Abstract

Expanding earlier studies on the boreal spring and autumn rainy seasons in equatorial East Africa, pending challenges on the mechanisms of rainfall variability, are investigated. Eastward pressure gradient and slack south Indian Ocean trade winds allow surface equatorial westerlies in spring and autumn. Complementing that, upper-tropospheric easterlies are required for the development of a zonal vertical circulation cell along the Indian Ocean equator. Because of the summer warming and high stand of upper-tropospheric topography over South Asia, strong upper-tropospheric easterlies over the tropical northern and equatorial Indian Ocean persist from summer into autumn, thus allowing the development of a zonal vertical circulation cell. By contrast, the winter cooling entails low stand of upper-tropospheric topography in the north, thus hindering easterlies over the equator. Consequently, an equatorial zonal circulation cell does not develop in boreal spring. The equatorial zonal circulation cell, with subsidence over East Africa, strongly controls the boreal autumn rains, as evidenced in their tight correlation with the equatorial westerlies. In a related vein, rain gauge stations show much shared variance in boreal autumn as compared to spring. Plausibly consistent with this, boreal autumn rather than spring has brought the extreme flood and drought disasters in the course of the past half-century.

1. Introduction

In two recent reports in this journal (Hastenrath et al. 2007, 2010), we promptly diagnosed the circulation mechanisms that led to extreme flood and drought disasters in Kenya during 2005–08. These accounts have raised awareness of pending challenges. Thus, while Kenya experiences rainy seasons in boreal spring and autumn, extreme events have occurred in autumn. Strong westerlies sweep the central equatorial Indian Ocean in both spring and autumn; however, in autumn only, they are closely accompanied by the rainfall variability. In autumn the equatorial westerlies become part of a zonal vertical circulation cell along the Indian Ocean equator, but in spring such a cell does not develop.

The circulation in the Indian Ocean sector is dominated by the winter and summer monsoons, and these have been extensively explored for more than a century (Ramage 1971; Hastenrath 1985; Pant and Kumar 1997). Less attention has been given to the monsoon transitions, which are the rainy seasons in equatorial East Africa. The extreme flood event of 1961 in eastern Africa sparked interest (Thompson and Mörth 1965; Lamb 1966; Hastenrath 1984, 42–49; Reverdin et al. 1986; Flohn 1987; Kapala et al. 1994). A study of the circulation mechanisms of climate anomalies (Hastenrath et al. 1993) revealed the extremely tight correlation (−0.85) between the surface westerlies over the central equatorial Indian Ocean and the boreal autumn rains in East Africa. Renewed interest in the climate of the equatorial Indian Ocean and East Africa has resumed in recent years, with attention to recent extreme events (Birkett et al. 1999; Behera et al. 1999; Webster et al. 1999; Latif et al. 1999; Hastenrath et al. 2007, 2010), processes in the upper ocean and the atmosphere (Anderson 1999; Saji et al. 1999; Baquero-Bernal et al. 2002; Hastenrath and Polzin 2003, 2004; Black et al. 2003; Clark et al. 2003; Lau and Nath 2004; Behera et al. 2005; Ummenhofer et al. 2009), and seasonal forecasting (Farmer 1988; Mutai et al. 1998; Mutai and Ward 2000; Philippon et al. 2002; Hastenrath et al. 2004).

Section 2 describes the data sources, section 3 reviews the background, section 4 presents the findings on the Kenya rainy seasons, section 5 gives a comparison with the other equatorial ocean sectors, section 6 focuses on the year 2009, and a synthesis is offered in the closing section 7.

2. Data

As in earlier investigations on related issues (Hastenrath 2000, 2007; Hastenrath and Polzin 2003, 2004, 2005; Hastenrath et al. 2007, 2010), the present work draws on rain gauge measurements and data archives of circulation information.

For the period 1948–87 and as indicated in Fig. 1, rain gauge records are available for two ensembles of stations, namely, at the coast—Lamu (LM), Malindi (MD), Mombasa (MB), and Voi (VO)—and in the interior highlands—Nanyuki (NK), Nyeri (NY), Meru (ME), Embu (EM), and Machakos (MK). In addition, we have from earlier work (Hastenrath and Polzin 2003, 2004, 2005) two rainfall indices—rain during October–November (RON) and rain during April–May (RAM)—compiled over the period 1958–97 from seven coastal stations [additional stations include Tanga (TG), Bagamoyo (BA), and Dar es Salaam (DS)] as “all-station-average normalized departure,” using the procedure first introduced by Hastenrath (1976).

Fig. 1.

Orientation map of Kenya, showing domains of two groups of rain gauge stations. Stations available for 1948–87 are at the coast and in the interior highlands. Additional stations used for the indices RAM and RON are TG, BA, and DS; refer to text for station names.

Fig. 1.

