The onset and end of wet seasons over Africa

Email: c.m.dunning@pgr.reading.ac.uk

For many Africans, the timing of the wet season is of crucial importance, especially for those reliant upon subsistence agriculture, who depend on the seasonal rains for crop irrigation. In addition, the wet season recharges lakes, rivers and water storage tanks which constitute the domestic water supply in some areas. The timing of the wet season also affects the availability of energy from hydroelectric schemes, and has impacts upon the prevalence of certain disease carrying vectors, such as mosquitoes.

Climate change is already threatening many vulnerable populations, and changes in the timing or intensity of the wet season, or increasing uncertainty in the timing of the onset, may lead to significant socio-economic impacts. But before we consider future projections or past changes in the seasonality, we need to go back a few steps.

The first step is to find a method for determining when the wet season starts and ends (its ‘onset’ and ‘cessation’). In order to look at large-scale shifts in the timing of the wet season and relate this to wider-scale drivers, this method needs to be applicable across the entirety of continental Africa. Most previous methods for determining the onset focus on the national to regional scale, and are dependent on the exceedance of a certain threshold e.g. the first week with at least 20mm of rainfall, with one rainfall event of more than 10mm, and no dry spell of more than 10 days after the rain event for the next month. While such definitions work well at a national scale they are not applicable at a continental scale where rainfall amounts vary substantially. A threshold suitable for the dry countries at the fringes of the Sahara would not be suitable in the wetter East African highlands.

In addition to a vast range of rainfall amounts, the African continent also spans multiple climatic regimes. The seasonal cycle of precipitation over continental Africa is largely driven by the seasonal progression of the ITCZ and associated rain belts, which follows the maximum incoming solar radiation. In the boreal summer, when the thermal equator sits between the equator and the Tropic of Cancer, the ITCZ sits north of the equator and West Africa and the Sahel experience a wet season. During the boreal autumn the ITCZ moves south, and southern Africa experiences a wet season during the austral summer, followed by the northward return of the ITCZ during the boreal spring. As a consequence of this, central African regions and the Horn of Africa experience two wet seasons per year – one as the ITCZ travels north, and a second as the ITCZ travels south. A method for determining the onset and cessation at the continental scale thus needs to account for regions with multiple wet seasons per year.

In our paper (available here) we propose such a method, based on the method of Liebmann et al (2012). The method has three steps:

  • Firstly, determine the number of seasons experienced per year at the location (or grid point) of interest. This is achieved using harmonic analysis – the amplitude of the first and second harmonic were computed, using the entire timeseries and their ratio compared. If the ratio was greater than 1.0, i.e. the amplitude of the second harmonic was greater than the amplitude of the first harmonic then the grid point was defined as having two wet seasons per year (biannual), if the ratio was less than one then it was defined as having an annual regime. Figure 1 shows the ratio for one African rainfall dataset (TARCATv2). Three regions are identified as biannual regions; the Horn of Africa, an equatorial strip extending from Gabon to Uganda and a small region on the southern West African coastline.

    blog_fig1
    Figure 1: Location of regions with one and two seasons per year, determined using harmonic analysis. Yellow indicates two seasons per year, while pink/purple indicates one season per year. Computed from TARCATv2 data.
  • Secondly the period of the year when the wet season occurs was determined. This was achieved by looking for minima and maxima in the climatological cumulative daily rainfall anomaly to identify one or two seasons.
  • The third and final stage is to calculate the onset and cessation dates for each year. This is done by looking for the minima and maxima in the cumulative daily rainfall anomaly, calculated for each season.

Figure 2 shows the seasonal progression of the onset and cessation, with the patterns observed in agreement with those expected from the driving physical mechanisms, and continuous progression across the annual/biannual boundaries. Over West Africa and the Sahel, Figure 2a-b shows zonally-contiguous progression patterns with onset following the onset of the long rains and moving north, and cessation moving southward, preceding the end of the short rains. Over southern Africa Figure 2c-d shows the onset over southern Africa starting in the north-west and south-east, following the onset of the short rains, reaching the East African coast last, and cessation starting at the Zimbabwe, Mozambique, South Africa border and spreading out radially into the cessation of the long rains.

As well as testing the method for compatibility with known physical drivers of African rainfall, agreement across multiple satellite-based rainfall estimates was also examined. In general, good agreement was found across the datasets, particularly for regions with an annual regime and over the biannual region of East Africa.

blog_fig2
Figure 2: Southward and northward progression of the onset and cessation across the annual/biannual boundaries, computed using GPCP daily rainfall data 1998-2013.

