Analysis of climate variability and droughts in East Africa using high-resolution climate data products

https://doi.org/10.1016/j.gloplacha.2020.103130Get rights and content

Highlights

  • High-resolution datasets help detect changes and variability in climate at local-scale.

  • Variability in rainfall during the short rain season is linked with Indian Ocean Dipole.

  • The 2000s is the warmest and driest decade compared to 1980s and 1990s.

  • Droughts are increasing in large parts of the region and it is linked with Nĩno3.4 index.

  • High-resolution drought maps allow the development of adaptation measures at a local scale.

Abstract

Analysis of climate variability and change as a basis for adaptation and mitigation strategies requires long-term observations. However, the limited availability of ground station data constrains studies focusing on detecting variability and changes in climate and drought monitoring, particularly in developing countries of East Africa. Here, we use high-resolution precipitation (1981–2016) and maximum and minimum temperature (T-max and T-min) (1979–2012) datasets from international databases like the Climate Hazard Group (CHG), representing the most accurate data sources for the region. We assessed seasonal, annual, and decadal variability in rainfall, T-max and T-min and drought conditions using the Standardized Precipitation Index (SPI). The impact of changes in Sea Surface Temperature on rainfall variability and droughts is assessed using the Nino3.4 and Indian Ocean Dipole (IOD) indices. The results show maximum variability in rainfall during October–December (OND, short rainy season) followed by March–May (MAM, long rainy season). Rainfall variability during OND showed a significant correlation with IOD in Ethiopia (69%), Kenya (80%), and Tanzania (63%). In Ethiopia, the period June–September (JJAS) showed a significant negative correlation (−56%) with the Nino3.4. Based on the 12-month SPI, the eastern and western parts of the region are getting drier and wetter, respectively with an average of mild, moderate, and severe droughts of more than 37%, 6%, and 2% of the study period, respectively. The observed severe droughts (e.g., 1999/2000) and extreme floods (e.g., 1997/1998) were found to be linked to respective negative and positive anomalies of the Nino3.4. In general, climate data products with high spatial resolution and accuracy help detect changes and variability in climate at local scale where adaptation is required.

Graphical abstract

In the 2000s, the eastern parts of the region are drier than the 1980s and 1990s. On the other hand, the western parts of Ethiopia and Kenya are wetter in the 2000s compare to the 1980s. Regarding temperature, however, the 2000s is the warmest decade compared to the 1980s and 1990s, which is in line with global warming.

Unlabelled Image
  1. Download : Download high-res image (453KB)
  2. Download : Download full-size image

Introduction

Countries in East Africa are particularly prone to climate variability (e.g., inter-annual variability in rainfall) and extreme climate events such as droughts and floods (WWF, 2006; Niang et al., 2014). The region is facing an increasing trend in maximum and minimum temperature and temperature extremes (e.g., increase in high and low percentiles of the temperature distribution) and high variability in seasonal rainfall and daily rainfall extremes (e.g., monthly maximum 1-day and 5-day precipitation) (Cattani et al., 2018; Gebrechorkos et al., 2018a). Due to the high variability in seasonal rainfall and occurrence of extreme events, the region is becoming one of the most food-insecure regions in the world looking for humanitarian assistance (ActionAid, 2016; Sidahmed, 2018). Moreover, as a result of high variability in rainfall, poor management of environmental resource, and absence of improved technologies, agricultural production is very low in Sub-Saharan Africa (IFPRI, 2009). Agriculture is the dominant sector in Africa and more than 80% of the population in East Africa depends on it, which also provides a significant contribution (up to 40%) to the economy of the region (FAO, 2014). In addition to agriculture, sectors such as water and energy and environmental resources (e.g., land) are increasingly affected by the changes and variability in climate (Niang et al., 2014). Agriculture in East Africa is largely rain-fed and it is based on the long (March–May) and short (October–December) rainy seasons, which makes the agriculture sector highly vulnerable to inter-annual rainfall variability. The change and shift in rainfall during March–May and October–December rainy seasons, which are the main cropping periods in the region, led to devastating droughts affecting the socio-economic welfare and environment (Haile et al., 2019). In East Africa, droughts are becoming a recurring event, every three years science 2005, and it has been difficult to manage the impacts due to limited forecasting skills and other anthropogenic and natural factors (Haile et al., 2019).