Orientation map of Kenya, showing domains of two groups of rain gauge stations. Stations available for 1948–87 are at the coast and in the interior highlands. Additional stations used for the indices RAM and RON are TG, BA, and DS; refer to text for station names.

The essential data archives of large-scale circulation during 1958–97 are the Comprehensive Ocean–Atmosphere Data Set (COADS; Woodruff et al. 1987, 1993) and the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis (Kalnay et al. 1996; Kistler et al. 2001). The latitude–longitude resolution in COADS is 2.0° and 1.0°, and in NCEP–NCAR it is 2.5°. In our earlier work (Hastenrath et al. 1993, 2004, 2007, 2010; Hastenrath and Polzin 2004, 2005), various index series were compiled from these data, as detailed in Fig. 2b. UEQ (4°N–4°S, 60°–90°E) is an index of the zonal component of surface wind over the central equatorial Indian Ocean. The pressure index PW is for a domain in the west (8°N–8°S, 40°–50°E), the pressure index PE is for a block in the east (8°N–8°S, 90°–100°E), and PWE = PW − PE represents the zonal pressure along the equator. SIW (4°–12°S, 60°–90°E) is an index of the total wind speed in the downstream portion of the southern Indian Ocean trade winds. VV (2°N–6°S, 46°–50°E) is the meridional wind component over the western equatorial Indian Ocean. From the NCEP–NCAR upper-air data and as indicated in Fig. 2a, these indices were compiled for 200 mb and the sector 60°–90°E: Z is 200-mb topography at 35°–25°N, UN is zonal wind at 20°–10°N, US is zonal wind at 5°N–5°S, and W5ω is 500-mb vertical motion (2.5°N–2.5°S, 30°–50°E). Also from NCEP–NCAR, meridional profiles of 200-mb topography were constructed for the sectors 60°–90°E, 150°E–90°W, and 0°–40°W.

Fig. 2.

Orientation maps of circulation and climate indices, 60°–90°E. (a) At 200 mb, 60°–90°E: Z is topography at 35°–25°N, UN is zonal wind at 20°–10°N, and US is zonal wind at 5°N–5°S; and at 500-mb, W5ω is vertical motion (2.5°N–2.5°S, 30°–50°E). (b) At the surface: UEQ is zonal wind (4°N–4°S, 60°–90°E); PW is pressure (8°N–8°S, 40°–50°E); PE is pressure (8°N–8°S, 90°–100°E); SIW is wind speed in downstream portion of south Indian Ocean trade winds (4°–12°S, 60°–90°E); and VV is meridional wind component over western equatorial Indian Ocean (2°N–6°S, 46°–50°E). Also shown are indices of RAM and RON.

Fig. 2.

Orientation maps of circulation and climate indices, 60°–90°E. (a) At 200 mb, 60°–90°E: Z is topography at 35°–25°N, UN is zonal wind at 20°–10°N, and US is zonal wind at 5°N–5°S; and at 500-mb, W5ω is vertical motion (2.5°N–2.5°S, 30°–50°E). (b) At the surface: UEQ is zonal wind (4°N–4°S, 60°–90°E); PW is pressure (8°N–8°S, 40°–50°E); PE is pressure (8°N–8°S, 90°–100°E); SIW is wind speed in downstream portion of south Indian Ocean trade winds (4°–12°S, 60°–90°E); and VV is meridional wind component over western equatorial Indian Ocean (2°N–6°S, 46°–50°E). Also shown are indices of RAM and RON.

3. Background

For comprehensive documentation of the annual cycle of circulation and climate in the equatorial Indian Ocean, reference is made to a series of atlases, a book, and journal articles (Hastenrath and Lamb 1979a,b, 2004; Hastenrath and Greischar 1989, 1991; Hastenrath 1995, 57–66, 186–197; Hastenrath 2000, 2001a; Hastenrath et al. 2002, 2007, 2010). A brief synopsis must suffice here.

The annual cycle of the surface wind field over the Indian Ocean is dominated by the alternation between the boreal winter (Fig. 3a) and summer monsoons (Fig. 3c). In boreal winter (Fig. 3a), the surface ship observations show flow from southern Asia recurving near the equator and a confluence in the Southern Hemisphere; in boreal spring (Fig. 3b), recurvature to the south of the equator and confluence in the equatorial region; in boreal summer (Fig. 3c), Southern Hemispheric trade winds recurving near the equator and not reaching any confluence within the map domain; and in boreal autumn (Fig. 3d), recurvature of southern trade winds south of the equator and confluence in the equatorial zone.

Fig. 3.

Surface wind field for (a) January, (b) April, (c) July, and (d) October. Isotach spacing is 2 m s−1. Shading highlights the domain of index UEQ. Dotted line traces the boundary of Kenya as shown in Fig. 1.

Fig. 3.

