Search

Influences of El Nino southern oscillation (ENSO) on weather systems have increased the frequency and amplitude of extreme events over the last century. This even continues to exacerbate the already warming earth, with 2014–2016 which coincided with the strongest El Nino years observed as the warmest period in recent past. This study presents an approach of characterizing droughts at various timescales and establishes teleconnections between ENSO and drought severity in Botswana. The study uses Standardized Precipitation Evaporation Index (SPEI) at timescales of 1, 3, 6, 12 and 24 month to characterize droughts and Pearson’s correlations to study the teleconnections between SPEIs and ENSO. Results from the study reveal that extreme droughts are a rare occurrence in Botswana though it is more prone to moderate droughts at 12 month SPEI with a probability of 19% in the north. The highest severe drought probability was 7% recorded in the east. Linear trends indicate increasing dryness of around 0.8% per decade. These results have demonstrated that warm sea surface temperatures combined with negative Southern Oscillation Index correspond to persistent negative SPEI values and thus are likely to result in dry conditions. Significant correlations were observed in the mid austral summer in December and January. Due to this relationship, the drought early warning systems could use ENSO as one of the instruments for predicting drought over the study area and hence in its management.

Semiarid areas exhibit high climate variability whose frequency has increased in the recent past. This variability ultimately influences the spatial and temporal characteristics of droughts which in turn require a reliable drought index that can adequately characterize drought both in time and space. This study applied two drought indices viz.: Standardized precipitation index (SPI) and Standardized precipitation and evapotranspiration index (SPEI) at timescales of 1-, 3-, 6-, 12-, 18- and 24-months for a period of 1960–2016. The monotonic changes in drought severity were studied using Mann-Kendall Z-statistic and Sen's slope estimator. Both drought indices were able to detect the historical drought events of 1961/62–1965/66, 1980/81–1986/87, 1991/92, 2001/02–2005/06, 2009/10–2012/13 and 2014/15–2015/16. These events mainly coincided with El Nino years, hence the influence of El Nino southern oscillation (ENSO) on drought occurrence could not be ruled out. The drought evolutions were characterized by low frequency and longer durations at timescales of 12-, 18-, and 24-months. It was also observed that SPI overestimated drought severity during dry winter months while SPEI showed higher spatial coverage of drought vulnerability compared to SPI. Spatial and temporal trends of droughts showed significant wetting trends during the period 1960–1979 while during the entire period it was mostly drying tendencies after 1979/80. Pandamatenga and Shakawe showed higher vulnerability towards drying conditions during the period of analysis. The association between SPI and SPEI was also investigated using Spearman's rank correlation. Results from this analysis indicate that SPI accounts for >50% of the variations in SPEI and the association was strongest (96%) at 12-months timescale. The influence of temperature on drought evolutions was demonstrated at Francistown with SPI recording positive trends and SPEI negative ones. This study concludes that with the ongoing global warming, SPEI proves more robust in characterizing droughts in semiarid areas. It is hoped that findings from this study will augment ongoing efforts of mitigating negative impacts of increased incidences of climate variability especially in areas of agriculture and water resource management.

Uncertainties in rainfall have increased in the recent past exacerbating climate risks which are projected to be higher in semiarid environments. This study investigates the associated features of rainfall such as rain onset, cessation, length of the rain season (LRS), and dry spell frequency (DSF) as part of climate risk management in Botswana. Their trends were analysed using Mann-Kendall test statistic and Sen’s Slope estimator. The rainfall-evapotranspiration relationships were used in formulating the rain onset and cessation criteria. To understand some of the complexities arising from such uncertainties, artificial neural network (ANN) is used to predict onset and cessation of rain. Results reveal higher coefficients of variation in onset dates as compared to cessation of rain. Pandamatenga experiences the earliest onset on 28th of November while Tsabong the latest on 14th of January. Likewise, earliest cessation is observed at Tshane on 22nd of February and the latest on 30th of March at Shakawe. The shortest LRS of 45 days is registered at Tsabong whereas the northern locations show LRS greater than 100 days. Stations across the country experience strong negative correlation between onset and LRS of − 0.9. DSF shows increasing trends in 50% of the stations but only significant at Mahalapye, Pandamatenga, and Shakawe. Combining the LRS criteria and DSF, Kasane, Pandamatenga, and Shakawe were identified to be suitable for rainfed agriculture in Botswana especially for short to medium maturing cereal varieties. Predictions of onset and cessation indicate the possibility of delayed onset by 2–5 weeks in the next 5 years. Information generated from this study could help Botswana in climate risk management in the context of rainfed farming.

Climate variability and change is expected to affect agricultural productivity among other sectors. Studying the influence of this variability on crop production is one measure of generating climate change resilience strategies. In this study, the influence of climate variability on crop yield is investigated by determining the degree of association between climatic indices and crop yields of maize and sorghum using spearman’s rank correlation. The climatic indices used in this study are aridity index (AI), standardised precipitation evapotranspiration index (SPEI) at timescales of 1, 3, 6 and 12 months and southern oscillation index (SOI) representing El Nino southern oscillation (ENSO) influence on local climate. Local rainfall characteristics are expressed through length of the rainy season (LRS). Results reveal that ENSO influence is the most dominant across Botswana accounting for 85% and 78% variations in maize and sorghum yields respectively. Whereas AI and SPEI accounts for 70% and 65% variations in maize and sorghum respectively, LRS accounts for only 50% and 62% respectively. To facilitate agricultural planning, crop yield projections have been made using artificial neural network (ANN) models. The ANN projections indicate a likelihood of maize and sorghum yields declining by 51% and 70% respectively in the next 5 years. The high association between ENSO and crop yields in Botswana could further facilitate yield projections. Information generated from this study is useful in agricultural planning and hence strengthens farmers’ strategies in mitigating impacts of climate variability and change in semiarid areas.