The advantage of having a method that works at the continental scale is the ability to look at the impact of large-scale oscillations on wider-scale variability. One application of this method was to investigate the impact of El Niño upon both the annual rains and short rains (Figure 3). In Figure 3 we see the well-documented dipole in rainfall anomaly, with higher rainfall totals over 0–15°S and the Horn of Africa in El Niño years and the opposite between 15°S and 30°S.  This anomaly is stronger when we use this method compared with using standard meteorological seasons. We can also see that while the lower rainfall to the south is colocated with later onset dates and a consequentially shorter season, the higher rainfall over the Horn of Africa is associated with later cessation of the short rains, with only small differences in onset date.

blog_fig3
Figure 3: a-c) Composite of onset, cessation and wet season rainfall in El Niño years for annual rains and short rains, minus the mean over 1982-2013, computed using CHIRPS data d) Oct-Feb rainfall anomaly in  years (CHIRPS).

In addition to using this method for research purposes, its application within an operational setting is also being explored. Hopefully, the method will be included within the Rainwatch platform, which will be able to provide users with a probabilistic estimate of whether or not the season has started, based on the rainfall experienced so far that year, and historical rainfall data.

For more details, please see the paper detailing this work:

Dunning, C.M., E Black, and R.P. Allan (2016) The onset and cessation of seasonal rainfall over Africa, Journal of Geophysical Research: Atmospheres, 121 11,405-11,424, doi: 10.1002/2016JD025428

References:

Liebmann, B., I. Bladé, G. N. Kiladis, L. M. Carvalho, G. B. Senay, D. Allured, S. Leroux, and C. Funk (2012), Seasonality of African precipitation from 1996 to 2009, J. Clim.25(12), 4304–4322.

Stationary Orographic Rainbands

Email: c.j.wright@pgr.reading.ac.uk

Small-scale rainbands often form downwind of mountainous terrain. Although relatively small in scale (a few tens of km across by up to ~100 km in length), these often poorly forecast bands can cause localised flooding as they can be associated with intense precipitation over several hours due to the anchoring effect of orography (Barrett et al., 2013).   Figure 1 shows a flash flood caused by a rainband situated over Cockermouth in 2009.  In some regions of southern France orographic banded convection can contribute 40% of the total rainfall (Cosma et al., 2002).  Rainbands occur in various locations and under different synoptic regimes and environmental conditions making them difficult to examine their properties and determine their occurrence in a systematic way (Kirshbaum et al. 2007a,b, Fairman et al. 2016).  My PhD considers the ability of current operational forecast models to represent these bands and the environmental controls on their formation.

blogfig1
Figure 1: Flash flood event caused by a rainband situated over Cockermouth, Cumbria, UK in 2009

 

What is a rainband?

  • A cloud and precipitation structure associated with an area of rainfall which is significantly elongated
  • Stationary (situated over the same location) with continuous triggering
  • Can form in response to moist, unstable air following over complex terrain
  • Narrow in width ~2-10 km with varying length scales from 10 – 100’s km

 

blogfig2
Figure 2: Schematic showing the difference between cellular and banded convection

To examine the ability of current operational forecast models to represent these bands a case study was chosen which was first introduced by Barrett, et al. (2016).  The radar observations during the event showed a clear band along The Great Glen Fault, Scotland (Figure 3).  However, Barrett, et al. (2016) concluded that neither the operational forecast or the operational ensemble forecast captured the nature of the rainband.  For more information on ensemble models see one of our previous blog posts by David Flack Showers: How well can we predict them?.

blogfig3
Figure 3: Radar observations of precipitation accumulation over a six hour period (between 3-9 am) showing a rainband located over The Great Glen Fault, Scotland on 29 December 2012.

Localised convergence and increased convective available potential energy along the fault supported the formation of the rainband.  To determine the effect of model resolution on the model’s representation of the rainband, a forecast was performed with the horizontal gird spacing decreased to 500 m from 1.5 km.  In this forecast a rainband formed in the correct location which generated precipitation accumulations close to those observed, but with a time displacement.  The robustness of this forecast skill improvement is being assessed by performing an ensemble of these convection-permitting simulations.  Results suggest that accurate representation of these mesoscale rainbands requires resolutions higher than those used operationally by national weather centres.

