CHIRPS anomalies, expressed as percentages.
Figure 8. Analog CHIRPS anomalies, expressed as percentages.

The prospect of a poor E. Kenya/S. Somalia season can be quantified formally via a simple cross-validated regression using October CHIRPS observations (Figure 2) and Indo-Pacific and East Pacific sea surface temperatures (Figure 7). I first predicted November-December precipitation and then added in the observed October values. The forecast (Figure 8) predicted 1996-2015 OND rains very well (cross-validated R2 of 0.8) and projects an expected OND total of 83 mm, with a standard error of +/- 53 mm. So it seems very likely that the season will be very poor.

This poor season will follow a very poor 2016 MAM rainy season. I made a MAM+OND ‘forecast’ by combining the observed MAM and October E. Kenya/S. Somalia values with our November-December forecasts (Figure 9). The story that emerges is that this region may experience the worst MAM+OND rains in the past 20 years. Since we can be almost certain that short rains outcomes are going to be poor, especially for farmers in Eastern Kenya and Southern Somalia, it may be possible to engage in proactive response planning. While pasture conditions can change more rapidly if anomalous late season rains arrive, current  estimates of vegetation health look quite low.

Figure 9. Cross-validated forecasts of E Kenya/S Somalia MAM+OND 2016 rains.
Figure 9. Cross-validated forecasts of E Kenya/S Somalia MAM+OND 2016 rains.

FEWS NET continues to develop new tools for monitoring and modeling drought.  For example, support from NASA SERVIR is enabling us to provide CHIRPS-compatible Global Ensemble Forecast Systems (GEFS) forecasts, which can be viewed on our Early Warning eXplorer.  Collaboration with the Desert Research Institute has created the FEWS Engine which leverages the incredible processing power of the Google Earth Engine (GEE). We can use the GEE to produce NDVI composites for our analog years (Figure 10) reinforcing our concerns.

Figure 10. Google Earth Engine composites of OND MODIS NDVI for 2005 and 2010.
Figure 10. Google Earth Engine composites of OND MODIS NDVI for 2005 and 2010.

All indicators converge on a very high probability of a very poor short rains. Looming on the horizon is the potential of La Nina-like conditions in the spring. While current NOAA CPC assessments indicate that this is unlikely, if a La Nina did arrive, and the West Pacific continued to be very warm, we might see another poor season in the spring of 2017. If so, providing an adequate response to a 2016 short rains drought would help increase East African resilience in 2017.

Concerns about the Kenya/Somalia short rains

globalpopulation-2
Figure 1. Schematic representation of person years, based on human population from 8000 BC through 2050 AD.

By Chris Funk

Welcome to the first installment of the Climate Hazards Group blog, which we have started to discuss potential climate hazards, current climate extremes, and climate change. I am a federal scientist who has worked for many years with the US Geological Survey’s Center for Earth Resources Observations and Science and the US Agency for International Development’s Famine Early Warning Systems Network (FEWS NET) and collaborating FEWS NET scientists at the Climate Prediction Center, NASA’s Goddard Space Flight Center Hydrology Lab, and the Earth Science Research Laboratory.  I am also an affiliated research professor with the University of California, Santa Barbara Geography Department, and the Research Director for the Climate Hazards Group. We (the CHG) works closely with the NASA SERVIR project, using state-of-the-art satellite observations to improve food security decision making.  In this blog we will analyze and discuss potential climate extremes related to Indo-Pacific sea surface temperatures, focusing on time scales of one to three months and food insecure countries.   We will also leverage new scientific data sets and models to examine potential real world extremes, especially droughts.

Please note, however, that this blog only represents our personal insights and opinions, and not the USGS, FEWS NET, or any of our partner agencies.

The objective of this web space is to monitor climate and put it in historic context. We are living in a unique period of rapid human transition, but also have an unprecedented ability to observe our planet. Since I am a climate watcher, this blog will capture some of that watching and relate what we see to potential food security hazards. Since climate impacts millions of people, my hope is that these posts may be of interest to colleagues in the developing world, FEWS NET, SERVIR, partner agencies like the World Food Programme and other humanitarian relief agencies.

My first observation is that we are living in a time of tremendous change. Think of it this way (Fig. 1). Begin in 1966, and add up all the people living in that year with the number of people living in 1967. Continue adding through to 2055. That’s a measure of ‘human power’ – the total number of person-years people have available to laugh, cook, cry, work and do whatever. Calculating this number we get an estimated total of 631 billion person-years. This staggering amount of time is about 49 times the age of our universe (~13 billion years).

Given such an epic span even I could tell a joke that my teenagers found funny (though of course they wouldn’t laugh).

Now let’s start counting backwards from 1966. We get all the way back to 1000 AD before we accumulate a similar number of person-years. Between 1966 and 2055 as much will happen, in terms of human action, insight, suffering and achievement, as in the 960 years before 1966. So if you feel a little overwhelmed by the world, take heart, the world is probably a little overwhelmed by you as well. Medieval history, the Renaissance, most of the great Chinese empires, the rise of great and powerful Islamic societies, the age of reason, all the wars, the art, the discoveries, and the industrialization between 966 and 1965 – we are likely to achieve that much human activity in our uncertain age of consequence.

Going back another ~631 billion person-years takes us back another two and a half millennia, to around 1500 B.C. Then the next 631 billion years brings us 6,500 years back to about 8000 B.C., the age of the ‘Neolithic’ (new rock) revolution that saw the first continuous human settlements, the advent of farming, and the first domestication of animals. So our current era of the anthropocene will contain about one quarter of human history, expressed in person-years.

