Temperature and precipitation extremes in western Central Africa, Guinea and Zimbabwe: an international effort

by E. Aguilar1, A. Aziz Barry2, M. Brunet3, L. Ekang4, A. Fernandes5, M. Massoukina6, J. Mbah7, A. Mhanda8, D.J. do Nascimento9 , T.C. Peterson10, O.Thamba11, M. Tomou12, X. Zhang13

Introduction

Anthropogenically induced climate change is one of the major challenges to global society in the 21st century. The Nairobi Work Programme (NWP) of the United Nations Framework Convention on Climate Change (UNFCCC) highlights the critical need to improve knowledge about current and future climate change, especially in developing countries, in order to assist all parties to make informed and effective adaptation decisions (UNFCC, 2007). These changes, especially those observed in temperature and precipitation, not only affect the mean values of these climate elements but also produce changes in the frequency and intensity of extreme events. Understanding how extremes are changing globally, regionally and locally is a key issue for planning adaptation measures as changes in extremes have major impacts on our societies. The latest Intergovernmental Panel on Climate Change Assessment Report (IPCC, 2007) was not able to describe in depth changes in temperature and precipitation extremes over western Central Africa as adequate analysis and internationally exchanged long-term daily data set were not available.

To remedy this situation, the WMO Joint Expert Team on Climate Change Detection and Indices (ETCCDI)14 organized a regional climate change workshop in Brazzaville, Congo, 23-27 April 2007. The event followed a successful format evolved through 12 previous workshops (see Peterson and Manton, 2008) and was a combination of theoretical lectures on climate change issues, quality control, homogenization and climate change indices, as well as multiple hands-on-data sessions. The Brazzaville workshop was supported by the United Kingdom Met Office through the WMO Voluntary Cooperation Programme and coordinated by WMO’s World Climate Data and Monitoring Programme (WCDMP). Representatives from nine countries participated, including six from western Central Africa (Cameroon, Central African Republic, Democratic Republic of Congo, Gabon, Sao Tome e Principe and Congo), as well as Angola, Guinea and Zimbabwe (see Table 1 for more details). It was hosted by Alphonse Kanga, Direction de la Météorologie Nationale, Congo, and logistic and scientific support was provided by Omar Baddour and Hamma Kontongomde, WMO.

Table 1 — Participants in the ETCCDI workshop in Brazzaville, 23-27 April 2007

The Workshop represented the start of a cooperative effort that has resulted, for the first time, in a true analysis of temperature and precipitation extreme indices over western Central Africa, Guinea and Zimbabwe (see Aguilar et al., 2009)

Data and methods

As suggested in the introductory section, data scarcity is a key issue for the analysis of climate extremes in Africa. The preliminary results produced by the participants at the Workshop fostered an appreciation of the need for data rescue and digitization. The international data gathered at the Workshop was largely increased by information collected after the venue. For example, Cameroon’s digital daily data available for the workshop was from 1966 to 2005 and only ~80 per cent complete. A few months after the Workshop, the data were ~95% complete between 1951 and 2007. Another good example (see Figure 1) is the increased data availability in Bangui (Central African Republic) after the Workshop. These efforts made it possible to produce a daily temperature and precipitation dataset, including 66 stations (see Figure 2), which was completed with information from the Global Summary of the Day, a dataset available from the NOAA’S National Climatic Data Center.

Figure 1 — Daily maximum temperature time series for Berberati (Central African Republic), available after post-Workshop digitalization efforts by the Central African Republic’s Direction générale de l’aviation civile et de la météorologie (top) and data available at the Workshop (bottom)

The next step was quality-control of the dataset, focusing on the detection of non-systematic errors usually caused by data processing, most frequently during digitization. The statistical and visual procedures contained in the RClimDex package (available at the ETCCDI Website) were applied and complemented with other tests, following the guidelines given by Brunet et al., 2008.