Surface wind field for (a) January, (b) April, (c) July, and (d) October. Isotach spacing is 2 m s−1. Shading highlights the domain of index UEQ. Dotted line traces the boundary of Kenya as shown in Fig. 1.

Boreal spring and autumn are the rainy seasons in equatorial East Africa; the former is called “long rains” or “mvua ya masika” and the latter “short rains” or “mvua ya vuli,” hereafter referred to as “masika” and “vuli,” respectively. During these limited time spans in the annual cycle, strong westerlies sweep the central equatorial Indian Ocean (Figs. 3b and 3d). The development of the westerlies (UEQ) is favored by the strong eastward pressure gradient along the equator (PWE) and slack winds in the downstream portion of the southern Indian Ocean trade winds (SIW), which allows a recurvature of the flow relatively far south of the equator (Hastenrath and Polzin 2004). In boreal autumn, but not in spring, the surface westerlies (UEQ) become the backbone of a powerful zonal vertical circulation cell along the Indian Ocean equator (Hastenrath 2001a,b, 2007; Hastenrath et al. 2002).

As detailed before (Hastenrath 2000, 2007; Hastenrath and Polzin 2003; Hastenrath et al. 2007, 2010), the equatorial zonal circulation cell features ascending motion over Indonesia, divergent westward flow in the upper troposphere, and subsidence over the western Indian Ocean and East Africa. This feeds into the surface westerlies (UEQ), which are forced by the strong zonal pressure gradient (PWE), with higher pressure in the west and lower pressure in the east. The westerly surface winds drive the eastward equatorial jet (EEJ) or Wyrtki jet in the upper ocean (Wyrtki 1973). The wind exerts forcing on the sea surface temperature (SST) field, which can, by hydrostatic forcing, affect the pressure field. Subsidence, high pressure, and cold SST are not conducive for precipitation. As advised by a reviewer of a previous report (Hastenrath et al. 2007), attention is called to some confusion in the literature. Reference is made to that previous article (Hastenrath et al. 2007) for synopsis and bibliographic details. As evidenced before (Hastenrath and Polzin 2004, 2005; Hastenrath 2007), there is no dipole/seesaw between west and east in either pressure or temperature and no indication of local forcing of temperature on pressure. Fast UEQ enhances the zonal temperature gradient, thus tightening the inverse relationship between the zonal gradients of pressure and temperature.

In boreal autumn, the rainfall at the coast of equatorial East Africa is closely related to the westerlies (UEQ), backbone, and surface manifestation of a powerful zonal vertical circulation cell along the Indian Ocean equator. Over four decades, the correlation amounts to −0.85 (Hastenrath et al. 1993; Hastenrath and Polzin 2003, 2004), arguably the tightest such correlation on the planet. Regarding individual extreme years, 1961, 1994, 1997, and 2006 brought disastrous floods along with slack westerlies, and 2005 and 2008 suffered severe drought with extremely fast UEQ (Hastenrath and Polzin 2003; Hastenrath et al. 2007, 2010).

4. Boreal spring and autumn rainy seasons

The synopsis in the preceding section should open perspective on pending challenges. Flood and drought disasters tend to be more frequent during boreal autumn compared with spring. Why? Why are the boreal spring rains RAM not correlated with the equatorial westerlies as the autumn rains RON, although UEQ is as strong during spring as during autumn? Why does an equatorial zonal circulation cell not develop also during spring, although the UEQ is as strong as during autumn? Some of these challenges shall be explored in the following.

Figure 4 illustrates the annual cycle of rainfall at the two ensembles of rain gauges—at the coast and the interior highlands—indicated in Fig. 1. The two rainy seasons are concentrated around April–May (AM) and October–November (ON). Totals are larger in boreal spring but variance is greater in autumn. For AM and ON, this is further detailed in Table 1. Totals are larger at the coast than in the interior during AM but not during ON. In absolute units the variability in the interior is much larger during ON than during AM. Set in proportion to the long-term mean, both coast and interior exhibit much larger variability during ON than during AM, and the variability at the coast is even larger than in the interior. The findings from Fig. 4 and Table 1 stimulate exploration of the coherence of variability between the rain gauges in each of the two ensembles. The correlation matrices in Table 2 present the basics. Correlations are overall larger during ON than during AM and larger for the interior ensemble than the coast ensemble. When we compacted the content of the correlation matrices, we calculated the variance (correlation coefficient squared) averaged for each of the two ensembles. The variance for the coast during AM is 26% and during ON it is 46%; for the interior, it is 22% during AM and 61% during ON; and for the two ensembles combined during AM it is 15% and during ON it is 36%. Thus, this appraisal shows a remarkable coherence of rainfall variability during ON with compared with AM. The causes for this must be sought in the circulation mechanisms of interannual climate variability.

Fig. 4.

Annual cycle of rainfall, 1948–87 mean, at (a) Kenya coast and (b) interior highlands.

