The Southern African Heat Low: Structure, Seasonal and Diurnal Variability, and Climatological Trends

Kitty Attwood aSchool of Geography and the Environment, University of Oxford, Oxford, United Kingdom

Search for other papers by Kitty Attwood in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0009-0006-1652-947X
,
Richard Washington aSchool of Geography and the Environment, University of Oxford, Oxford, United Kingdom

Search for other papers by Richard Washington in
Current site
Google Scholar
PubMed
Close
, and
Callum Munday aSchool of Geography and the Environment, University of Oxford, Oxford, United Kingdom

Search for other papers by Callum Munday in
Current site
Google Scholar
PubMed
Close
Open access

Abstract

Heat lows are key features of subtropical climates and monsoon systems. In southern Africa, they are pivotal to understanding divergent climate change projections, in particular the veracity of future rainfall decline. Compared to other heat lows, including in West Africa and Australia, the southern African heat low remains poorly documented. Here, we analyze the diurnal cycle, seasonal variability, and trends of the heat low in reanalysis data. In ERA5, 462 strong heat low days are detected between September and March from 1990 to 2019, equating to 7.3% of days sampled. These events feature ascent (exceeding −0.2 Pa s−1) at low levels (strongest between 800 and 600 hPa) and subsidence aloft, generating low-level cyclonic flow with anticyclonic flow above. This flow exhibits strong diurnal variability, with peak windspeeds between 0600 and 0900 UTC and maximum ascent at ∼2300 UTC. Heat lows form preferentially over Angola in September (∼14°S) and October (15°–20°S), and in Namibia from November to March (∼20°–26°S). Strongest ascent occurs over areas of high elevation. Finally, we show a rapidly increasing frequency of strong heat low days, with a 175% increase between 1960–89 and 1990–2019. The greatest increase (459%) has occurred in the early summer months of September and October, consistent with projections of delayed rainfall onset. Strikingly, more strong heat lows are detected in the most recent 5 years of analysis (2014–19) than in the 30-yr period from 1960 to 1989. These results suggest the heat low is an important feature in determining drying trends over southern Africa and is a vital indicator of climate model accuracy.

Significance Statement

This work documents the heat low that forms in southern Africa in the lowest levels of the atmosphere. The feature is present during austral summer (from September to March) and is associated with below average rainfall across much of the subcontinent. The frequency of strong heat lows has rapidly increased in line with regional amplified warming trends. The heat low is identified as an important control on circulation and precipitation patterns and changes in the frequency or intensity of the feature in the future are likely to influence the strength of declining rainfall trends across southern Africa.

© 2024 American Meteorological Society. This published article is licensed under the terms of a Creative Commons Attribution 4.0 International (CC BY 4.0) License .

Corresponding author: Kitty Attwood, kitty.attwood@ouce.ox.ac.uk

Abstract

Heat lows are key features of subtropical climates and monsoon systems. In southern Africa, they are pivotal to understanding divergent climate change projections, in particular the veracity of future rainfall decline. Compared to other heat lows, including in West Africa and Australia, the southern African heat low remains poorly documented. Here, we analyze the diurnal cycle, seasonal variability, and trends of the heat low in reanalysis data. In ERA5, 462 strong heat low days are detected between September and March from 1990 to 2019, equating to 7.3% of days sampled. These events feature ascent (exceeding −0.2 Pa s−1) at low levels (strongest between 800 and 600 hPa) and subsidence aloft, generating low-level cyclonic flow with anticyclonic flow above. This flow exhibits strong diurnal variability, with peak windspeeds between 0600 and 0900 UTC and maximum ascent at ∼2300 UTC. Heat lows form preferentially over Angola in September (∼14°S) and October (15°–20°S), and in Namibia from November to March (∼20°–26°S). Strongest ascent occurs over areas of high elevation. Finally, we show a rapidly increasing frequency of strong heat low days, with a 175% increase between 1960–89 and 1990–2019. The greatest increase (459%) has occurred in the early summer months of September and October, consistent with projections of delayed rainfall onset. Strikingly, more strong heat lows are detected in the most recent 5 years of analysis (2014–19) than in the 30-yr period from 1960 to 1989. These results suggest the heat low is an important feature in determining drying trends over southern Africa and is a vital indicator of climate model accuracy.

Significance Statement

This work documents the heat low that forms in southern Africa in the lowest levels of the atmosphere. The feature is present during austral summer (from September to March) and is associated with below average rainfall across much of the subcontinent. The frequency of strong heat lows has rapidly increased in line with regional amplified warming trends. The heat low is identified as an important control on circulation and precipitation patterns and changes in the frequency or intensity of the feature in the future are likely to influence the strength of declining rainfall trends across southern Africa.

© 2024 American Meteorological Society. This published article is licensed under the terms of a Creative Commons Attribution 4.0 International (CC BY 4.0) License .

Corresponding author: Kitty Attwood, kitty.attwood@ouce.ox.ac.uk

1. Introduction

Southern Africa is a region with a particularly unique and complex climate owing to a combination of tropical and midlatitude influences, contrasting sea surface temperatures on the east and west coasts and a high interior plateau (Tyson and Preston-Whyte 2000). Considerable gaps remain in our knowledge of the southern African climate system, hindering prediction relating to interannual variability and trends associated with anthropogenic climate change. As an arid region with high exposure to climate across many economic sectors, southern Africa is particularly vulnerable to climate change (Conway et al. 2015). Annual precipitation rates are projected to decrease across the region throughout the twenty-first century owing to a delayed onset of the summer rainy season (Dunning et al. 2018; Ranasinghe et al. 2021). Despite intermodel agreement in the future drying trend, there are considerable differences in the projected magnitude of change between models (Munday and Washington 2019). In addition, many models overestimate current rainfall—some by as much as 300%—casting doubt on their ability to simulate future precipitation accurately (Munday and Washington 2018). Improving our understanding of contemporary climate dynamics at a regional scale is therefore essential, not only to improve forecasting of interannual and interseasonal variability but also to better constrain models and evaluate future projections (James et al. 2018).

A key feature of southern African climate that remains understudied is the heat low that forms during austral summer due to intense solar heating, initially as the Angola (heat) low and subsequently as the Kalahari heat low, as it shifts southward to the Kalahari Desert around 25°S (Vizy and Cook 2016; Munday and Washington 2017; Howard and Washington 2018). The heat low in southern Africa exerts a strong influence on regional rainfall and the accuracy of model rainfall estimates (Munday and Washington 2017). Furthermore, the Angola and Kalahari heat lows are expected to strengthen due to anthropogenic climate change, which will affect millions of people, particularly via effects on agriculture (Vizy et al. 2015; Vizy and Cook 2016).

Considering the importance of the regional heat low for the current and future climate of southern Africa, it is essential that the feature and its seasonal evolution are understood, yet no paper to date has provided a full climatology of the feature. In this study, we address the following questions:

  1. How does the location and structure of the heat low vary through the annual cycle?

  2. How does the heat low evolve through the diurnal cycle?

  3. Are there trends in the strength of heat low events?

Section 2 summarizes the understanding of heat lows gained from theoretical work and observed data. Section 3 details the data and metric used to detect heat lows and section 4 describes how this was used to detect strong heat low days and identifies the mean structure of the feature during these cases. Section 5 deals with the seasonal variability of the heat low, and section 6 discusses the diurnal cycle of strong heat lows. Section 7 documents trends in the strongest heat lows. The final section provides a summary and conclusions.

2. Heat lows and heat low circulation

a. Heat lows

Heat lows occur in the lower troposphere of arid regions, particularly during summer months, where intense solar heating and low soil moisture levels initiate dry convection (Spengler and Smith 2008), creating an area of low surface pressure that drives a cyclonic circulation near the surface. Ascent is capped by subsiding air above at approximately 600 hPa, where divergence generates an anticyclone. The rhythm of this circulation is set by the diurnal cycle of solar radiation; greatest ascent occurs after the afternoon surface heating maximum (Rácz and Smith 1999). This diurnal cycle can dictate low-level winds, rainfall, heatwave occurrence, and monsoonal inflow (Hoinka and Castro 2003).

Numerical models can reproduce the circulation patterns associated with heat lows and have demonstrated the importance of topography, sea breezes, and low-level jet inflows in their formation (Reichmann and Smith 2003). Model simulations exhibit a nocturnal peak in low-level winds and a corresponding overnight maximum in cyclonic vorticity (Rácz and Smith 1999). Daytime turbulent mixing in the boundary layer weakens the cyclonic flow. The absence of this inflow allows surface pressure to decrease throughout the day until surface heating reduces and the flow toward the low pressure center of the heat low can resume, reducing the pressure gradient (Zängl and Chico 2006; Spengler and Smith 2008). Model experiments have also demonstrated that heat lows often occur in areas of high orography where heating is concentrated over a shallower column of air (Zängl and Chico 2006; Smith and Spengler 2011).