Figure 2 — The location of the stations available during or after the Brazzaville Workshop. Regional series were calculated for three different regions: Guinea (dark blue and light blue circles), central (red and orange circles) and Zimbabwe (dark green and light green circles). Large circles in red, blue and dark green represent time series used for the final analysis; smaller circles in light blue, orange and light green represent time series removed after failing quality control or homogeneity assessment.

The quality-controlled dataset was checked for homogeneity using RhTest software, developed at the Meteorological Service of Canada, and also available from the ETCCDI Website. This program allowed us to identify inhomogeneities in the time series, using a two-phase regression model with a linear trend for the entire series (Wang, 2003; 2008(a); 2008(b). This, together with the shortness of some series or their lack of continuity, led to the removal of some stations from the final analysis (see Figure 2), as homogenization of daily data is still a challenging problem, that will be hopefully solved in the near future15.

A set of 27 indices formulated and coordinated by ETCCDI were calculated using again the RClimdex software, although some of them were suppressed as they were not relevant to the studied region (i.e. number of days with temperature below 0°C). The indices (fully described at the Expert Team’s Website) are mostly based on station-level thresholds calculated over a base period, such as the 10th percentile of maximum temperature or the 95th percentile of precipitation. These thresholds are determined for each day of the year, using data from that day and two days on either side of it over the course of the base period.

The time series were split into three geographical groups to calculate regional time series: Guinea, Zimbabwe and central (containing data points from the other represented countries). Trends were evaluated for annual and seasonal values using the adaptation of the Sen slope indicator (Sen, 1968) described by Zhang et al., 2000.

Results

Trends for the temperature indices (given in Table 2, together with indicators of global trends) show significant warming produced by the increases in warm extremes and decreases in cold extremes. For example, the warmest day and night of the year are warming at a rate approximately comparable to the global average. The coldest day and night of the year are warming slower than the global average, although the planetary trend for the coldest day is insignificant.

Table 2 — Regional trends in temperature indices. The trends for the globe are from Alexander et al. (2006), based on the period 1955-2003. A trend significant at the 5 per cent level is given in bold.

The diurnal temperature range—which is increasing globally—does not show the same behaviour in the studied regions and the percentile based indices (number of days/nights that are above or below the 90th or 10th percentile) show as well warming slopes, with the central region warming faster than average. Figure 3 shows, as an example, the regional time series of annual counts of warm days. The identified warming matches well with the results by New et al. (2006) in their study for southern Africa, including Zimbabwe.

Figure 3 — Regional time series for number of warm days. The three panels represent regionally averaged time series of anomalies in °C to the 1971-1995 reference period for Central region, Zimbabwe and Guinea.

In relation to precipitation indices (shown in Table 3, together with indicators of global trends), the most striking feature is the decrease in total precipitation amount (see Figure 4) and in the indicators of heavy precipitation both in Guinea and in the central region, contrasting with the global increases. Zimbabwe does not show any significant trends in precipitation. This is in agreement with Nicholson (2000, 2001), who highlights in her study of monthly African precipitation data a shift from relatively wet conditions from the 1920s to the early 1950s to dry conditions from the 1970s onwards, especially in the Gulf of Guinea. Results are described and discussed more in depth in Aguilar et al., 20090.

Table 3 — Regional and global trends in precipitation indices. The global trends are from Alexander et al. (2006), based on the period 1955- 2003. A trend significant at the 5 per cent level is given in bold.

Figure 4 — Regional time series for total precipitation. The three panels represent regionally averaged time series of anomalies in mm to the 1971-1995 reference period for CENTRAL region, Zimbabwe and Guinea.

Conclusions

We identified through our study a clear warming pattern in temperature extremes and less change in precipitation extremes over western central Africa, Guinea and Zimbabwe. This analysis highlights the benefits of international cooperation and regional climate change workshops. The participating countries belong to three WMO subregions: Central Africa, West Africa and southern Africa. WMO, through the World Climate Data and Monitoring Programme, helps these countries, in particular the least developed ones, to acquire the capacity of implementing this plan.

The calculated indices are available at the ETCCDI Website, thanks to the generosity of the National Meteorological and Hydrological Services which participated in the Workshop. In a region with limited international data exchange, this is an important step forward and opens up possibilities for many additional lines of research related to measuring observed climate change, such as links between climate change and variability food security or alterations in ecosystems.