In addition to seasonal rainfall variability, one of the main challenges that make the region highly vulnerable to droughts and climate variability lies in the fact that the major portion of the agricultural land is owned by smallholder farmers. The land owned by the smallholder farmers provides about 90% of the total agricultural production (Salami et al., 2010) and they have less knowledge and capacity to adapt to any change in weather and climate, adding to the vulnerability of the agricultural sector (Kotir, 2011). Climate projections, in line to the observed change, show an increase in maximum and minimum temperature (Gebrechorkos et al., 2019a) and frequency of extreme events (e.g., droughts, floods, and heavy rainstorms) in East Africa (IPCC, 2007; Niang et al., 2014), which will pose a negative impact on the environment and the region's economy, health and wellbeing. According to IPCC (2007), by the end of 2100, the region's economy is projected to decline by 2%–7% as the result of the impact of climate change and variability on agriculture. Compared to the change in temperature, the change and variability in rainfall induce a significant impact on agriculture. In general, considering the vulnerability of the region to climate change and variability and the projected change in climate (increase in temperature and variability in precipitation), development of adaptation measures to reduce the possible impacts is urgently needed. In this region, there is already a growing interest in understanding the climate condition e.g., identifying the possible drivers of seasonal variability in rainfall at different spatial scales (watershed to regional scale) based on different datasets such as remote sensing-based rainfall products and climate model output (Wolff et al., 2011; Endris et al., 2013; Endris et al., 2015; Fer et al., 2017; Mpelasoka et al., 2018).

While most of the studies in this region focus on identifying the drivers, the number of studies dealing with the spatial and temporal variability providing high-resolution maps to identify hot spot areas which should be prioritized in adaptation plans are limited (e.g., Daron, 2014; Rowell et al., 2015; Seregina et al., 2014; Tierney et al., 2013). In addition, most of the earlier studies are confined to watersheds scale based on limited information from field-based meteorological stations and relay on regional averages based on coarse resolution of climate data such as output from Global Climate Models (GCMs). However, due to the limited, in terms of spatial and temporal coverage, availability of ground station data large parts of the region, particularly the remote parts are less studied. Moreover, application of coarse resolution of climate datasets from GCMs or RCMs (Regional Climate Models) can only provide average information on a global or regional scale, respectively. Therefore, for a better understanding of climate variability and for developing climate change adaptation measures at a local scale, climate information with high spatial resolution and temporal coverage are required.

Globally, in order to overcome the data challenges, a number of climate data products based on remote sensing (e.g., satellite-based rainfall products) such as Climate Hazards Group InfraRed Precipitation (CHIRP) and CHIRP with Station data (CHIRPS) (Funk et al., 2015) are developed for climate and hydrological studies. In addition, for Africa, a number of high-resolution satellite-based rainfall products such as the African Rainfall Climatology (Novella et al., 2013), Tropical Applications of Meteorology using Satellite and ground-based observations (TAMSAT) African Rainfall Climatology And Time series (TARCAT) (Maidment et al., 2017; Maidment et al., 2014; Tarnavsky et al., 2014), and Enhancing National Climate Services (ENACTS) (Dinku et al., 2014) are available at different spatial and temporal resolutions. The products differ in their development process (e.g., methodology, input data), temporal and spatial resolution, and spatial and temporal coverages. Therefore, before direct application of the products, assessing their accuracy by comparison with observed data is a prerequisite to identify the most accurate product and to produce valuable results. Hence, in our previous study (Gebrechorkos et al., 2018b), we evaluated different products based on climate models (RCMs), reanalysis, and satellite-based rainfall estimations such as CHIRPS and the African Rainfall Climatology. Finally, two products (for rainfall and maximum and minimum temperature) with high resolution (spatial and temporal) and coverage (spatial and temporal) were selected after a detailed evaluation over 21 regions of East Africa. In this study, therefore, using the selected datasets we assessed the variability in rainfall and maximum and minimum temperature (T-max and T-min) on decadal, seasonal, and annual time scales. In addition, the impact of large-scale climate variables such as the El Nĩno Southern Oscillation (ENSO) indices (e.g., Nĩno3.4) on rainfall variability and droughts are assessed. The results help identify areas with high variability in rainfall and an increase in droughts and develop adaptation measures at a local scale. In addition, the results will allow assessing the application of remote sensing and reanalysis based climate data products in climate and hydrological studies by comparing the results with historical observations (e.g., drought and flood events).