Model experiments have been supported by studies of heat lows using both reanalysis and observational datasets, such as for the Iberian Peninsula (e.g., Hoinka and Castro 2003), the Arabian Peninsula (e.g., Blake et al. 1983; Smith 1986), Australia (e.g., Lavender 2017), North America (e.g., Rowson and Colucci 1992), and West Africa (e.g., Parker et al. 2005; Lavaysse et al. 2009; Engelstaedter et al. 2015). Comparatively few studies have focused on southern Africa, where the heat low has, however, been identified as an important feature (e.g., Vizy and Cook 2016; Munday and Washington 2017).

b. Heat lows in southern Africa

The southern African heat low initially forms in the plateau region of southern Angola as the Angola low (Vizy and Cook 2016). The Angola low is only a heat low in the early summer season (September–November); from December the low shifts from dry to moist convection (Howard and Washington 2018), becoming a tropical low as the heat low migrates southward to become the Kalahari heat low in the core summer months (December–February) (Munday and Washington 2017). The tropical low and heat low forms of the Angola low can be distinguished via their vertical profiles; the deeper, tropical Angola low features negative relative vorticity from the surface to 400 hPa, whereas negative relative vorticity values associated with the heat low do not extend above 600 hPa (Fig. 3 in Howard and Washington 2018; Crétat et al. 2019; Pascale et al. 2019). At low levels, the heat low (tropical low) features lower (higher) specific humidity and a lower (higher) potential temperature lapse rate (Howard and Washington 2018). The heat low is embedded in the regional atmospheric circulation; the pressure gradient between the South Atlantic and Indian Ocean anticyclones and the heat low drives moisture flux and low-level winds across the subcontinent (Cook et al. 2004; Munday and Washington 2019) particularly via low-level jet inflow from the east through the Limpopo and Zambezi valleys (Spavins-Hicks et al. 2021).

Howard and Washington (2019) show that the southern African heat low is linked to the presence of a dryline known as the Congo air boundary (CAB). The CAB marks the meeting point of moist northerly winds and dry southeasterlies in a convergence line feature which is a key determinant of rainfall onset over southern Africa (Howard and Washington 2019). The CAB lies to the north of the heat low and sets up the cloud-free skies to the south required for strong solar heating and heat low formation. As the season progresses the dryline shifts to the Kalahari Desert, forming to the east of the heat low (Howard and Washington 2019; Van Schalkwyk et al. 2022). The drylines are themselves maintained by the strong temperature and moisture gradients generated by the heat low circulation (Howard and Washington 2019).

The formation of the heat low is also linked to the African easterly jet south, which is driven by the anticyclone (the Botswana high) that forms between heat-low-induced ascent and upper-level subsidence from 600 to 500 hPa (Adebiyi and Zuidema 2016; Reason 2016; Vizy and Cook 2016; Kuete et al. 2020). Subsidence associated with the Botswana high generates the clear skies required for intense surface heating (Driver and Reason 2017), thus strengthening the heat low, yet the heat low simultaneously strengthens the Botswana high by providing inflow via ascent in the lower troposphere (Kuete et al. 2023). The heat low is thus an important control on interannual rainfall and temperature variability and the onset of the summer rainy season over southern Africa (Mulenga 1999; Vizy et al. 2015).

Studies have indicated a strengthening of the heat low in recent decades due to amplified surface warming trends over the Kalahari region (Vizy et al. 2015; Vizy and Cook 2016). Cook and Vizy (2013) and Engelbrecht et al. (2009) show that the heat low circulation strengthens in response to increased greenhouse gas concentrations in two separate regional climate model simulations. Potential changes in the heat low have been posed as a mechanism of late rainfall onset over southern Africa in future projections (Dunning et al. 2018). The capacity of climate models to accurately resolve the heat low is linked to model ability to simulate current climate over the region. The majority of coupled climate models simulate a large positive precipitation bias (Munday and Washington 2017). The tropical low form of the Angola low in late summer has been shown to influence model precipitation biases (Reason and Jagadheesha 2005; Dieppois et al. 2015; Munday and Washington 2017). As a stronger southern African heat low has been linked to suppressed rainfall across the subcontinent (Vizy et al. 2015), it is possible that models with anomalously weak or absent heat lows exhibit the greatest positive precipitation biases. Assessing the structure, diurnal cycle and seasonal evolution of the regional heat low are hence important steps from which to progress understanding of both current and future climate over southern Africa.

3. Data

This study uses hourly ECMWF Reanalysis v5 (ERA5) data (with a horizontal resolution of 0.281 25°) to investigate the structure and variability of the heat low over southern Africa, focusing primarily on 1990–2019 and extending to 1960 to assess observed trends in section 7. ERA5 has a high spatial resolution (a native resolution of approximately 31 km × 31 km) across 137 vertical levels (Hersbach et al. 2020). The dataset performs well over Africa, with greatly reduced biases in temperature and precipitation compared to its predecessor, ERA-Interim (Gleixner et al. 2020; Steinkopf and Engelbrecht 2022; Gbode et al. 2023). Despite this, the accuracy of ERA5 over areas of southern Africa, where station data are limited, is less well known and tropospheric values in particular are hard to evaluate due to a lack of upper air sounding data (Hersbach et al. 2020; Roffe and van der Walt 2023). Additionally, care must be taken when analyzing trends in reanalysis datasets due to potential inhomogeneities introduced by changes in the observational network and data assimilation. In acknowledgment of these limitations, we test the robustness of key results by comparison with Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2) reanalysis data at a horizontal resolution of 0.5° × 0.625°, which is available from 1980 to present at 3-hourly timesteps (Gelaro et al. 2017).

Following previous studies, primary heat low detection is performed using low-level atmospheric thickness (LLAT)—a metric used to determine the dilation of the lower levels of the troposphere (Lavaysse et al. 2009). LLAT is taken here as the difference in geopotential heights at 700 and 850 hPa. A lower bound of 950 hPa falls below the surface level across much of the southern African plateau.

Trends in heat low frequency are assessed in two ways: 1) the Wilcoxon signed rank test for difference between the periods 1960–90 and 1990–2020 and 2) Poisson regression for the time series of strong heat low days.

4. Heat low detection

A 30-yr time series of austral summer (September–March) LLAT was generated across southwest Africa (10°–30°S, 12°–24°E) at 1500 coordinated universal time (UTC). Based on our analysis of the diurnal cycle, the 1500 UTC time step was chosen to coincide with the afternoon LLAT maximum (demonstrated in section 5). Days where the spatial mean 1500 UTC LLAT was above the 90th percentile of 1990–2019 values were added to an initial dataset of strong heat low days. This absolute rather than relative monthly detection threshold was applied to enable a comparison of the strength of the heat low throughout the season. As there is no absolute threshold for determining the presence of a heat low, this stringent threshold enables an analysis of the structure and variability of particularly strong and well-defined heat low events. Similar thresholds have been employed for other heat lows, such as in West Africa and Australia (Lavaysse et al. 2009; Lavender 2017). Events exceeding the 90th percentile of 1500 UTC LLAT were included in the final sample if the grid point of maximum LLAT met the additional criteria of featuring subsidence in the daily mean at 300 hPa to remove events associated with deep moist convection. Events were also removed if the LLAT maximum occurred outside the core study area, where the Angola and Kalahari heat lows are known to occur (10°–30°S, 12°–24°E). Monthly composite vertical profiles of potential temperature and vertical velocity were assessed to ensure the characteristics were not contingent on the chosen detection scheme.

These criteria yielded a total of 462 strong heat low days (7.3% of all days sampled), with the highest number detected in October (114) followed by January (89), November (75), February (57), December (55), September (37), and with the fewest detected in March (35). The spatial distribution of events demonstrates the transition from the Angola (heat) low in the early season (September and October) to the Kalahari heat low, following the seasonal shift of the solar radiation maximum (Fig. 1). However, the LLAT maxima are clustered with the heat low favoring two broad positions, rather than undergoing a gradual southward progression. LLAT maxima are clustered in areas of high elevation, such as the Huila and Bié plateaus in Angola and Khomas Highland in Namibia, indicating the role of topography, which is discussed further in section 5. Due to the high elevation of the Namibian plateaus, when defined by maximum thickness, most strong heat lows that form in the late season (as the Kalahari heat low) are centered to the northwest of the Kalahari Desert. The heat low detection scheme was also applied to MERRA-2, which yields a similar spatial distribution of events, although slight differences in the location of the LLAT maxima are inevitable in part due to the higher resolution of ERA5 (Fig. 1 in the online supplemental material).

Fig. 1.
Fig. 1.

The location of detected strong heat lows in southwest Africa in ERA5 from September to March 1990–2019, defined by the grid point of maximum low-level atmospheric thickness (LLAT). The size of the marker corresponds to the number of heat lows at that grid point.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

Latitude–height composite profiles of potential temperature, vertical velocity, and zonal and meridional winds point to a coherent structure of the lower atmosphere for strong heat low days across the annual cycle (Fig. 2). High surface potential temperature values indicate clear skies and high surface heating in the heat low region. Little change in potential temperature with height at the center of the heat low indicates the central region of instability where the atmosphere is well mixed due to dry convection; at 1200 UTC the 317-K isoline extends from 650 hPa to the surface (Fig. 2). This corresponds to negative vertical velocity values of −0.2 Pa s−1 from 700 to 600 hPa at the center of the heat low, denoting ascent at low levels. Weaker values extend up to 500 hPa, above which the heat low is capped by subsidence. Zonal winds highlight the midlevel African easterly jet south with mean wind speeds exceeding 7 m s−1. Westerly flow associated with the westerly subtropical jet dominates toward the south. Meridional winds show cyclonic converging flow below 700 hPa and divergent, anticyclonic flow from 700 to 500 hPa (∼2–3 m s−1). The composites capture a classic heat low structure.

Fig. 2.
Fig. 2.