References

Aguilar, E., A. Barry, M. Brunet, L. Ekang, A. Fernandes, M. Massoukina, J. Mbah, A. Mhanda, D. J. do Nascimento, T. C. Peterson, O. Thamba, M. Tomou, X. Zhang, 2009: Changes in temperature and precipitation extremes in western Central Africa, Guinea Conakry and Zimbabwe, 1955-2006, Journal of Geophysical Research-Atmospheres (in press).

Alexander, L. V. et al., 2006: Global observed changes in daily climate extremes of temperature and precipitation, Journal of Geophysical Research-Atmospheres, 111.

Barnett, T. P. et al., 2008: Human-induced changes in the hydrology of the western United States, Science, 319, 1080-1083.

Brunet, M., O. Saladié, P. Jones, E. Aguilar, A. Moberg, D. Lister, A. Walther, and C. Almarza, 2008: A case-study/guidance on the development of long-term daily adjusted temperature datasets, WCDMP-66, 46 pp, WMO, Geneva.

IPCC,2007: Climate Change 2007. The Physical Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Pannel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and NY, USA.

Peterson, T. C. and Manton, M.J., 2008: Monitoring changes in climate extremes: A tale of international collaboration. Bulletin of the American Meteorological Society (in press).

New, M. et al., 2006: Evidence of trends in daily climate extremes over southern and west Africa, Journal of Geophysical Research-Atmospheres, 111.

Nicholson, S. E., 2000: The nature of rainfall variability over Africa on time scales of decades to millenia, Global Planet. Change, 26, 137-158.

Nicholson, S.E., 2001: Climatic and environmental change in Africa during the last two centuries, Climate Research, 17, 123-144.

Sen, P. K., 1968: Estimates of regression coefficient based on Kendalls Tau, Journal of the American Statistical Association, 63, 1379-end.

Wang, X. L., 2003: Comments on “Detection of undocumented changepoints: a revision of the two-phase regression model”, J. Clim., 16, 3383-3385.

Wang, X. L., 2008(a): Accounting for autocorrelation in detecting mean-shifts in climate data series using the penalized maximal t or F test, J. Appl. Meteorol. Climatol., 47-9, 2423-2444

Wang, X. L., 2008(b): Penalized maximal F test for detecting undocumented mean shift without trend change, J. Atmos. Ocean. Technol., 25, 368-384.

1. Centre on Climate Change (C3), Geography Department, University Rovira i Virgili, Tarragona, Spain
2. Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich, United Kingdom
3. Centre on Climate Change, University Rovira i Virgili, Department of Geography, Tarragona, Spain
4. Cabinet du Ministre des Transports, Gabon
5. Institut National de Météorologie, Sao Tomé et Principe
6. Direction de la Météorologie Nationale, Congo
7. Direction de la Météorologie Nationale, Cameroon
8. Zimbabwe Meteorological Services Department
9. Instituto Nacional de Hidrometeorología e Geofísica, Angola
10. NOAA’s National Climatic Data Center, Asheville, North Carolina, USA
11. Agence Nationale de Météorologie et de Télédétection par Satellite, METTELSAT
12. Direction générale de l'aviation civile et de la météorologie, République Centrafricaine
13. Climate Research Branch, Meteorological Service of Canada, Downsview, Ontario, Canada
14. The expert team is jointly sponsored by the WMO) Commission for Climatology (CCl), the World Climate Research Programme (WCRP) project on Climate Variability and Predictability (CLIVAR) and, since 2006, the Joint WMO-Intergovernmental Oceanographic Commission (IOC) of the United National Educational, Scientific and Cultural Organization Technical Commission for Oceanography and Marine Meteorology (JCOMM).
15. Several groups are working on new methods for homogenization of climatological elements on a daily basis. Some of these efforts are concentrated on the COST action ES0601: Advances in homogenization methods of climate series: an integrated approach (HOME). See www.homogenisation.org for more information.