Idealised numerical simulations have been used to investigate the environmental conditions leading to the formation of these rainbands.  The theoretical dependence of the partitioning of dry flow over and around mountains on the non-dimensional mountain height is well understood.  For this project I examine the effect of this dependence on rainband formation in a moist environment.  Preliminary analysis of the results show that the characteristics of rainbands are controlled by more than just the non-dimensional mountain height, even though this parameter is known to be sufficient to determine flow behaviour relative to mountains.

This work has been funded by the Natural Environmental Research Council (NERC) under the project PREcipitation STructures over Orography (PRESTO), for more project information click here.

References

Barrett, A. I., S. L. Gray, D. J. Kirshbaum, N. M. Roberts, D. M. Schultz, and J. G. Fairman, 2015: Synoptic Versus Orographic Control on Stationary Convective Banding. Quart. J. Roy. Meteorol. Soc., 141, 1101–1113, doi:10.1002/qj.2409.

— 2016: The Utility of Convection-Permitting Ensembles for the Prediction of Stationary Convective Bands. Mon. Wea. Rev., 144, 10931114, doi:10.1175/MWR-D-15-0148.1.

Cosma, S., E. Richard, and F. Minsicloux, 2002: The Role of Small-Scale Orographic Features in the Spatial Distribution of Precipitation. Quart. J. Roy. Meteorol. Soc., 128, 75–92, doi:10.1256/00359000260498798.

Fairman, J. G., D. M. Schultz, D. J. Kirshbaum, S. L. Gray, and A. I. Barrett, 2016: Climatology of Banded Precipitation over the Contiguous United States. Mon. Wea. Rev., 144,4553–4568, doi: 10.1175/MWR-D-16-0015.1.

Kirshbaum, D. J., G. H. Bryan, R. Rotunno, and D. R. Durran, 2007a: The Triggering of Orographic Rainbands by Small-Scale Topography. J. Atmos. Sci., 64, 1530–1549, doi:10.1175/JAS3924.1.

Kirshbaum, D. J., R. Rotunno, and G. H. Bryan, 2007b: The Spacing of Orographic Rainbands Triggered by Small-Scale Topography. J. Atmos. Sci., 64, 4222–4245, doi:10.1175/2007JAS2335.1.

Showers: How well can we predict them?

Email: d.l.a.flack@pgr.reading.ac.uk

Showers are one of the many examples of convective events experienced in the UK, other such events include thunderstorms, supercells and squall lines. These type of events form most often in the summer but can also form over the sea in the winter. They form because the atmosphere is unstable, i.e. warm air over a cooler surface, this results in the creation of thermals. If there is enough water vapour in the air and the thermal reaches high enough the water vapour will condense and eventually form a convective cloud. Convective events produce intense, often very localised, rainfall, which can result in flash floods, e.g. Boscastle 2004.

boscastle04
Boscastle flood 2004 – BBC News

Flash floods are very difficult to predict, unlike flood events that happen from the autumnal and winter storms e.g. floods from Storms Desmond and Frank last winter, and the current floods (20-22 November). So often there is limited lead time for emergency services to react to flash flood events. One of the main reasons why flash floods are difficult to predict is the association with convective events because these events only last for a few hours (6 hours at the longest) and only affect a very small area.

One of the aspects of forecasting the weather that researchers look into is the predictability of certain events. My PhD considers the predictability of convective events within different situations in the UK.

The different situations I am considering are generally split into two regimes: convective quasi-equilibrium and non-equilibrium convection.

In convective quasi-equilibrium any production of instability in the atmosphere is balanced by its release (Arakawa and Schubert, 1974). This results in scattered showers, which could turn up anywhere in a region where there is large-scale ascent. This is typical of areas behind fronts and to the left of jet stream exit regions. Because there are no obvious triggers (like flow over mountains or cliffs) you can’t pin-point the exact location of a shower.  We often find ourselves in this sort of situation in April, hence April showers.

equilibrium
Classic convective quasi-equilibrium conditions in the UK – scattered showers on 20 April 2012 – Dundee Satellite Receiving Station

On the other hand in non-equilibrium convection the instability is blocked from being released so energy in the system builds-up over time. If this inhibiting factor is overcome all the instability can be released at once and will result in ‘explosive’ convection (Emanuel, 1994).  Overcoming the inhibiting factor usually takes place locally, such as a sea breeze or flow up mountains, etc. so these give distinct triggers and help tie the location of these events down. These are the type of situations that occur frequently over continents in the spring and often result in severe weather.

nonequilibrium
Non-equilibrium convection – convergence line along the North Cornish Coast, 2 August 2013 – Dundee Satellite Receiving Station

It’s useful having these regimes to categorise events to help determine what happens in the forecasts of different situations but only if we understand a little bit about their characteristics. For the initial part of my work I considered the regimes over the British Isles and found that  we mainly have convective events in convective quasi-equilibrium (showers) – on average roughly 85% of convective events in the summer are in this regime (Flack et al., 2016). Therefore it is pertinent to ask how well can we predict showers?