Composite daily mean latitude–height vertical profiles of strong heat low days in ERA5 from September to March for potential temperature (K), vertical velocity (Pa s−1), zonal wind (m s−1), and meridional wind (m s−1). Profiles are averaged over 5° and centered over the LLAT maximum to account for the seasonal movement of the heat low. Negative (positive) latitude values represent degrees north (south) from the heat low center. Negative (positive) longitude values represent degrees west (east) from the heat low center.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

5. The seasonal cycle of low-level atmospheric thickness and heat low presence

This section considers the annual cycle of the heat low. At 1500 UTC, LLAT over southwest Africa reaches its maximum (∼1660 m) between late September and early March, declining to a climatological minimum of 1620 m in austral winter (Fig. 3). Both ERA5 and MERRA-2 demonstrate similar annual cycles, although ERA5 exhibits a slightly lower LLAT during austral summer. The climatological LLAT is nonetheless similar between the two reanalyses at 1656.6 and 1658.9 m in ERA5 and MERRA-2 respectively. Average LLAT remains somewhat constant within the study period of September to March, although the LLAT maximum (defined as values exceeding the 99th percentile) moves southward by 10° latitude (Fig. 4). In September, the LLAT maximum is situated above the Angolan highlands (at approximately 14°S), indicating the dominance of the Angola (heat) low in the early season. October marks the point of transition when the LLAT maximum is found between Angola and Namibia. By November, the LLAT maximum is situated across western-central Namibia, where it remains until the end of the summer season in April (22°–24°S).

Fig. 3.
Fig. 3.

The annual cycle of LLAT at 1500 UTC for southwest Africa (10°–30°S, 12°–24°E) in ERA5 and MERRA-2. The red and blue lines denote the multiyear mean from 1990 to 2019 of ERA5 and MERRA-2, respectively, and individual lines show each year for ERA5 only.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

Fig. 4.
Fig. 4.

The annual cycle of LLAT in ERA5 at 1500 UTC averaged from 1990 to 2019. Dashed contours denote the 95th percentile, and solid contours denote the 99th percentile of LLAT relative to each month.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

The relationship between topography and heat low location was tested by comparing ERA5 surface elevation for all land grid points across detected heat low days (n = 3338) and a subset of grid points with strong ascent in the seasonal mean (September to March). Grid points of strong ascent are defined by integrated 800–600-hPa vertical velocity values below −0.15 Pa s−1 (n = 205). A left-tailed z test reveals a statistically significant difference in surface elevation at grid points with strong ascent, demonstrating that high elevation is associated with strong overlying ascent (z score = 19.36, p value < 0.0001). Strong ascent in the seasonal mean occurs above land that is an average of 267 m higher in elevation than the average for southwest Africa (1110 m). Dry convection is enhanced above areas of high elevation, where surface heating induces strong heating of the comparatively shallow atmospheric column (Smith and Spengler 2011).

Figure 5 shows the seasonal shift in vertical velocity during strong heat low days. In September and October, low-level ascent (below 600 hPa) is favored equatorward of 16°S, above which there is subsiding air. As the season progresses, low-level ascent favors more southerly latitudes, with a northern limit of 14°S in November, shifting to 17°S by February. This southward movement is accompanied by a reduction in the latitudinal extent of the heat low; a region of deep ascent encroaches from the north as the summer progresses, visible through the vertical extent of the troposphere and reaching a southern limit of ∼16°S in January and February. This deep layer of ascent is associated with increasing specific humidity—the 5.0 g kg−1 specific humidity contour at 600 hPa moves by over 10° latitude throughout the season—indicating tropical moist convection associated with the southward movement and subsequent breakdown of the CAB. Potential temperature contours further support this conclusion; in the heat low region, weak vertical gradients of potential temperature indicate low static stability (Fig. 5). In addition, high surface potential temperatures point to strong surface heating due to clear skies and thus dry, rather than moist, convection. The area with the weakest vertical potential temperature gradient tracks southward through the season and, at lower latitudes, gradients of potential temperature increase with growing deep ascent, indicating the release of latent heat associated with moist convection. Figure 5 thus captures the seasonal cycle of the heat low whereby dry convection shifts from the Angola (heat) low to the Kalahari heat low as tropical moist convection becomes dominant at low latitudes from late October.

Fig. 5.
Fig. 5.

Average daily mean latitude–height composites of vertical velocity (Pa s−1), specific humidity (g kg−1; solid black lines), and potential temperature (K; dotted blue lines) during strong heat low days (n = 462) in ERA5, averaged from 12° to 24°E. Red shades indicate negative (upward) vertical velocity. Black shading indicates averaged topography.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

In alignment with this latitudinal shift, the zone of greatest convergence moves southward and the low-level winds providing the inflow shift from being dominantly associated with the Zambezi low-level jet (September–November) to the Limpopo low-level jet (December–March) (not shown). Portela and Castro (1996) similarly demonstrate that surface winds flowing into the Iberian heat low tend to follow major river valleys and the importance of the heat low in setting up the subcontinental pressure gradient for low-level jet flow has been noted for southern Africa (Spavins-Hicks et al. 2021; Van Schalkwyk et al. 2022). These studies highlight the importance of the heat low for low-level winds across the entire subcontinent.

A striking feature of strong heat low events is that, across all months, they are associated with below average precipitation across much of southern Africa, particularly over southern Angola, Namibia, Botswana, South Africa, and southern Zambia (Fig. 6). The average daily negative precipitation anomaly increases each month as the mean climatological rainfall increases. From December, the northern extent of negative precipitation anomalies follows the southwest–northeast alignment of the CAB (Howard and Washington 2019). In January and February, positive rainfall anomalies are visible north of the expected boundary, with anomalously high precipitation on heat low days in northern Angola (January) and the southern Congo basin (February). This dipole pattern suggests that heat lows are linked to the CAB, whereby dry convection south of approximately 14°S associated with the heat low prevents equatorial moisture from moving south (Kuete et al. 2020). In the late rainy season months of January and February, the dipole anomaly thus reflects an anomalously late-season CAB structure. The earlier summer months (September–November) do not exhibit this dipole anomaly as the CAB is present on most days during this period (Howard and Washington 2019).

Fig. 6.
Fig. 6.

ERA5 daily average precipitation anomalies (mm day−1) during heat low days from September to March, relative to the 1990–2019 climatology. Stippling denotes areas where the precipitation anomaly exceeds the 0.05 significance level using a Mann–Whitney U test and where the monthly average precipitation exceeds 30 mm.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

6. The diurnal cycle of strong heat lows

Average LLAT across southwest Africa undergoes a diurnal oscillation of approximately 12 m (Fig. 7). Maximum LLAT occurs in the afternoon (1500 UTC; 1700 local time) in response to daytime surface heating and the minimum is reached in the early morning (0700 UTC; 0900 local time). Diurnal characteristics are consistent across all months studied, albeit with varying magnitudes of the daytime maximum in agreement with the seasonal cycle outlined in section 5. Composites of strong heat lows show that winds in the low-level cyclone peak in the morning between 0500 and 0900 UTC, whereas upper-level winds exhibit less variability (Fig. 8). Low-level winds weaken during the day following increased boundary layer turbulence (Parker et al. 2005). Low-level moisture flux also peaks overnight, demonstrating the role of the nocturnal low-level jets in bringing moisture from the southwest Indian Ocean (not shown) (Spavins-Hicks et al. 2021).

Fig. 7.
Fig. 7.

Average hourly ERA5 LLAT across southwest Africa (10°–30°S, 12°–24°E) from 1990 to 2019.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

Fig. 8.
Fig. 8.

Cross sections of ERA5 (left) vertical velocity (Pa s−1) and contours of potential temperature (K), (center) meridional winds (m s−1), and (right) zonal winds (m s−1) during detected heat low events from September to March, at 6-hourly (UTC) timesteps. Data are centered over the LLAT maximum of each event, averaged over 5° longitude/latitude to account for the seasonal movement of the heat low. Green/blue shades of meridional wind (center) indicate positive values (northward flow). Green shades of zonal wind (right) indicate positive (westerly) flow. Negative (positive) latitude values represent degrees north (south) from the heat low center. Negative (positive) longitude values represent degrees west (east) from the heat low center.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

The afternoon peak in LLAT coincides with the steepest potential temperature gradient where values of 317 K extend below 850 hPa, indicating strong vertical mixing associated with daytime turbulence (Fig. 8). Vertical velocity also exhibits a strong diurnal cycle at low levels; subsidence occurs throughout the troposphere in the morning hours and is replaced by ascent up to 500 hPa during the afternoon, which reaches its maximum strength overnight at ∼2300 UTC (0100 local time), in agreement with Howard and Washington (2019). The diurnal cycle of ascent is uniform across all months of analysis and is consistent between ERA5 and MERRA-2 (supplemental Fig. 2). Streamlines show that low-level ascent begins in the early afternoon, with upward motion extending up to 400 hPa at 1500 UTC (1700 local time) but persisting at lower levels until the early morning (Fig. 9). Comparable strong evening ascent in the heat low region of northern Australia was documented by Hart (1990), and similarly Huaman et al. (2023) found that the ascent maximum of the West African heat low occurs around 0000 UTC in ERA5 reanalysis. It is not immediately clear what drives the strong nocturnal ascent. One hypothesis is that nocturnal ascent occurs due to the convergence of inflow from the Zambezi and Limpopo low-level jets [e.g., as documented by Spavins-Hicks et al. (2021)]. However, since the maximum in convergence in the early morning occurs when subsidence dominates the lower troposphere (Fig. 8), other processes may be important.