Fig. 4.

Annual cycle of rainfall, 1948–87 mean, at (a) Kenya coast and (b) interior highlands.

Table 1.

Rainfall characteristics during 1948–87 of the two ensembles of stations at the coast and the interior highlands indicated in Fig. 1 during AM and ON: mean and sigma (standard deviation), and ratio (sigma/mean).

Rainfall characteristics during 1948–87 of the two ensembles of stations at the coast and the interior highlands indicated in Fig. 1 during AM and ON: mean and sigma (standard deviation), and ratio (sigma/mean).
Rainfall characteristics during 1948–87 of the two ensembles of stations at the coast and the interior highlands indicated in Fig. 1 during AM and ON: mean and sigma (standard deviation), and ratio (sigma/mean).
Table 2.

Matrices of correlation coefficients (×100) during 1948–87 between rainfall at stations near the Kenya coast (LM, MD, MB, and VO) and in the interior highlands (NY, NK, ME, EM, MK). Refer to text for station names.

Matrices of correlation coefficients (×100) during 1948–87 between rainfall at stations near the Kenya coast (LM, MD, MB, and VO) and in the interior highlands (NY, NK, ME, EM, MK). Refer to text for station names.
Matrices of correlation coefficients (×100) during 1948–87 between rainfall at stations near the Kenya coast (LM, MD, MB, and VO) and in the interior highlands (NY, NK, ME, EM, MK). Refer to text for station names.

Considering the tight association during boreal autumn of rainfall (RON) in equatorial East Africa with the westerlies over the central equatorial Indian Ocean, the development of UEQ is of immediate interest. This has been explored for boreal autumn in earlier work (Hastenrath and Polzin 2004). As indicated in the preceding section 3, the equatorial westerlies UEQ are favored by strong zonal pressure gradient PWE and weak wind in the downstream portion of the southern Indian Ocean trade winds SIW. The domains of the pertinent circulation indices are indicated in the orientation map Fig. 2b, their average annual cycle is plotted in Fig. 5, and Table 3 presents matrices of correlation between the indices. Figure 5 shows the steepest eastward pressure gradient PWE, the strongest southern trade winds SIW, and the fastest cross-equatorial flow in the west VV during boreal summer. The equatorial westerlies UEQ are confined to the boreal spring and autumn: the favorable zonal pressure gradient PWE is moderately large, while the unfavorable southern trade winds SIW are weak enough. The matrices in Table 3 present the correlations between the pertinent indices. The strong correlations of VV with UEQ and PWE may be appreciated in a chain of causality not detailed here, which merits further attention. Table 3 shows the high correlations between the pertinent indices, including the high correlations of UEQ with PWE, SIW, and VV, lesser correlations between these indices, and the very strong correlation with RON. As can be seen, correlations during AM are of same sign as during ON but overall weaker, and most notably there is little correlation of UEQ with RAM.

Fig. 5.

Annual cycle of selected surface circulation indices in the Indian Ocean sector, 1958–97 mean, with domains indicated in the orientation map (Fig. 2b): (a) UEQ zonal wind (m s−1); (b) PWE (mb); (c) SIW (m s−1); and (d) VV (m s−1).

Fig. 5.

Annual cycle of selected surface circulation indices in the Indian Ocean sector, 1958–97 mean, with domains indicated in the orientation map (Fig. 2b): (a) UEQ zonal wind (m s−1); (b) PWE (mb); (c) SIW (m s−1); and (d) VV (m s−1).

Table 3.

Matrices of correlation coefficients (×100) during 1958–97 between indicated indices.

Matrices of correlation coefficients (×100) during 1958–97 between indicated indices.
Matrices of correlation coefficients (×100) during 1958–97 between indicated indices.

As indicated in the preceding section 3, during ON the surface westerlies UEQ become the backbone of a powerful zonal vertical circulation cell along the Indian Ocean equator, but such a cell does not develop during AM. In addition to the surface westerlies, such a cell requires divergent/convergent easterly flow in the upper troposphere, 200 mb. With that perspective, the annual cycle of some pertinent 200-mb circulation indices is plotted in Fig. 6, with domains of the three indices shown in the orientation map (Fig. 2a). Reflecting the temperature and layer-mean thickness over South Asia, Fig. 6a shows low topography during boreal winter and high stand during summer. Consistent with that is the seasonal evolution of the tropical easterly jet (Koteswaram 1958), reflected in the alternation between westerlies during winter and easterlies during summer, apparent in Fig. 6b. Proceeding from there farther southward to the equatorial zone, Fig. 6c shows a similar annual cycle, albeit with easterlies throughout. When comparing the seasons, remarkable differences become apparent: during the tail of boreal winter in AM, the easterlies are relatively weak; during ON following the summer, the easterlies are stronger. Thus, during AM as compared with ON, upper airflow is not favorable for the development of a zonal vertical circulation cell. This has implications for vertical motion over East Africa: conditions are not conducive to spatially coherent anomalies of vertical motion and associated spatial coherence of interannual rainfall variability.