Section snippets

Study area

The study is conducted in East Africa, also called the Greater Horn of Africa, particularly in Ethiopia, Kenya, and Tanzania (Fig. 1). The region is characterized by diverse topography and climate (e.g., rainfall varies within tens of kilometres) (WWF, 2006). The region is heavily dependent on rainfall and agriculture is the main sector. The commonality in the region is the occurrence of droughts and floods which is affecting millions of people (Nicholson, 2016). In this region, when extreme

Methodology

The Climate data Operator (CDO) (Schulzweida et al., 2009) is used to merge and aggregate multiple NetCDF files and compute monthly and annual average (temperature) and sums (precipitation). CDO combines multiple command line operators to analyze and manipulate different climate datasets in different formats such as NetCDF and GRIB.

The monthly total rainfall and monthly average T-max and T-min were computed using CDO for the periods January–February (JF), March–May (MAM), June–September (JJAS),

Decadal variability in rainfall and temperature

For decadal analysis, the anomalies for each grid cell are computed as a departure from the mean of the period 1981–2010 (decadal mean) of rainfall, T-max, and T-min and the results are classified as the 1980s, 1990s, and 2000s (Fig. 2). The results show that in the 1980s large parts of Ethiopia, particularly western, northern and northeastern parts were drier than during the 1990s and 2000s, and the 2000s were drier than 1990s. On the other hand, the central part of Ethiopia (Arsi Zone) is

Discussion and conclusion

In data-sparse regions such as Africa, satellite-based climate data products with high spatial and temporal resolution and accuracy are widely used in hydro-climate studies (e.g. Camberlin et al., 2007; Rojas et al., 2011; Vrieling et al., 2016; Agutu et al., 2017; Cattani et al., 2018). However, some products, although bias-corrected at a regional and global scale, face large bias and disagreement when compared with ground stations (Kimani et al., 2018). Studies in East Africa (Cattani et al.,

Declaration of Competing Interest

None.

References (71)

  • P. Camberlin

    Temperature trends and variability in the Greater Horn of Africa: interactions with precipitation

    Clim. Dyn.

    (2017)
  • P. Camberlin et al.

    The East African March–May Rainy Season: associated atmospheric dynamics and predictability over the 1968–97 Period

    J. Clim.

    (2002)
  • D. Camuffo et al.

    Western Mediterranean precipitation over the last 300 years from instrumental observations

    Clim. Chang.

    (2013)
  • E. Cattani et al.

    East Africa rainfall trends and variability 1983–2015 using three long-term satellite products

    Remote Sens.

    (2018)
  • CDKN

    The IPCC’s Fifth Assessment Report : Whats in it for Africa

    (2014)
  • N.W. Chaney et al.

    Development of a high-resolution gridded daily meteorological dataset over Sub-Saharan Africa: spatial analysis of trends in climate extremes

    J. Clim.

    (2014)
  • J.D. Daron

    Regional climate Messages: East Africa

  • M.A. Degefu et al.

    Teleconnections between Ethiopian rainfall variability and global SSTs: observations and methods for model evaluation

    Meteorol. Atmospheric Phys.

    (2017)
  • T. Dinku et al.