Fig. 9.
Fig. 9.

Average streamlines [u (m s−1), w (Pa s−1) × 103] across all heat low days in ERA5 from September to March, averaged over 5° latitude centered over the low-level atmospheric thickness (LLAT) maximum of each event to account for the seasonal movement of the heat low. Negative (positive) longitude values represent degrees west (east) from the heat low center.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

The lack of strong daytime ascent is surprising. Typically, model studies and observations of heat lows indicate that strong ascent should occur throughout the day (e.g., Blake et al. 1983; Gaertner et al. 1993; Peyrillé and Lafore 2007). As noted by Howard and Washington (2019), the diurnal cycle of ascent within the heat low occurs out of phase with the diurnal cycle of potential temperature. Little change in potential temperature with height in the afternoon—as seen at 1200 UTC when only 2 K separates potential temperature values at 600 hPa and the surface (Fig. 8)—indicates the atmosphere is at its least statically stable due to turbulent mixing, and thus strong ascent is expected at this time (Rácz and Smith 1999). Howard and Washington (2019) suggest that the absence of this strong daytime ascent may result from turbulence parameterization in reanalysis; at a horizontal resolution of 30 km, ERA5 is unable to explicitly represent subgrid-scale turbulent fluxes that account for daytime ascent in reality (Peyrillé and Lafore 2007). Likewise, MERRA-2 has a spatial resolution of approximately 50 km. Following this argument, ascent in reanalyses is simulated at night when vertical velocity is a product of the large-scale flow but the ascent that occurs as the sum of daytime turbulent fluxes may not be captured. Observational data would be invaluable in assessing the accuracy of the diurnal cycle in reanalysis and establishing whether strong daytime ascent is also absent in reality.

7. Heat low trends from 1960 to 2019

Section 4 defined 462 strong heat low days from 1990 to 2019. Here, we analyze trends in the frequency of strong heat low days in the extended period of 1960–2019 using hourly ERA5 data. Using the same thickness threshold, we find only 168 strong heat low days between 1960 and 1989. This represents a significant 175% increase between the 1960–89 and 1990–2019 study periods (p < 0.0001). Figure 10 depicts the increasing frequency of the strongest heat lows and the corresponding temperature and geopotential height trends. Figure 10b shows a high increase in heat low frequency in the most recent decade; more heat low days were detected in the most recent 5-yr period (2015–19) than in the 30-yr period from 1960 to 1989. A Poisson regression on the annual frequency of strong heat lows from 1960 to 2019 indicates that approximately 42% of the variability in strong heat low frequency can be explained by the temporal trend.

Fig. 10.
Fig. 10.

ERA5 trends in (a) annual average September–March 850-hPa temperature (K) and 400-hPa geopotential height (m) from 1960 to 2019 across southwest Africa (10°–30°S, 12°–24°E) and (b) the detected number of strong heat lows from 1960 to 2019, using the initial detection criteria (detailed in section 4), for each month of the study period.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

To confirm the robustness of this result, we analyze trends in the MERRA-2 data, which are available from 1980. We find a similarly strong, and significant trend (supplemental Fig. 3). In MERRA-2 the increase in the frequency of strong heat lows from 1980 to 1999 compared to 2000–19 is 73.3%. The increase in ERA5 between the same time periods is 74.9%. The similarity in results between the two reanalysis datasets supports the reliability of the trend irrespective of model structure and changes in the observations assimilated into both products. The increasing frequency of strong heat lows in both datasets supports the findings of Vizy et al. (2015), who document a strengthening heat low in southern Africa from 1982 to 2013 in multiple reanalysis products. Lavaysse et al. (2016) and Fonseca et al. (2022) similarly detect increases in the Saharan heat low and Arabian heat low, respectively, in reanalysis data.

Daily time series of temperature at 850 hPa and geopotential height at 400 hPa in ERA5 (in addition to 300- and 500-hPa geopotential height; not shown) for the same period depict a similar trend (Fig. 10a). Both of these trends are determined to be statistically significant at the 0.05 significance level via using a Mann–Kendall test (in both cases, p < 1 × 10−7). The positive lower tropospheric temperature trend of approximately 2°C from 1960 to 2019 implies a strengthening heat low circulation. The trend in 400-hPa geopotential height (an increase of approximately 40 m since 1960) indicates tropospheric expansion in line with this increased heating. Correspondingly, the average LLAT at 1500 UTC across southwest Africa has increased in all months between 1960–89 and 1990–2019 (Fig. 11a). Despite increases in LLAT, no shift in the location of monthly maximum LLAT was detected, further demonstrating the controlling influence of topography on constraining heat low location.

Fig. 11.
Fig. 11.

The difference in 1960–89 and 1990–2019 ERA5 low-level atmospheric thickness (LLAT) at 1500 UTC for (a) the spatial mean across southwest Africa (10°–30°S, 12°–24°E) throughout the annual cycle and (b) the monthly mean spatial change.

Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0522.1

The greatest trend in LLAT is attributable to the early season (September and October), where the mean increase is approaching the same magnitude as the diurnal cycle (Fig. 11b). This corresponds to an increase in the relative contribution of detected September and October heat low days to the seasonal total—by 7.4% and 8.9% respectively from 1960–89 to 1990–2019—and an absolute increase in the number of strong heat low days during these two months of over a factor of 5 (Fig. 10). Evidence of early season strengthening of the heat low is supported by the results of Moses et al. (2023) who identify a warming and drying trend due to a strengthening of the Botswana high in October and November in ERA5. LLAT increases are also noteworthy in February and March, when the number of strong heat low days detected has increased by almost a factor of 3, suggesting an overall lengthening of the season in which strong heat lows occur (Figs. 10 and 11). As heat low events are associated with largely negative precipitation anomalies (Fig. 6), these results support projections of shortening rainy season length over southern Africa and indicate that the heat low circulation may be an important factor in the dynamics of this trend (Engelbrecht et al. 2009; Cook and Vizy 2013; Dunning et al. 2018).

8. Summary and conclusions

This study provides the first detailed account of the characteristics, seasonal variation, and diurnal variability of the southern African heat low. Using ERA5 and MERRA-2 reanalyses, the heat low is shown to be an important feature of the summertime southern African climate, present in the climatology between September and March. The heat low initially forms over Angola, shifting southward to Namibia in the core summer months as moist convection in the rainbelt moves south from the Congo basin (Fig. 5). Analyzing the strongest heat lows yields a daily mean vertical structure common to theoretical and observation-based studies of heat lows: ascent occurs at low levels, capped by subsiding air above (Fig. 2). As this ascent is associated with dry convection, clear skies and intense surface heating, strong heat lows are associated with negative precipitation anomalies across much of southern and central Africa (Fig. 6). Ascent is driven by converging low-level winds that generate a cyclonic circulation below 700 hPa. Above the layer of ascent, outflow generates an anticyclonic circulation. At its diurnal maximum, ascent occurs throughout much of the troposphere (above 500 hPa), refuting the classification of heat lows as only shallow features confined to levels below 600 hPa. This reflects both the high elevation of the southern African plateau and the strength of the circulation during strong heat low days.

Strongest ascent occurs over areas of high elevation, demonstrating the role of orography in constraining the heat low location, as has been noted in model-based studies (Zängl and Chico 2006; Smith and Spengler 2011). Poor resolution of topography may thus explain why low-resolution climate models often fail to reproduce the heat low circulation and generate positive rainfall biases across southern Africa (Munday and Washington 2018). Our results support the conclusion that the ability of models to simulate the location and structure of the heat low may be indicative of their ability to accurately reproduce the rainfall climatology over the region, due to the influence of the heat low on regional precipitation (Fig. 6).

Composite plots of strong heat low days demonstrate a marked diurnal cycle of low-level atmospheric thickness which peaks in the late afternoon following the diurnal cycle of solar radiation (Fig. 7). Accordingly, vertical potential temperature gradients and instability are strongest in the afternoon (Fig. 8). Ascent, however, is weak at this time—out of phase with the afternoon surface potential temperature maximum—in agreement with the results of Howard and Washington (2019). Clarifying whether the lack of daytime ascent is due to turbulence parameterization in reanalysis is an important avenue of future research. If such a signal is present due to the parameterization of subgrid-scale processes, this has implications for the ability of the heat low to be accurately simulated not only in reanalysis datasets but also climate models, which typically operate at coarser resolutions. Considering the lack of observed data with which to constrain models and reanalysis data over Africa, observational analysis of the diurnal cycle would provide invaluable insight into this key feature of the southern African climate. This is the subject of future research.

Analysis of strong heat lows from 1960 to 2019 shows a significantly increasing trend: the number of heat lows detected from 1990 to 2019 has increased by 175% with respect to 1960–89. Over 40% of the variability in strong heat low frequency is attributable to the temporal trend. This is in agreement with surface temperature and midlevel geopotential height trends (Fig. 10). Dunning et al. (2018) suggest that changes to the relative strength of the Saharan and Angolan heat lows may be responsible for future drying trends over southern Africa owing to their influence on the pressure gradient associated with the movement of the tropical rainbelt. Comparing daily LLAT between 1960–89 and 1990–2019 shows greatest increases during early and (to a lesser extent) late austral summer, indicating a lengthening of season in which strong heat lows occur. A continuation of this trend would align with the projected shortening of the rainy season over Africa, considering the occurrence of largely negative precipitation anomalies during strong heat low days. Late rainy season onset is expected to occur due to the delayed breakdown of the Congo air boundary (Howard and Washington 2020). The results presented in this study support the conclusion that the circulation associated with heat lows helps to maintain the Congo air boundary (and vice versa) and indicate that this relationship may be important for future rainfall trends (Howard and Washington 2019).