To see how well we can predict showers, and other types of convection, the forecast itself is examined. This is done by adding small-scale variability into the model, throughout the forecast, to determine what would happen if the starting conditions (or any other time in the model) changed. This is run a number of times to create an ensemble.

ensembles
Deterministic forecast vs Ensemble forecast schematic, dotted lines represent model trajectories, the bright red represents the truth, darker red represents the forecast

Using ensembles we can determine the uncertainty in the weather forecast, this can either be in terms of spatial positioning, timing or intensity of the event. My work has mainly considered the spatial positioning and intensity of the convection, and is to be submitted shortly to Monthly Weather Review. The intensity in my ensemble shows similar variation in both regimes, suggesting that there are times when the amount of rainfall predicted can be spot on. Most of the interesting results appear to be linked to the location of the events. The ensembles for the non-equilibrium cases generally show agreement between location of the events, so we can be fairly confident about their location (so here your weather app would be very good). On the other hand, when it comes to showers there is no consistency between the different forecasts so they could occur anywhere  (so when your app suggests showers be careful – you may or may not get one).

So I’ll answer my question that I originally posed with another question: What do you want from a forecast? If the answer to this question is “I want to know if there is a chance of rain at my location” then yes we can predict that you might get caught by a shower. If on the other hand your answer is “I want exact details, for my exact location, e.g. is there going to be a shower at 15:01 on Saturday at Stonehenge yes or no?” Then the answer is, although we are improving forecasts, we can’t give that accurate a forecast when it comes to scattered showers, simply because of their very nature.

With forecasts improving all the time and the fact that they are looking more realistic it does not mean that every detail of a forecast is perfect. As with forecasting in all areas (from politics to economy) things can take an unexpected turn so caution is advised. When it comes to the original question of showers then it’s always best to be prepared.

This work has been funded by the Natural Environmental Research Council under the project Flooding From Intense Rainfall, for more project details and project specific blogs visit: www.met.reading.ac.uk/flooding

References

Arakawa, A. and W. H. Schubert, 1974: Interaction of a Cumulus Cloud Ensemble with the Large-Scale Environment, Part I. J. Atmos. Sci., 31, 674-701.

Emanuel, K. A., 1994: Atmospheric convection, Oxford University Press, 580 pp.

Flack, D. L. A., R. S. Plant, S.L. Gray, H. W. Lean, C. Keil and G. C. Craig, 2016: Characterisation of Convective Regimes over the British Isles. Quart. J. Roy. Meteorol. Soc., 142, 1541-1553.  

 

NERC Into the Blue – the Science We Live and Breathe

Email: d.l.a.flack@pgr.reading.ac.uk

One of the key aspects of science is communicating our work, not only to other scientists but also to the public. As part of the Manchester Science Festival the Natural Environment Research Council (NERC) have been holding a number of events and last week (25 – 29 Oct) Into the Blue (a science showcase) was held at the Runway Visitor Centre underneath the wings of a Concorde. Along with a fellow PhD student from Reading (Kieran Hunt, who helped out on a stand about the monsoon) I was privileged to help man a stand (on flash flooding).

The event was used to showcase all the science that NERC funds from the atmosphere through to ecology. There were 40 exhibits and the chance to take tours of Concorde and the FAAM aircraft.

Concorde (left) and FAAM aircraft (right)

Exhibits involved a variety of interactive activities from making clouds in a bottle, using Infra-red cameras, making rivers in sand boxes, meeting Boaty McBoatface and a virtual reality flash flood!

During the quieter moments at their stands the exhibitors were allowed to wander around the rest of the event (including getting tours on the planes). In doing this we were able to talk to a number of different scientists about their work and engage in all the activities.

Personal highlights for me were touring both the Concorde and the FAAM aircraft. Although the best bit was the interaction with the public and being able to give everyone (no matter the age, from kids to adults) a “wow moment”.