Fig. 6.

Annual cycle of selected 200-mb circulation indices in the Indian Ocean sector, 1958–97 mean, 60°–90°E, with domains indicated in the orientation map (Fig. 2a): (a) Z (gpm; 35°–25°N); (b) UN (m s−1; 20°–10°N); and (c) US (m s−1; 5°N–5°S).

Fig. 6.

Annual cycle of selected 200-mb circulation indices in the Indian Ocean sector, 1958–97 mean, 60°–90°E, with domains indicated in the orientation map (Fig. 2a): (a) Z (gpm; 35°–25°N); (b) UN (m s−1; 20°–10°N); and (c) US (m s−1; 5°N–5°S).

5. Comparison with the Pacific and Atlantic

When considering the development of equatorial zonal circulation, some elementary requirements must be recognized (Hastenrath 2001a, 2007; Hastenrath et al. 2002). First, well-developed zonal flow at the surface in the equatorial zone is essential, thus excluding the continental sectors. Even over the equatorial oceans another prerequisite must be met, namely sufficiently strong zonal flow in the upper troposphere directed opposite of the surface winds. As background information, Fig. 7 presents meridional profiles of 200-mb topography for the three equatorial ocean sectors.

Fig. 7.

Meridional profiles of 200-mb topography (gpm) for (a) Pacific Ocean sector 150°E–90°W; (b) Atlantic Ocean sector 0°–40°W, and (c) Indian Ocean sector 60°–90°E. Dotted line with crosses is January, dashed line with open circles is April, dash–dotted line with triangles is July, and solid line with dots is October.

Fig. 7.

Meridional profiles of 200-mb topography (gpm) for (a) Pacific Ocean sector 150°E–90°W; (b) Atlantic Ocean sector 0°–40°W, and (c) Indian Ocean sector 60°–90°E. Dotted line with crosses is January, dashed line with open circles is April, dash–dotted line with triangles is July, and solid line with dots is October.

Most favorable are conditions in the Pacific sector: strong surface easterlies sweep the equatorial zone all year round; regarding the upper troposphere, Fig. 7a shows the highest topographies near the equator all year round, geostrophically commensurate with westerly flow, opposite of the surface easterlies. Consistent with this, a strong equatorial zonal circulation cell persists throughout the year. Turning to the Atlantic sector, the surface wind field is not conducive through most of the annual cycle (Hastenrath and Lamb 1977; Hastenrath 2001b) because of the persistent cross-equatorial flow from the Southern Hemisphere; only around boreal spring do easterlies cover the equatorial zone. The meridional profiles in Fig. 7b, with maxima in the equatorial zone, entail conditions favoring upper-tropospheric westerlies throughout the year. So in the Atlantic sector, limitations stem from the surface wind field, and accordingly an equatorial zonal circulation cell develops only around boreal spring (Hastenrath 2001b). By comparison, consider now the Indian Ocean sector with its strong surface equatorial westerlies during boreal spring and autumn (Fig. 3). Complementing the annual cycle plots of upper-tropospheric circulation shown in Fig. 6, Fig. 7c displays the meridional profiles of 200-mb topography. Given surface westerlies, the development of zonal circulation cell requires upper-tropospheric easterlies. The evolution of the meridional profile of 200-mb topography in Fig. 7c from July to October is favorable for this, whereas topographies stand low in the north in the evolution from January to April, consistent with the annual cycle plots of zonal flow in Fig. 6. This limits the development of a zonal vertical circulation cell along the Indian Ocean equator to boreal autumn.

6. The year 2009

As our investigation was coming to completion, information became available on the 2009 events, including rainfall data as presented in Table 1. The AM 2009 rains also were deficient, which aggravated the disastrous drought of 2008. Then during October precipitation was abundant, followed by drier conditions during November. As a result, the ON 2009 rains became abundant at the coast and near average in the interior. Attention was called to these evolutions in news releases of the Kenya Meteorological Department (2009a,b) and in the international press (New York Times, 8 September 2009; Der Spiegel, 14 September 2009), with the expectation of abundant rainfall during ON 2009.

During the deficient rainy season AM 2009, UEQ was 0.6 m s−1 faster than the 1958–97 mean, and W5ω 1.6 × 10−4 mb s−1 more subsident, although little association with the zonal circulation cell is indicated for the long-term mean, as indicated in section 3. Interannual variability of boreal spring circulation merits further exploration. In the wetter to near-average ON 2009, UEQ was 0.4 m s−1 faster and W5ω 0.1 × 10−4 mb s−1 more subsident than the 1958–97 mean. However, in detail, the abundant rains of October came with weak UEQ and subsidence W5ω, while in the dry conditions of November UEQ and W5ω were strong, all consistent with the known circulation diagnostics.