A strengthening of the southern African heat low is likely to play a role in future amplified warming the region, where rates of temperature change are expected to exceed the global rate (Engelbrecht et al. 2015; Vizy and Cook 2016; Landman et al. 2018; Fan et al. 2021). Alongside the link between heat lows and precipitation, this demonstrates that understanding the regional heat low is essential to constrain model projections of future change across southern Africa in addition to informing forecasting and understanding interannual variability. Such efforts provide vital opportunities to minimize risks to health, infrastructure, and agriculture (e.g., Lazenby et al. 2014; Landman et al. 2018). The increasing frequency of strong heat lows is symptomatic of the rapidly changing nature of the southern African climate system, which requires careful observation to inform responses to changes in seasonality and extremes.

Acknowledgments.

The first author is funded through the U.K. National Environmental Research Council (NERC) Doctoral Training Partnership (NE/S007474/1) and a Met Office CASE Studentship (R87185/CN001). The authors are also supported by the NERC DRY-CAB award (NE/V011928/1).

Data availability statement.

All ERA5 reanalysis data used in this study are available from the Copernicus Climate Data Store at https://doi.org/10.24381/cds.143582cf as documented by Hersbach et al. (2020).

REFERENCES

  • Adebiyi, A. A., and P. Zuidema, 2016: The role of the southern African Easterly Jet in modifying the southeast Atlantic aerosol and cloud environments. Quart. J. Roy. Meteor. Soc., 142, 15741589, https://doi.org/10.1002/qj.2765.

    • Search Google Scholar
    • Export Citation
  • Blake, D. W., T. N. Krishnamurti, S. V. Low-Nam, and J. S. Fein, 1983: Heat low over the Saudi Arabian desert during May 1979 (summer MONEX). Mon. Wea. Rev., 111, 17591775, https://doi.org/10.1175/1520-0493(1983)111<1759:HLOTSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Conway, D., and Coauthors, 2015: Climate and southern Africa’s water–energy–food nexus. Nat. Climate Change, 5, 837846, https://doi.org/10.1038/nclimate2735.

    • Search Google Scholar
    • Export Citation
  • Cook, C., C. J. C. Reason, and B. Hewitson, 2004: Wet and dry spells within particularly wet and dry summers in the South African summer rainfall region. Climate Res., 26, 1731, https://doi.org/10.3354/cr026017.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., and E. K. Vizy, 2013: Projected changes in East African rainy seasons. J. Climate, 26, 59315948, https://doi.org/10.1175/JCLI-D-12-00455.1.

    • Search Google Scholar
    • Export Citation
  • Crétat, J., B. Pohl, B. Dieppois, S. Berthou, and J. Pergaud, 2019: The Angola Low: Relationship with southern African rainfall and ENSO. Climate Dyn., 52, 17831803, https://doi.org/10.1007/s00382-018-4222-3.

    • Search Google Scholar
    • Export Citation
  • Dieppois, B., M. Rouault, and M. New, 2015: The impact of ENSO on southern African rainfall in CMIP5 ocean atmosphere coupled climate models. Climate Dyn., 45, 24252442, https://doi.org/10.1007/s00382-015-2480-x.

    • Search Google Scholar
    • Export Citation
  • Driver, P., and C. J. C. Reason, 2017: Variability in the Botswana high and its relationships with rainfall and temperature characteristics over southern Africa. Int. J. Climatol., 37, 570581, https://doi.org/10.1002/joc.5022.

    • Search Google Scholar
    • Export Citation
  • Dunning, C. M., E. Black, and R. P. Allan, 2018: Later wet seasons with more intense rainfall over Africa under future climate change. J. Climate, 31, 97199738, https://doi.org/10.1175/JCLI-D-18-0102.1.

    • Search Google Scholar
    • Export Citation
  • Engelbrecht, F., J. L. McGregor, and C. J. Engelbrecht, 2009: Dynamics of the conformal-cubic atmospheric model projected climate-change signal over southern Africa. Int. J. Climatol., 29, 10131033, https://doi.org/10.1002/joc.1742.

    • Search Google Scholar
    • Export Citation
  • Engelbrecht, F., and Coauthors, 2015: Projections of rapidly rising surface temperatures over Africa under low mitigation. Environ. Res. Lett., 10, 085004, https://doi.org/10.1088/1748-9326/10/8/085004.

    • Search Google Scholar
    • Export Citation
  • Engelstaedter, S., R. Washington, C. Flamant, D. J. Parker, C. J. T. Allen, and M. C. Todd, 2015: The Saharan heat low and moisture transport pathways in the central Sahara—Multiaircraft observations and Africa-LAM evaluation. J. Geophys. Res. Atmos., 120, 44174442, https://doi.org/10.1002/2015JD023123.

    • Search Google Scholar
    • Export Citation
  • Fan, X., C. Miao, Q. Duan, C. Shen, and Y. Wu, 2021: Future climate change hotspots under different 21st century warming scenarios. Earth’s Future, 9, e2021EF002027, https://doi.org/10.1029/2021EF002027.

    • Search Google Scholar
    • Export Citation
  • Fonseca, R., D. Francis, N. Nelli, and M. Thota, 2022: Climatology of the heat low and the intertropical discontinuity in the Arabian Peninsula. Int. J. Climatol., 42, 10921117, https://doi.org/10.1002/joc.7291.

    • Search Google Scholar
    • Export Citation
  • Gaertner, M. A., C. Fernandez, and M. Castro, 1993: A two-dimensional simulation of the Iberian summer thermal low. Mon. Wea. Rev., 121, 27402756, https://doi.org/10.1175/1520-0493(1993)121<2740:ATDSOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gbode, I. E., T. E. Babalola, G. T. Diro, and J. D. Intsiful, 2023: Assessment of ERA5 and ERA-Interim in reproducing mean and extreme climates over West Africa. Adv. Atmos. Sci., 40, 570586, https://doi.org/10.1007/s00376-022-2161-8.

    • Search Google Scholar
    • Export Citation
  • Gelaro, R., and Coauthors, 2017: The Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2). J. Climate, 30, 54195454, https://doi.org/10.1175/JCLI-D-16-0758.1.

    • Search Google Scholar
    • Export Citation
  • Gleixner, S., T. Demissie, and G. T. Diro, 2020: Did ERA5 improve temperature and precipitation reanalysis over East Africa? Atmosphere, 11, 996, https://doi.org/10.3390/atmos11090996.

    • Search Google Scholar
    • Export Citation
  • Hart, T. L., 1990: Estimates of horizontal divergence and vertical velocity over northern Australia, and implications for biogeophysical feedback. Aust. Meteor. Oceanogr. J., 38, 107122.

    • Search Google Scholar
    • Export Citation
  • Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 19992049, https://doi.org/10.1002/qj.3803.

    • Search Google Scholar
    • Export Citation
  • Hoinka, K. P., and M. D. Castro, 2003: The Iberian Peninsula thermal low. Quart. J. Roy. Meteor. Soc., 129, 14911511, https://doi.org/10.1256/qj.01.189.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2018: Characterizing the synoptic expression of the Angola low. J. Climate, 31, 71477165, https://doi.org/10.1175/JCLI-D-18-0017.1.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2019: Drylines in southern Africa: Rediscovering the Congo air boundary. J. Climate, 32, 82238242, https://doi.org/10.1175/JCLI-D-19-0437.1.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2020: Tracing future spring and summer drying in southern Africa to tropical lows and the Congo Air Boundary. J. Climate, 33, 62056228, https://doi.org/10.1175/JCLI-D-19-0755.1.

    • Search Google Scholar
    • Export Citation
  • Huaman, L., C. Schumacher, A. H. Fink, and E. Buttitta, 2023: Diurnal variations of the meridional overturning circulations over West Africa during the premonsoon and monsoon seasons. Quart. J. Roy. Meteor. Soc., 149, 28202837, https://doi.org/10.1002/qj.4533.

    • Search Google Scholar
    • Export Citation
  • James, R., and Coauthors, 2018: Evaluating climate models with an African lens. Bull. Amer. Meteor. Soc., 99, 313336, https://doi.org/10.1175/BAMS-D-16-0090.1.

    • Search Google Scholar
    • Export Citation
  • Kuete, G., W. Pokam Mba, and R. Washington, 2020: African Easterly Jet South: Control, maintenance mechanisms and link with southern subtropical waves. Climate Dyn., 54, 15391552, https://doi.org/10.1007/s00382-019-05072-w.

    • Search Google Scholar
    • Export Citation
  • Kuete, G., W. Pokam Mba, R. James, E. Dyer, T. Annor, and R. Washington, 2023: How do coupled models represent the African easterly jets and their associated dynamics over central Africa during the September–November rainy season? Climate Dyn., 60, 29072929, https://doi.org/10.1007/s00382-022-06467-y.