The stand I was helping run was called FlashFlood! This stall was run predominantly by the University of Hull on behalf of the Flooding From Intense Rainfall (FFIR) project. They had created a virtual reality flash flood that was based on a real event (Thinhope Burn, 17 July 2007) which enabled us to place the stand’s visitor into a river valley and take them through the process of flooding from intense rainfall and how floods can change the characteristics of the rivers. It also gave us the ability (because of the case we had chosen) to show people that just because its not raining heavily at your location does not mean you won’t get flooded.

img_20161029_0915221

Having virtual reality was a massive draw for people to come to our stand so we were always fairly busy, but the feedback we had was very positive with the most frequent comments being,

  • “It felt like I was really there”
  • “It really helps me to visualise the science”
  • “Wow, this is really amazing”.

Comments like this really make events such as Into the Blue worth while for us as scientists as we then realise we are getting our messages through to people, and it shows the usefulness of scientific research to the public.

Events like this can be exhausting, but they are definitely worth the effort as you get to see the delight of the public as they learn about different science and have fun at the same time.

A big thank you must be said to NERC and Manchester Runway Visitor Centre for organizing and hosting the event and to all the exhibitors who did a great job in communicating science to the public.

The effect of local topography on severe tropical convective rainfall development.

Email: m.f.f.b.mohdnor@pgr.reading.ac.uk

The occurrence of severe convective rainfall is common over the tropical rainforest region. While the basic mechanism of the development of severe convective rainfall over the tropics is well understood in previous studies, the effect of local topography may yield a unique development process.

One part of my PhD project is to look at how local topography modifies severe rainfall events over the western Peninsular Malaysia. This was examined via a case study of severe rainfall that took place on 2nd May 2012. On that day, heavy rainfall caused flash floods and landslides over Klang Valley (red box in Fig. 1). Although the total rainfall on the 2nd May was above the Apr-May average, it was not extremely high.

fig1_geography_malaysia

Fig. 1. The study area, specifically over the western Peninsular Malaysia. The red box is Klang valley area.

Looking at observational data was not enough to understand the processes involved in the development of severe rainfall event on 2nd May 2012 and therefore a simulation study was conducted using the UK Met Office Unified Model (1.5km horizontal resolution).

One theory which could explain  the rainfall event on 2nd May 2012 is the influence of a series of rainfall events that developed earlier. There were rainfall events over the Peninsular Malaysia and Sumatra Island in the early evening of 1st May 2012 along the Titiwangsa mountains (Peninsular Malaysia) and Barisan Mountains (Sumatra Island). These rainfall events influenced the development of rainfall over the Malacca Strait overnight. The rainfall event over the strait strengthened by the morning of 2nd May. In the afternoon of 2nd May, the western peninsula had the right atmospheric conditions to develop convective rainfall, and the rainfall over the strait influenced the intensification of rainfall over the western peninsula. Thus, we believe that the local topography has a large impact on the development of the 2nd May rainfall event.

So, how do we test the hypothesis? One way is to perform sensitivity experiments. Four sensitivity experiments were conducted, modifying the orography of both the peninsula and Sumatra, and removing Sumatra altogether (Fig. 2).

fig2_experiments_all

Fig. 2. Sensitivity experiments on the local orography and Sumatra Island. Control run on the first panel, flatPM (flat peninsula to sea level), flatSI (flat Sumatra), flatALL(both peninsula and Sumatra are flat), and noSI (Sumatra is removed)

The results show that orography influenced and modified the development of late evening rainfall over both landmasses on both days. On 2nd May, total rainfall in the experiments are as follows:
1. flatPM : Klang valley received less rainfall than control,
2. flatSI : Klang valley received less rainfall than control but more than flatPM,
3. flatALL : Klang valley received more rainfall than control, flatPM and flatSI experiments,
4. noSI : Klang valley received triple the amount of rainfall of the control and other experiments.
These results hint the complex relationship between local topography and rainfall. Moreover, both the peninsula and Sumatra are important for the development of the morning rainfall over the Malacca Strait, regardless of the orographic variability.

Whilst looking at one case study is not enough to draw a general conclusion, this will definitely be a step forward on broadening the information that we already have. A more robust conclusion would require further studies to be taken.

(This PhD project is supervised by Pete Inness and Christopher Holloway, and funded by MARA Malaysia).