7. Conclusions

The average annual cycle of circulation in the Indian Ocean sector is well documented (Hastenrath and Lamb 1977): it is dominated by the alternation between the winter and summer monsoons; the transition between the monsoons, boreal spring and autumn, are the rainy seasons masika and vuli in equatorial East Africa. The objective of the present study was to explore the circulation mechanisms of interannual variability for these seasons.

Rain gauge stations show much common variance during vuli but not masika. Also, vuli brought the fiercest climatic events in the course of the past half-century: the disastrous floods of 1961, 1994, 1997, and 2006 and the droughts of 2005 and 2008. Since our work two decades ago (Hastenrath et al. 1993), we know that the rains of vuli are correlated at −0.85, with the concurrent westerlies over the central equatorial Indian Ocean (UEQ), arguably the highest such correlation on the planet. There is no such correlation in masika, although the westerlies UEQ develop well in both seasons, favored as we know (Hastenrath and Polzin 2004) by strong eastward pressure gradient PWE and weak southern trade winds SIW. On from there, during vuli but not masika, a zonal vertical circulation cell develops along the Indian Ocean equator (Hastenrath 2000, 2001a; Hastenrath et al. 2002), of which the surface westerlies UEQ are a crucial part; it features subsidence over equatorial East Africa, which is tightly correlated with the rainfall variations. The evidence shows for masika in contrast to vuli no equatorial zonal circulation cell, no correlation with the surface westerlies UEQ, less organized rainfall variability, and less extreme flood and drought disasters.

Here we explored the chain of causality. Development of a zonal vertical circulation cell along the equator, with westerlies at the surface, requires easterly flow in the upper troposphere. While this is found in the evolution from the summer monsoon to vuli, it is hindered by the low stand of upper-tropospheric topography in the north during and following the Asian winter monsoon. Without such a zonal circulation cell in masika, there is no strong spatially organized subsidence over equatorial East Africa. This is not conducive to spatially coherent anomalies of vertical motion. Consequently, there is less spatial coherence in the rainfall variability. Plausibly, during vuli and not masika, Kenya experienced its most severe flood and drought disasters in the course of the past half-century.

Acknowledgments

This study was supported by the Variability of Tropical Climate Fund of the University of Wisconsin Foundation. We thank the anonymous reviewers for their helpful comments.