    • Search Google Scholar
    • Export Citation
  • Landman, W. A., F. Engelbrecht, B. Hewitson, J. Malherbe, and J. van de Merwe, 2018: Towards bridging the gap between climate change projections and maize producers in South Africa. Theor. Appl. Climatol., 132, 11531163, https://doi.org/10.1007/s00704-017-2168-8.

    • Search Google Scholar
    • Export Citation
  • Lavaysse, C., C. Flamant, S. Janicot, D. J. Parker, J.-P. Lafore, B. Sultan, and J. Pelon, 2009: Seasonal evolution of the West African heat low: A climatological perspective. Climate Dyn., 33, 313330, https://doi.org/10.1007/s00382-009-0553-4.

    • Search Google Scholar
    • Export Citation
  • Lavaysse, C., C. Flamant, A. Evan, S. Janicot, and M. Gaetani, 2016: Recent climatological trend of the Saharan heat low and its impact on the West African climate. Climate Dyn., 47, 34793498, https://doi.org/10.1007/s00382-015-2847-z.

    • Search Google Scholar
    • Export Citation
  • Lavender, S. L., 2017: A climatology of Australian heat low events. Int. J. Climatol., 37, 534539, https://doi.org/10.1002/joc.4692.

  • Lazenby, M. J., W. A. Landman, R. M. Garland, and D. G. DeWitt, 2014: Seasonal temperature prediction skill over southern Africa and human health. Meteor. Appl., 21, 963974, https://doi.org/10.1002/met.1449.

    • Search Google Scholar
    • Export Citation
  • Moses, O., R. C. Blamey, and C. J. C. Reason, 2023: Drought metrics and temperature extremes over the Okavango River basin, southern Africa, and links with the Botswana high. Int. J. Climatol., 43, 64636483, https://doi.org/10.1002/joc.8215.

    • Search Google Scholar
    • Export Citation
  • Mulenga, H. M., 1999: Southern African climate anomalies, summer rainfall and the Angola Low. Ph.D. thesis, University of Cape Town, 232 pp.

  • Munday, C., and R. Washington, 2017: Circulation controls on southern African precipitation in coupled models: The role of the Angola Low. J. Geophys. Res. Atmos., 122, 861877, https://doi.org/10.1002/2016JD025736.

    • Search Google Scholar
    • Export Citation
  • Munday, C., and R. Washington, 2018: Systematic climate model rainfall biases over southern Africa: Links to moisture circulation and topography. J. Climate, 31, 75337548, https://doi.org/10.1175/JCLI-D-18-0008.1.

    • Search Google Scholar
    • Export Citation
  • Munday, C., and R. Washington, 2019: Controls on the diversity in climate model projections of early summer drying over southern Africa. J. Climate, 32, 37073725, https://doi.org/10.1175/JCLI-D-18-0463.1.

    • Search Google Scholar
    • Export Citation
  • Parker, D. J., and Coauthors, 2005: The diurnal cycle of the West African monsoon circulation. Quart. J. Roy. Meteor. Soc., 131, 28392860, https://doi.org/10.1256/qj.04.52.

    • Search Google Scholar
    • Export Citation
  • Pascale, S., B. Pohl, S. B. Kapnick, and H. Zhang, 2019: On the Angola low interannual variability and its role in modulating ENSO effects in southern Africa. J. Climate, 32, 47834803, https://doi.org/10.1175/JCLI-D-18-0745.1.

    • Search Google Scholar
    • Export Citation
  • Peyrillé, P., and J.-P. Lafore, 2007: An idealized two-dimensional framework to study the West African monsoon. Part II: Large-scale advection and the diurnal cycle. J. Atmos. Sci., 64, 27832803, https://doi.org/10.1175/JAS4052.1.

    • Search Google Scholar
    • Export Citation
  • Portela, A., and M. Castro, 1996: Summer thermal lows in the Iberian Peninsula: A three-dimensional simulation. Quart. J. Roy. Meteor. Soc., 122 (529), 122, https://doi.org/10.1002/qj.49712252902.

    • Search Google Scholar
    • Export Citation
  • Rácz, Z., and R. K. Smith, 1999: The dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 125, 225252, https://doi.org/10.1002/qj.49712555313.

    • Search Google Scholar
    • Export Citation
  • Ranasinghe, R., and Coauthors, 2021: Climate change information for regional impact and for risk assessment. Climate Change 2021: The Physical Science Basis, V. Masson Delmotte et al., Eds., Cambridge University Press, 1767–1926, https://doi.org/10.1017/9781009157896.014.

  • Reason, C. J. C., 2016: The Bolivian, Botswana, and Bilybara Highs and Southern Hemisphere drought/floods. Geophys. Res. Lett., 43, 12801286, https://doi.org/10.1002/2015GL067228.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., and D. Jagadheesha, 2005: A model investigation of recent ENSO impacts over southern Africa. Meteor. Atmos. Phys., 89, 181205, https://doi.org/10.1007/s00703-005-0128-9.

    • Search Google Scholar
    • Export Citation
  • Reichmann, Z., and R. K. Smith, 2003: Terrain influences on the dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 129, 17791793, https://doi.org/10.1256/qj.00.194.

    • Search Google Scholar
    • Export Citation
  • Roffe, S. J., and A. J. van der Walt, 2023: Representation and evaluation of southern Africa’s seasonal mean and extreme temperatures in the ERA5-based reanalysis products. Atmos. Res., 284, 106591, https://doi.org/10.1016/j.atmosres.2022.106591.

    • Search Google Scholar
    • Export Citation
  • Rowson, D. R., and S. J. Colucci, 1992: Synoptic climatology of thermal low-pressure systems over south-western North America. Int. J. Climatol., 12, 529545, https://doi.org/10.1002/joc.3370120602.

    • Search Google Scholar
    • Export Citation
  • Smith, E. A., 1986: The structure of the Arabian heat low. Part I: Surface energy budget. Mon. Wea. Rev., 114, 10671083, https://doi.org/10.1175/1520-0493(1986)114<1067:TSOTAH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Smith, R. K., and T. Spengler, 2011: The dynamics of heat lows over elevated terrain. Quart. J. Roy. Meteor. Soc., 137, 250263, https://doi.org/10.1002/qj.737.

    • Search Google Scholar
    • Export Citation
  • Spavins-Hicks, Z. D., R. Washington, and C. Munday, 2021: The Limpopo low-level jet: Mean climatology and role in water vapor transport. J. Geophys. Res. Atmos., 126, e2020JD034364, https://doi.org/10.1029/2020JD034364.

    • Search Google Scholar
    • Export Citation
  • Spengler, T., and R. K. Smith, 2008: The dynamics of heat lows over flat terrain. Quart. J. Roy. Meteor. Soc., 134, 21572172, https://doi.org/10.1002/qj.342.

    • Search Google Scholar
    • Export Citation
  • Steinkopf, J., and F. Engelbrecht, 2022: Verification of ERA5 and ERA-Interim precipitation over Africa at intra-annual and interannual timescales. Atmos. Res., 280, 106427, https://doi.org/10.1016/j.atmosres.2022.106427.

    • Search Google Scholar
    • Export Citation
  • Tyson, P. D., and R. A. Preston-Whyte, 2000: The Weather and Climate of Southern Africa. 2nd ed. Oxford University Press, 396 pp.

  • Van Schalkwyk, L., R. C. Blamey, L. L. Dyson, and C. J. C. Reason, 2022: A climatology of drylines in the interior of subtropical southern Africa. J. Climate, 35, 64116430, https://doi.org/10.1175/JCLI-D-21-1005.1.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., and K. H. Cook, 2016: Understanding long-term (1982–2013) multi-decadal change in the equatorial and subtropical South Atlantic climate. Climate Dyn., 46, 20872113, https://doi.org/10.1007/s00382-015-2691-1.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., K. H. Cook, J. Chimphamba, and B. McCusker, 2015: Projected changes in Malawi’s growing season. Climate Dyn., 45, 16731698, https://doi.org/10.1007/s00382-014-2424-x.

    • Search Google Scholar
    • Export Citation
  • Zängl, G., and S. G. Chico, 2006: The thermal circulation of a grand plateau: Sensitivity to the height, width, and shape of the plateau. Mon. Wea. Rev., 134, 25812600, https://doi.org/10.1175/MWR3207.1.

    • Search Google Scholar
    • Export Citation

Supplementary Materials

Save
  • Adebiyi, A. A., and P. Zuidema, 2016: The role of the southern African Easterly Jet in modifying the southeast Atlantic aerosol and cloud environments. Quart. J. Roy. Meteor. Soc., 142, 15741589, https://doi.org/10.1002/qj.2765.

    • Search Google Scholar
    • Export Citation
  • Blake, D. W., T. N. Krishnamurti, S. V. Low-Nam, and J. S. Fein, 1983: Heat low over the Saudi Arabian desert during May 1979 (summer MONEX). Mon. Wea. Rev., 111, 17591775, https://doi.org/10.1175/1520-0493(1983)111<1759:HLOTSA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Conway, D., and Coauthors, 2015: Climate and southern Africa’s water–energy–food nexus. Nat. Climate Change, 5, 837846, https://doi.org/10.1038/nclimate2735.

    • Search Google Scholar
    • Export Citation
  • Cook, C., C. J. C. Reason, and B. Hewitson, 2004: Wet and dry spells within particularly wet and dry summers in the South African summer rainfall region. Climate Res., 26, 1731, https://doi.org/10.3354/cr026017.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., and E. K. Vizy, 2013: Projected changes in East African rainy seasons. J. Climate, 26, 59315948, https://doi.org/10.1175/JCLI-D-12-00455.1.