    Bridging critical gaps in climate services and applications in Africa

    Earth Perspect.

    (2014)
  • H.S. Endris et al.

    Assessment of the performance of CORDEX regional climate models in simulating East African rainfall

    J. Clim.

    (2013)
  • H.S. Endris et al.

    Teleconnection responses in multi-GCM driven CORDEX RCMs over Eastern Africa

    Clim. Dyn.

    (2015)
  • FAO

    Adapting to Climate Change through Land and Water Management in Eastern Africa

    (2014)
  • I. Fer et al.

    The influence of El Niño–Southern Oscillation regimes on eastern African vegetation and its future implications under the RCP8.5 warming scenario

    Biogeosciences

    (2017)
  • C. Funk et al.

    Warming of the Indian Ocean threatens eastern and southern African food security but could be mitigated by agricultural development

    Proc. Natl. Acad. Sci.

    (2008)
  • C. Funk et al.

    The climate hazards infrared precipitation with stations—a new environmental record for monitoring extremes

    Sci. Data

    (2015)
  • S.H. Gebrechorkos et al.

    Changes in temperature and precipitation extremes in Ethiopia, Kenya, and Tanzania

    Int. J. Climatol.

    (2018)
  • S.H. Gebrechorkos et al.

    Evaluation of multiple climate data sources for managing environmental resources in East Africa

    Hydrol. Earth Syst. Sci.

    (2018)
  • S.H. Gebrechorkos et al.

    Regional climate projections for impact assessment studies in East Africa

    Environ. Res. Lett.

    (2019)
  • S.H. Gebrechorkos et al.

    Long-term trends in rainfall and temperature using high-resolution climate datasets in East Africa

    Sci. Rep.

    (2019)
  • G.G. Haile et al.

    Droughts in East Africa: Causes, impacts and resilience

    Earth-Sci. Rev.

    (2019)
  • M. Herrero et al.

    Climate Variability and Climate Change and their Impacts on Kenya’s Agricultural Sector (Report)

    (2010)
  • G.J. Huffman et al.

    The TRMM multisatellite precipitation analysis (TMPA): quasi-global, multiyear, combined-sensor precipitation estimates at fine scales

    J. Hydrometeorol.

    (2007)
  • IFPRI

    Economywide Impacts of Climate Change on Agriculture in Sub-Saharan Africa

    (2009)
  • IPCC
  • IPCC

    Climate Change 2013: The Physical Science Basis (eds Stocker et al.) (WG1)

    (2013)
  • Cited by (41)

    • Extreme climate events in sub-Saharan Africa: A call for improving agricultural technology transfer to enhance adaptive capacity

      2022, Climate Services
      Citation Excerpt :

      This is because key livelihoods depend on rainfall, as nearly 80 % of the agricultural land and crop production in this part of the world is rain-fed (Gérardeaux et al., 2018, Sarr et al., 2021). From 1970 up until recent years, nearly 14 % and 22 % of the East and West African populations respectively have been affected by extreme climate events such as windstorms, and multiple effects of drought including extreme temperatures (Diakite et al., 2020, Ekwezuo and Ezeh, 2020, Gebrechorkos et al., 2020). Other recent studies have established a warming trend through extreme heat events, with changes in patterns and amounts of rainfall (Ongoma et al., 2018, Rahimi et al., 2021, Marcotullio et al., 2021, Iyakaremye et al., 2022).

    • Food and nutrition security impacts of resilience capacity: Evidence from rural Ethiopia

      2022, Journal of Agriculture and Food Research
      Citation Excerpt :

      The empirical analysis also reveals the worsening of food insecurity by crippling covariate and idiosyncratic shocks in multiple channels. Concurrent and recurrent shocks cause such a wide range of adverse consequences as a decline in crop productivity, agricultural incomes, and per capita consumption, and deteriorate the welfare of rural households [57–59]. They also obliterate households' asset gains and eventually cause social instability asserting prior evidence [60,61].

    View all citing articles on Scopus
    View full text