I think that we better be paying close attention in this age of compressed activity. Luckily, we have satellites, climate models and networked observations systems. We know that climate change will make our oceans much warmer, and in today’s blog we will examine how warming in the West Pacific and Eastern Indian ocean, combined with La Niña-like cool conditions in the central Pacific, may increase the chances of drought during the upcoming ‘short’ October-November-December (OND) rains in Eastern Kenya and Southern Somalia. More specifically, I am concerned about potential back-to-back droughts in this region in the spring and fall of 2016. We are facing a situation that combines very warm Eastern Indian Ocean/Western Pacific sea surface temperatures with La Niña-like cool  sea surface temperatures in the eastern Pacific: I do some a simple statistical forecast to argue that we need to be concerned about poor East African OND rains.

This part of the world has two rainy seasons each year, one in the spring and one in the fall, and repeated dry seasons can be very hard on the millions of food insecure farmers and pastoralists that live in this region. The spring 2016 March-May rains this year were very low (Fig. 2). This region/season has experienced a substantial decrease in rainfall due to warming in the West Pacific and increases in the West Pacific Warming Mode. A warm West Pacific can contribute to droughts in both the spring and fall.  In the spring we’ve seen a substantial increase in drought frequencies – note in Fig. 2 that ~10 out of the last 16 March-to-May seasons have been dry in comparison to the 1900-1950 mean.

timeserieskensom_mam_rain
Figure 2. Eastern Kenya/Southern Somalia standardized rainfall anomalies. Based on CHG CHIRPS and CenTrends data sets.

 

Here I present a very simple statistical forecast of OND Eastern Kenya/Southern Somalia rains based on observed September sea surface temperatures. We begin by taking a look at the correlation between OND rains and September sea surface temperatures (Fig. 3, left). What we see here is that there is a strong negative (<-0.7) correlation with sea surface temperatures over the west Pacific (boxed area with blue shading), and a similar positive (>+0.7) correlation with sea surface temperatures over the equatorial east Pacific (boxed area with red shading). What we see in this figure is that cool La Niña-like conditions are associated with low East African rains, as are warm west Pacific sea surface temperatures.

What concerns me, right now, is that the moderately cool  La Niña conditions will combine with the very warm conditions in the western Pacific to produce a very strong sea surface temperature gradient, which tends to produce east African droughts. The right panel in Fig. 2 shows observed September sea surface temperature anomalies, expressed as standardized deviations. What we see is modest cooling in the central Pacific, and some very warm (>+2 standard deviations) warm conditions in the west Pacific. Note that the Indian Ocean pattern in the right panel of Fig. 3 also resembles the Indian Ocean Dipole — which is also conducive to dry conditions over the Horn of Africa.

Figure 3. Top – correlation between Kenya/S. Somalia OND rainfall and September sea surface temperatures. Bottom – current September sea surface temperatures expressed as standardized anomalies.

We can place these west Pacific sea surface temperatures in historic context by plotting them as a time series (Fig. 4). We have pointed out the very strong climate change component in West Pacific sea surface temperatures in several papers (a,b,c,d,e), and it is very obvious from this figure. There is a very strong warming trend starting around 1970. It is also interesting to note that all of the past four Septembers have been pretty hot. Right now, West Pacific sea surface temperatures are very warm (third warmest on record). These conditions are likely to persist given the strong thermal inertia of the West Pacific and the latest coupled climate model forecasts (not shown).

timeserieswestpacificssts
Figure 4. West Pacific September sea surface temperature anomalies [deg C].

We next use the observed sea surface temperatures to predict OND East Africa rains using a simple cross-validated regression model. I predicted 1996-2015 Kenya-Somalia OND rains using time series of west Pacific and central Pacific sea surface temperatures. These results are shown in Fig. 5 as blue dots.  While not perfect, this model does a good job of discriminating between most wet and dry seasons. The associated regression coefficients suggest that the influence of the West Pacific is about twice as important as central Pacific. I then used these coefficients to predict 2016 OND Eastern Kenya-Southern Somalia rains (red dot). The very warm West Pacific sea surface temperatures combine with modestly cool central Pacific conditions to produce a forecast for low Eastern Kenya/Southern Somalia rainfall (~-1Z). Note however, that the cross-validated standard error was fairly high (0.8 of a standardized anomaly), so there is a lot of uncertainty in this projection. Below normal rainfall seems likely, however, and this may mean repeated shocks following the poor rainfall performance this spring (Fig. 2).

Figure 5. 1996-2015 cross-validated forecasts of OND Eastern Kenya/Southern Somalia rains (blue circles), together with the 2016 forecast (red circle).

Please bear in mind that this forecast only applies for Eastern Kenya and Southern Somalia. Other nearby regions, like south-eastern Ethiopia or northeastern Tanzania may also be dry, but I did not explicitly evaluate these areas. On the other hand, standardized rainfall maps from past seasons with similar sea surface temperatures showed wide-spread drying  (Figure 6). Note also that current food security situations in Eastern Kenya and Southern Somalia are not at crisis levels. We should be concerned, however, about the possibility of two poor rainy seasons in the spring and fall of 2016.

Figure 6. CHIRPS OND standardized precipitation index maps for similar (analog) seasons.
Figure 6. CHIRPS OND standardized precipitation index maps for similar (analog) seasons.