REFERENCES

REFERENCES
Anderson
,
D.
,
1999
:
Extremes in the Indian Ocean.
Nature
,
401
,
327
328
.
Baquero-Bernal
,
A.
,
M.
Latif
, and
S.
Legutke
,
2002
:
On dipolelike variability of sea surface temperature in the tropical Indian Ocean.
J. Climate
,
15
,
1358
1368
.
Behera
,
S. K.
,
R.
Krishnan
, and
T.
Yamagata
,
1999
:
Unusual ocean-atmosphere conditions in the tropical Indian Ocean during 1994.
Geophys. Res. Lett.
,
26
,
3001
3004
.
Behera
,
S. K.
,
J. J.
Luo
,
S.
Masson
,
P.
Delecluse
,
S.
Gualdi
,
A.
Navarra
, and
T.
Yamagata
,
2005
:
Paramount impact of the Indian Ocean dipole on the East African short rains: A CGCM study.
J. Climate
,
18
,
4514
4530
.
Birkett
,
C.
,
R.
Murtugudde
, and
T.
Allan
,
1999
:
Indian Ocean climate event brings floods to East Africa’s lakes and the Sudd Marsh.
Geophys. Res. Lett.
,
26
,
1031
1034
.
Black
,
E.
,
J.
Slingo
, and
K. R.
Sperber
,
2003
:
An observational study of the relationship between excessively strong short rains in coastal East Africa and Indian Ocean SST.
Mon. Wea. Rev.
,
131
,
74
94
.
Clark
,
C. O.
,
P. J.
Webster
, and
J. E.
Cole
,
2003
:
Interdecadal variability of the relationship between the Indian Ocean zonal mode and East African coastal rainfall anomalies.
J. Climate
,
16
,
548
554
.
Farmer
,
G.
,
1988
:
Seasonal forecasting of the Kenya coast short rains, 1901–1984.
J. Climatol.
,
8
,
489
497
.
Flohn
,
H.
,
1987
:
East African rains of 1961/62 and the abrupt change of the White Nile discharge.
Palaeoecol. Afr.
,
18
,
3
18
.
Hastenrath
,
S.
,
1976
:
Variations in low-latitude circulation and extreme climatic events in the tropical Americas.
J. Atmos. Sci.
,
33
,
202
215
.
Hastenrath
,
S.
,
1984
:
The Glaciers of Equatorial East Africa.
Kluwer Academic, 353 pp
.
Hastenrath
,
S.
,
1985
:
Climate and Circulation of the Tropics.
Kluwer Academic, 455 pp
.
Hastenrath
,
S.
,
1995
:
Climate Dynamics of the Tropics.
Kluwer, 488 pp
.
Hastenrath
,
S.
,
2000
:
Zonal circulations over the equatorial Indian Ocean.
J. Climate
,
13
,
2746
2756
.
Hastenrath
,
S.
,
2001a
:
Equatorial zonal circulations from the NCAR-NCEP reanalysis.
J. Meteor. Soc. Japan
,
79
,
719
728
.
Hastenrath
,
S.
,
2001b
:
In search of zonal circulation in the equatorial Atlantic sector from the NCEP–NCAR reanalysis.
Int. J. Climatol.
,
21
,
37
47
.
Hastenrath
,
S.
,
2007
:
Circulation mechanisms of climate anomalies in East Africa and the equatorial Indian Ocean.
Dyn. Atmos. Oceans
,
43
,
25
35
.
Hastenrath
,
S.
, and
P. J.
Lamb
,
1977
:
Climatic Atlas of the Tropical Atlantic and Eastern Pacific Oceans.
University of Wisconsin Press, 97 pp
.
Hastenrath
,
S.
, and
P. J.
Lamb
,
1979a
:
Surface Climate and Atmospheric Circulation.
Vol. 1, Climatic Atlas of the Indian Ocean, University of Wisconsin Press, 116 pp
.
Hastenrath
,
S.
, and
P. J.
Lamb
,
1979b
:
The Oceanic Heat Budget.
Vol. 2, Climatic Atlas of the Indian Ocean, University of Wisconsin Press, 110 pp
.
Hastenrath
,
S.
, and
L.
Greischar
,
1989
:
Upper-Ocean Structure.
Vol. 3, Climatic Atlas of the Indian Ocean, University of Wisconsin Press, 273 pp
.
Hastenrath
,
S.
, and
L.
Greischar
,
1991
:
The monsoonal current regimes of the tropical Indian Ocean: Observed surface flow fields and their geostrophic and wind-driven components.
J. Geophys. Res.
,
96
, (
C7
).
12619
12633
.
Hastenrath
,
S.
, and
D.
Polzin
,
2003
:
Circulation mechanisms of climate anomalies in the equatorial Indian Ocean.
Meteor. Z.
,
12
,
81
93
.
Hastenrath
,
S.
, and
P. J.
Lamb
,
2004
:
Climate dynamics of atmosphere and ocean in the equatorial zone: A synthesis.
Int. J. Climatol.
,
24
,
1601
1612
.
Hastenrath
,
S.
, and
D.
Polzin
,
2004
:
Dynamics of the surface wind field over the equatorial Indian Ocean.
Quart. J. Roy. Meteor. Soc.
,
130
,
503
517
.
Hastenrath
,
S.
, and
D.
Polzin
,
2005
:
Mechanisms of climate anomalies in the equatorial Indian Ocean.
J. Geophys. Res.
,
110
,
D08113
.
doi:10.1029/2004JD004981
.
Hastenrath
,
S.
,
A.
Nicklis
, and
L.
Greischar
,
1993
:
Atmospheric-hydrospheric mechanisms of climate anomalies in the western equatorial Indian Ocean.
J. Geophys. Res.
,
98
, (
C11
).
20219
20235
.
Hastenrath
,
S.
,
D.
Polzin
, and
L.
Greischar
,
2002
:
Annual cycle of equatorial zonal circulations from the ECMWF reanalysis.
J. Meteor. Soc. Japan
,
80
,
755
766
.
Hastenrath
,
S.
,
D.
Polzin
, and
P.
Camberlin
,
2004
:
Exploring the predictability of the “short rains” at the coast of East Africa.
Int. J. Climatol.
,
24
,
1333
1354
.
Hastenrath
,
S.
,
D.
Polzin
, and
C.
Mutai
,
2007
:
Diagnosing the 2005 drought in equatorial East Africa.
J. Climate
,
20
,
4628
4637
.
Hastenrath
,
S.
,
D.
Polzin
, and
C.
Mutai
,
2010
:
Diagnosing the droughts and floods in equatorial East Africa during boreal autumn 2005–08.
J. Climate
,
23
,
813
817
.
Kalnay
,
E.
, and
Coauthors
,
1996
:
The NCEP/NCAR 40-Year Reanalysis Project.
Bull. Amer. Meteor. Soc.
,
77
,
437
471
.
Kapala
,
A.
,
K.
Born
, and
H.
Flohn
,
1994
:
Monsoon anomaly or an El Niño event at the equatorial Indian Ocean? Catastrophic rains 1961/62 in East Africa and their teleconnections.
Proceedings of the International Conference on Monsoon Variability and Prediction, Vol. 1, WCRP-84, WMO/TD 619, World Meteorological Organization, 119–126
.
Kenya Meteorological Department
,
2009a
:
Review of the weather in March-April-May (MAM) and June-July-August (JJA) 2009 seasons and the outlook for the October-November-December 2009 “short rains” season.
KMD Press Release Met/7/23, 7 pp. [Available online at http://www.meteo.go.ke/ranet/wx/OND2009FCST.PDF]
.
Kenya Meteorological Department
,
2009b
:
Weather review in the months of October and November 2009 and the forecast for December 2009.
KMD Press Release Met/7/28, 5 pp. [Available online at ochaonline.un.org/OchaLinkClick.aspx?link=ocha&docId=1141966]
.
Kistler
,
R.
, and
Coauthors
,
2001
:
The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation.
Bull. Amer. Meteor. Soc.
,
82
,
247
267
.
Koteswaram
,
P.
,
1958
:
The easterly jet stream in the tropics.
Tellus
,
10
,
43
57
.
Lamb
,
H. H.
,
1966
:
Climate in the 1960’s: Changes in the world’s wind circulation reflected in prevailing temperatures, rainfall patterns and the levels of the African lakes.
Geogr. J.
,
132
,
183
212
.
Latif
,
M.
,
D.
Dommenget
,
M.
Dima
, and
A.
Grötzner
,
1999
:
The role of Indian Ocean sea surface temperature in forcing East African rainfall anomalies during December–January 1997/98.
J. Climate
,
12
,
3497
3504
.
Lau
,
N-C.
, and
M. J.
Nath
,
2004
:
Coupled GCM simulation of atmosphere–ocean variability associated with zonally asymmetric SST changes in the tropical Indian Ocean.
J. Climate
,
17
,
245
265
.
Mutai
,
C. C.
, and
M. N.
Ward
,
2000
:
East African rainfall and the tropical circulation/convection on intraseasonal to interannual timescales.
J. Climate
,
13
,
3915
3939
.
Mutai
,
C. C.
,
M. N.
Ward
, and
A. W.
Colman
,
1998
:
Towards the prediction of the East Africa short rains based on sea-surface temperature–atmosphere coupling.
Int. J. Climatol.
,
18
,
975
997
.
Pant
,
G. B.
, and
K. R.
Kumar
,
1997
:
Climates of South Asia.
Wiley, 320 pp
.
Philippon
,
N.
,
P.
Camberlin
, and
N.
Fauchereau
,
2002
:
Emperical predictability study of the October–December East African rainfall.
Quart. J. Roy. Meteor. Soc.
,
128
,
2239
1256
.
Ramage
,
C. S.
,
1971
:
Monsoon Meteorology.
Academic Press, 296 pp
.
Reverdin
,
G.
,
D.
Cadet
, and
D.
Gutzler
,
1986
:
Interannual displacement of convection and surface circulation over the equatorial Indian Ocean.
Quart. J. Roy. Meteor. Soc.
,
112
,
43
67
.
Saji
,
N. H.
,
B. N.
Goswami
,
P. N.
Vinayachandran
, and
T.
Yamagata
,
1999
:
A dipole mode in the tropical Indian Ocean.
Nature
,
401
,
360
363
.
Thompson
,
B. W.
, and
H.
Mörth
,
1965
:
Notes from East Africa, No. 1.
Weather
,
20
,
226
227
.
Ummenhofer
,
C. C.
,
A.
Sen Gupta
,
M. E.
England
, and
C. J. C.
Reason
,
2009
:
Contribution of Indian Ocean sea surface temperatures to enhanced East African rainfall.
J. Climate
,
22
,
993
1013
.
Webster
,
P. J.
,
A. M.
Moore
,
J. P.
Loschnigg
, and
R. R.
Leben
,
1999
:
Coupled ocean-atmosphere dynamics in the Indian Ocean during 1997-98.
Nature
,
401
,
356
360
.
Woodruff
,
S.
,
R.
Slutz
,
R.
Jenne
, and
P.
Steurer
,
1987
:
A Comprehensive Ocean-Atmosphere Data Set.
Bull. Amer. Meteor. Soc.
,
68
,
1239
1250
.
Woodruff
,
S.
,
S.
Lubker
,
K.
Wolter
,
S.
Worley
, and
J.
Elms
,
1993
:
Comprehensive Ocean-Atmosphere Data Set (COADS) release 1a: 1980-92.
Earth Syst. Monit.
,
4
,
1
8
.
Wyrtki
,
K.
,
1973
:
An equatorial jet in the Indian Ocean.
Science
,
181
,
262
264
.

Footnotes

Corresponding author address: Stefan Hastenrath, Department of Atmospheric and Oceanic Sciences, University of Wisconsin—Madison, 1225 West Dayton Street, Madison, WI 53706. Email: slhasten@wisc.edu