    • Search Google Scholar
    • Export Citation
  • Crétat, J., B. Pohl, B. Dieppois, S. Berthou, and J. Pergaud, 2019: The Angola Low: Relationship with southern African rainfall and ENSO. Climate Dyn., 52, 17831803, https://doi.org/10.1007/s00382-018-4222-3.

    • Search Google Scholar
    • Export Citation
  • Dieppois, B., M. Rouault, and M. New, 2015: The impact of ENSO on southern African rainfall in CMIP5 ocean atmosphere coupled climate models. Climate Dyn., 45, 24252442, https://doi.org/10.1007/s00382-015-2480-x.

    • Search Google Scholar
    • Export Citation
  • Driver, P., and C. J. C. Reason, 2017: Variability in the Botswana high and its relationships with rainfall and temperature characteristics over southern Africa. Int. J. Climatol., 37, 570581, https://doi.org/10.1002/joc.5022.

    • Search Google Scholar
    • Export Citation
  • Dunning, C. M., E. Black, and R. P. Allan, 2018: Later wet seasons with more intense rainfall over Africa under future climate change. J. Climate, 31, 97199738, https://doi.org/10.1175/JCLI-D-18-0102.1.

    • Search Google Scholar
    • Export Citation
  • Engelbrecht, F., J. L. McGregor, and C. J. Engelbrecht, 2009: Dynamics of the conformal-cubic atmospheric model projected climate-change signal over southern Africa. Int. J. Climatol., 29, 10131033, https://doi.org/10.1002/joc.1742.

    • Search Google Scholar
    • Export Citation
  • Engelbrecht, F., and Coauthors, 2015: Projections of rapidly rising surface temperatures over Africa under low mitigation. Environ. Res. Lett., 10, 085004, https://doi.org/10.1088/1748-9326/10/8/085004.

    • Search Google Scholar
    • Export Citation
  • Engelstaedter, S., R. Washington, C. Flamant, D. J. Parker, C. J. T. Allen, and M. C. Todd, 2015: The Saharan heat low and moisture transport pathways in the central Sahara—Multiaircraft observations and Africa-LAM evaluation. J. Geophys. Res. Atmos., 120, 44174442, https://doi.org/10.1002/2015JD023123.

    • Search Google Scholar
    • Export Citation
  • Fan, X., C. Miao, Q. Duan, C. Shen, and Y. Wu, 2021: Future climate change hotspots under different 21st century warming scenarios. Earth’s Future, 9, e2021EF002027, https://doi.org/10.1029/2021EF002027.

    • Search Google Scholar
    • Export Citation
  • Fonseca, R., D. Francis, N. Nelli, and M. Thota, 2022: Climatology of the heat low and the intertropical discontinuity in the Arabian Peninsula. Int. J. Climatol., 42, 10921117, https://doi.org/10.1002/joc.7291.

    • Search Google Scholar
    • Export Citation
  • Gaertner, M. A., C. Fernandez, and M. Castro, 1993: A two-dimensional simulation of the Iberian summer thermal low. Mon. Wea. Rev., 121, 27402756, https://doi.org/10.1175/1520-0493(1993)121<2740:ATDSOT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gbode, I. E., T. E. Babalola, G. T. Diro, and J. D. Intsiful, 2023: Assessment of ERA5 and ERA-Interim in reproducing mean and extreme climates over West Africa. Adv. Atmos. Sci., 40, 570586, https://doi.org/10.1007/s00376-022-2161-8.

    • Search Google Scholar
    • Export Citation
  • Gelaro, R., and Coauthors, 2017: The Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2). J. Climate, 30, 54195454, https://doi.org/10.1175/JCLI-D-16-0758.1.

    • Search Google Scholar
    • Export Citation
  • Gleixner, S., T. Demissie, and G. T. Diro, 2020: Did ERA5 improve temperature and precipitation reanalysis over East Africa? Atmosphere, 11, 996, https://doi.org/10.3390/atmos11090996.

    • Search Google Scholar
    • Export Citation
  • Hart, T. L., 1990: Estimates of horizontal divergence and vertical velocity over northern Australia, and implications for biogeophysical feedback. Aust. Meteor. Oceanogr. J., 38, 107122.

    • Search Google Scholar
    • Export Citation
  • Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 19992049, https://doi.org/10.1002/qj.3803.

    • Search Google Scholar
    • Export Citation
  • Hoinka, K. P., and M. D. Castro, 2003: The Iberian Peninsula thermal low. Quart. J. Roy. Meteor. Soc., 129, 14911511, https://doi.org/10.1256/qj.01.189.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2018: Characterizing the synoptic expression of the Angola low. J. Climate, 31, 71477165, https://doi.org/10.1175/JCLI-D-18-0017.1.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2019: Drylines in southern Africa: Rediscovering the Congo air boundary. J. Climate, 32, 82238242, https://doi.org/10.1175/JCLI-D-19-0437.1.

    • Search Google Scholar
    • Export Citation
  • Howard, E., and R. Washington, 2020: Tracing future spring and summer drying in southern Africa to tropical lows and the Congo Air Boundary. J. Climate, 33, 62056228, https://doi.org/10.1175/JCLI-D-19-0755.1.

    • Search Google Scholar
    • Export Citation
  • Huaman, L., C. Schumacher, A. H. Fink, and E. Buttitta, 2023: Diurnal variations of the meridional overturning circulations over West Africa during the premonsoon and monsoon seasons. Quart. J. Roy. Meteor. Soc., 149, 28202837, https://doi.org/10.1002/qj.4533.

    • Search Google Scholar
    • Export Citation
  • James, R., and Coauthors, 2018: Evaluating climate models with an African lens. Bull. Amer. Meteor. Soc., 99, 313336, https://doi.org/10.1175/BAMS-D-16-0090.1.

    • Search Google Scholar
    • Export Citation
  • Kuete, G., W. Pokam Mba, and R. Washington, 2020: African Easterly Jet South: Control, maintenance mechanisms and link with southern subtropical waves. Climate Dyn., 54, 15391552, https://doi.org/10.1007/s00382-019-05072-w.

    • Search Google Scholar
    • Export Citation
  • Kuete, G., W. Pokam Mba, R. James, E. Dyer, T. Annor, and R. Washington, 2023: How do coupled models represent the African easterly jets and their associated dynamics over central Africa during the September–November rainy season? Climate Dyn., 60, 29072929, https://doi.org/10.1007/s00382-022-06467-y.

    • Search Google Scholar
    • Export Citation
  • Landman, W. A., F. Engelbrecht, B. Hewitson, J. Malherbe, and J. van de Merwe, 2018: Towards bridging the gap between climate change projections and maize producers in South Africa. Theor. Appl. Climatol., 132, 11531163, https://doi.org/10.1007/s00704-017-2168-8.

    • Search Google Scholar
    • Export Citation
  • Lavaysse, C., C. Flamant, S. Janicot, D. J. Parker, J.-P. Lafore, B. Sultan, and J. Pelon, 2009: Seasonal evolution of the West African heat low: A climatological perspective. Climate Dyn., 33, 313330, https://doi.org/10.1007/s00382-009-0553-4.

    • Search Google Scholar
    • Export Citation
  • Lavaysse, C., C. Flamant, A. Evan, S. Janicot, and M. Gaetani, 2016: Recent climatological trend of the Saharan heat low and its impact on the West African climate. Climate Dyn., 47, 34793498, https://doi.org/10.1007/s00382-015-2847-z.

    • Search Google Scholar
    • Export Citation
  • Lavender, S. L., 2017: A climatology of Australian heat low events. Int. J. Climatol., 37, 534539, https://doi.org/10.1002/joc.4692.

  • Lazenby, M. J., W. A. Landman, R. M. Garland, and D. G. DeWitt, 2014: Seasonal temperature prediction skill over southern Africa and human health. Meteor. Appl., 21, 963974, https://doi.org/10.1002/met.1449.

    • Search Google Scholar
    • Export Citation
  • Moses, O., R. C. Blamey, and C. J. C. Reason, 2023: Drought metrics and temperature extremes over the Okavango River basin, southern Africa, and links with the Botswana high. Int. J. Climatol., 43, 64636483, https://doi.org/10.1002/joc.8215.

    • Search Google Scholar
    • Export Citation
  • Mulenga, H. M., 1999: Southern African climate anomalies, summer rainfall and the Angola Low. Ph.D. thesis, University of Cape Town, 232 pp.

  • Munday, C., and R. Washington, 2017: Circulation controls on southern African precipitation in coupled models: The role of the Angola Low. J. Geophys. Res. Atmos., 122, 861877, https://doi.org/10.1002/2016JD025736.

    • Search Google Scholar
    • Export Citation
  • Munday, C., and R. Washington, 2018: Systematic climate model rainfall biases over southern Africa: Links to moisture circulation and topography. J. Climate, 31, 75337548, https://doi.org/10.1175/JCLI-D-18-0008.1.

    • Search Google Scholar
    • Export Citation
  • Munday, C., and R. Washington, 2019: Controls on the diversity in climate model projections of early summer drying over southern Africa. J. Climate, 32, 37073725, https://doi.org/10.1175/JCLI-D-18-0463.1.

    • Search Google Scholar
    • Export Citation
  • Parker, D. J., and Coauthors, 2005: The diurnal cycle of the West African monsoon circulation. Quart. J. Roy. Meteor. Soc., 131, 28392860, https://doi.org/10.1256/qj.04.52.

    • Search Google Scholar
    • Export Citation
  • Pascale, S., B. Pohl, S. B. Kapnick, and H. Zhang, 2019: On the Angola low interannual variability and its role in modulating ENSO effects in southern Africa. J. Climate, 32, 47834803, https://doi.org/10.1175/JCLI-D-18-0745.1.

    • Search Google Scholar
    • Export Citation
  • Peyrillé, P., and J.-P. Lafore, 2007: An idealized two-dimensional framework to study the West African monsoon. Part II: Large-scale advection and the diurnal cycle. J. Atmos. Sci., 64, 27832803, https://doi.org/10.1175/JAS4052.1.

    • Search Google Scholar
    • Export Citation
  • Portela, A., and M. Castro, 1996: Summer thermal lows in the Iberian Peninsula: A three-dimensional simulation. Quart. J. Roy. Meteor. Soc., 122 (529), 122, https://doi.org/10.1002/qj.49712252902.

    • Search Google Scholar
    • Export Citation
  • Rácz, Z., and R. K. Smith, 1999: The dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 125, 225252, https://doi.org/10.1002/qj.49712555313.

    • Search Google Scholar
    • Export Citation
  • Ranasinghe, R., and Coauthors, 2021: Climate change information for regional impact and for risk assessment. Climate Change 2021: The Physical Science Basis, V. Masson Delmotte et al., Eds., Cambridge University Press, 1767–1926, https://doi.org/10.1017/9781009157896.014.

  • Reason, C. J. C., 2016: The Bolivian, Botswana, and Bilybara Highs and Southern Hemisphere drought/floods. Geophys. Res. Lett., 43, 12801286, https://doi.org/10.1002/2015GL067228.

    • Search Google Scholar
    • Export Citation
  • Reason, C. J. C., and D. Jagadheesha, 2005: A model investigation of recent ENSO impacts over southern Africa. Meteor. Atmos. Phys., 89, 181205, https://doi.org/10.1007/s00703-005-0128-9.

    • Search Google Scholar
    • Export Citation
  • Reichmann, Z., and R. K. Smith, 2003: Terrain influences on the dynamics of heat lows. Quart. J. Roy. Meteor. Soc., 129, 17791793, https://doi.org/10.1256/qj.00.194.

    • Search Google Scholar
    • Export Citation
  • Roffe, S. J., and A. J. van der Walt, 2023: Representation and evaluation of southern Africa’s seasonal mean and extreme temperatures in the ERA5-based reanalysis products. Atmos. Res., 284, 106591, https://doi.org/10.1016/j.atmosres.2022.106591.

    • Search Google Scholar
    • Export Citation
  • Rowson, D. R., and S. J. Colucci, 1992: Synoptic climatology of thermal low-pressure systems over south-western North America. Int. J. Climatol., 12, 529545, https://doi.org/10.1002/joc.3370120602.

    • Search Google Scholar
    • Export Citation
  • Smith, E. A., 1986: The structure of the Arabian heat low. Part I: Surface energy budget. Mon. Wea. Rev., 114, 10671083, https://doi.org/10.1175/1520-0493(1986)114<1067:TSOTAH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Smith, R. K., and T. Spengler, 2011: The dynamics of heat lows over elevated terrain. Quart. J. Roy. Meteor. Soc., 137, 250263, https://doi.org/10.1002/qj.737.

    • Search Google Scholar
    • Export Citation
  • Spavins-Hicks, Z. D., R. Washington, and C. Munday, 2021: The Limpopo low-level jet: Mean climatology and role in water vapor transport. J. Geophys. Res. Atmos., 126, e2020JD034364, https://doi.org/10.1029/2020JD034364.

    • Search Google Scholar
    • Export Citation
  • Spengler, T., and R. K. Smith, 2008: The dynamics of heat lows over flat terrain. Quart. J. Roy. Meteor. Soc., 134, 21572172, https://doi.org/10.1002/qj.342.

    • Search Google Scholar
    • Export Citation
  • Steinkopf, J., and F. Engelbrecht, 2022: Verification of ERA5 and ERA-Interim precipitation over Africa at intra-annual and interannual timescales. Atmos. Res., 280, 106427, https://doi.org/10.1016/j.atmosres.2022.106427.

    • Search Google Scholar
    • Export Citation
  • Tyson, P. D., and R. A. Preston-Whyte, 2000: The Weather and Climate of Southern Africa. 2nd ed. Oxford University Press, 396 pp.

  • Van Schalkwyk, L., R. C. Blamey, L. L. Dyson, and C. J. C. Reason, 2022: A climatology of drylines in the interior of subtropical southern Africa. J. Climate, 35, 64116430, https://doi.org/10.1175/JCLI-D-21-1005.1.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., and K. H. Cook, 2016: Understanding long-term (1982–2013) multi-decadal change in the equatorial and subtropical South Atlantic climate. Climate Dyn., 46, 20872113, https://doi.org/10.1007/s00382-015-2691-1.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., K. H. Cook, J. Chimphamba, and B. McCusker, 2015: Projected changes in Malawi’s growing season. Climate Dyn., 45, 16731698, https://doi.org/10.1007/s00382-014-2424-x.

    • Search Google Scholar
    • Export Citation
  • Zängl, G., and S. G. Chico, 2006: The thermal circulation of a grand plateau: Sensitivity to the height, width, and shape of the plateau. Mon. Wea. Rev., 134, 25812600, https://doi.org/10.1175/MWR3207.1.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    The location of detected strong heat lows in southwest Africa in ERA5 from September to March 1990–2019, defined by the grid point of maximum low-level atmospheric thickness (LLAT). The size of the marker corresponds to the number of heat lows at that grid point.

  • Fig. 2.

    Composite daily mean latitude–height vertical profiles of strong heat low days in ERA5 from September to March for potential temperature (K), vertical velocity (Pa s−1), zonal wind (m s−1), and meridional wind (m s−1). Profiles are averaged over 5° and centered over the LLAT maximum to account for the seasonal movement of the heat low. Negative (positive) latitude values represent degrees north (south) from the heat low center. Negative (positive) longitude values represent degrees west (east) from the heat low center.

  • Fig. 3.

    The annual cycle of LLAT at 1500 UTC for southwest Africa (10°–30°S, 12°–24°E) in ERA5 and MERRA-2. The red and blue lines denote the multiyear mean from 1990 to 2019 of ERA5 and MERRA-2, respectively, and individual lines show each year for ERA5 only.

  • Fig. 4.

    The annual cycle of LLAT in ERA5 at 1500 UTC averaged from 1990 to 2019. Dashed contours denote the 95th percentile, and solid contours denote the 99th percentile of LLAT relative to each month.

  • Fig. 5.

    Average daily mean latitude–height composites of vertical velocity (Pa s−1), specific humidity (g kg−1; solid black lines), and potential temperature (K; dotted blue lines) during strong heat low days (n = 462) in ERA5, averaged from 12° to 24°E. Red shades indicate negative (upward) vertical velocity. Black shading indicates averaged topography.

  • Fig. 6.

    ERA5 daily average precipitation anomalies (mm day−1) during heat low days from September to March, relative to the 1990–2019 climatology. Stippling denotes areas where the precipitation anomaly exceeds the 0.05 significance level using a Mann–Whitney U test and where the monthly average precipitation exceeds 30 mm.

  • Fig. 7.

    Average hourly ERA5 LLAT across southwest Africa (10°–30°S, 12°–24°E) from 1990 to 2019.

  • Fig. 8.

    Cross sections of ERA5 (left) vertical velocity (Pa s−1) and contours of potential temperature (K), (center) meridional winds (m s−1), and (right) zonal winds (m s−1) during detected heat low events from September to March, at 6-hourly (UTC) timesteps. Data are centered over the LLAT maximum of each event, averaged over 5° longitude/latitude to account for the seasonal movement of the heat low. Green/blue shades of meridional wind (center) indicate positive values (northward flow). Green shades of zonal wind (right) indicate positive (westerly) flow. Negative (positive) latitude values represent degrees north (south) from the heat low center. Negative (positive) longitude values represent degrees west (east) from the heat low center.

  • Fig. 9.

    Average streamlines [u (m s−1), w (Pa s−1) × 103] across all heat low days in ERA5 from September to March, averaged over 5° latitude centered over the low-level atmospheric thickness (LLAT) maximum of each event to account for the seasonal movement of the heat low. Negative (positive) longitude values represent degrees west (east) from the heat low center.

  • Fig. 10.

    ERA5 trends in (a) annual average September–March 850-hPa temperature (K) and 400-hPa geopotential height (m) from 1960 to 2019 across southwest Africa (10°–30°S, 12°–24°E) and (b) the detected number of strong heat lows from 1960 to 2019, using the initial detection criteria (detailed in section 4), for each month of the study period.

  • Fig. 11.

    The difference in 1960–89 and 1990–2019 ERA5 low-level atmospheric thickness (LLAT) at 1500 UTC for (a) the spatial mean across southwest Africa (10°–30°S, 12°–24°E) throughout the annual cycle and (b) the monthly mean spatial change.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 650 650 108
PDF Downloads 663 663 89