In semi-arid catchments, hydrological modeling, water resources planning and management are hampered by insufficient spatial rainfall data which is usually derived from limited rain gauge networks. Satellite products are potential candidates to augment the limited spatial rainfall data in these areas. In this paper, the utility of the Tropical Rainfall Measuring Mission (TRMM) product (3B42 v7) is evaluated using data from the Notwane catchment in Botswana. In addition, rainfall simulations obtained from a multi-site stochastic rainfall model based on the generalised linear models (GLMs) were used as additional spatial rainfall estimates. These rainfall products were compared to the observed rainfall data obtained from six (6) rainfall stations available in the catchment for the period 1998–2012. The results show that in general the two approaches produce reasonable spatial rainfall estimates. However, the TRMM products provided better spatial rainfall estimates compared to the GLM rainfall outputs on an average, as more than 90% of the monthly rainfall variations were explained by the TRMM compared to 80% from the GLMs. However, there is still uncertainty associated mainly with limited rainfall stations, and the inability of the two products to capture unusually high rainfall values in the data sets. Despite this observation, rainfall indices computed to further assess the daily rainfall products (i.e. rainfall occurrence and amounts, length of dry spells) were adequately represented by the TRMM data compared to the GLMs. Performance from the GLMs is expected to improve with addition of further rainfall predictors. A combination of these rainfall products allows for reasonable spatial rainfall estimates and temporal (short term future) rainfall simulations from the TRMM and GLMs, respectively. The results have significant implications on water resources planning and management in the catchment which has, for the past three years, been experiencing prolonged droughts as shown by the drying of Gaborone dam (currently at a record low of 1.6% full), which is the main source of water supply to the city of Gaborone and neighbouring townships in Botswana.

This study is conducted over the data-poor Limpopo basin centered over southern Africa using reanalysis downscaled to useful resolution. Reanalysis products are of limited value in hydrological applications due to the coarse spatial scales they are available at. Dynamical downscaling of these products over a domain of interest offers a means to convert them to finer spatial scales in a dynamically consistent manner. Additionally, this downscaling also offers a way to resolve dominantatmospheric processes, leading to improved accuracy in the atmospheric variables derived. This study thus evaluates high-resolution downscaling of an objectively chosen reanalysis (ERA-I) over the Limpopo basin using Weather Research and Forecasting (WRF) as a regional climate model. The model generally under-estimates temperature and over-estimates precipitation over the basin, although reasonably consistent with observations. The model does well in simulating observed sustained hydrological extremes as assessed using the Standardized Precipitation Index (SPI) although it consistently under-estimates the severity ofmoisture deficit for the wettest part of the year during the dry years. The basin's aridity index (I) is above the severe drought threshold during summer and is more severe in autumn. This practically restricts rain-fed agriculture to around 3 months in a year over the basin. This study presents possible beneficial use of the downscaled simulations foroptimal hydrologic design and water resources planning in data scarce parts of the world.

Lateral and lower boundary conditions derived from a suitable global reanalysis dataset forms the basis for deriving a dynamically consistent finer resolution downscaled product for climate and hydrological assessment studies. A problem with this, however, is that systematic biases have been noted to be present in the global reanalysis datasets that form these boundaries, biases which can be carried into the downscaled simulations thereby reducing their accuracy or efficacy. In this work, three Weather Research and Forecasting (WRF) model downscaling experiments are undertaken to investigate the impact of bias correcting European Centre for Medium range Weather Forecasting Reanalysis ERA-Interim (ERA-I) atmospheric temperature and relative humidity using Atmospheric Infrared Sounder (AIRS) satellite data. The downscaling is performed over a domain centered over southern Africa between the years 2003 and 2012. The sample mean and the mean as well as standard deviation at each grid cell for each variable are used for bias correction. The resultant WRF simulations of near-surface temperature and precipitation are evaluated seasonally and annually against global gridded observational datasets and compared with ERA-I reanalysis driving field. The study reveals inconsistencies between the impact of the bias correction prior to downscaling and the resultant model simulations after downscaling. Mean and standard deviation bias corrected WRF simulations are, however, found to be marginally better than mean only bias corrected WRF simulations and raw ERA-I reanalysis driven WRF simulations. Performances, however, differ when assessing different attributes in the downscaled field. This raises questions about the efficacy of the correction procedures adopted.

Multiple reanalysis datasets currently exist that can provide boundary conditions for dynamic downscaling and simulating local hydro-climatic processes at finer spatial and temporal resolutions. Previous work has suggested that there are two reanalyses alternatives that provide the best lateral boundary conditions for downscaling over southern Africa. This study dynamically downscales these reanalyses (ERA-I and MERRA) over southern Africa to a high resolution (10 km) grid using the WRF model. Simulations cover the period 1981–2010. Multiple observation datasets were used for both surface temperature and precipitation to account for observational uncertainty when assessing results. Generally, temperature is simulated quite well, except over the Namibian coastal plain where the simulations show anomalous warm temperature related to the failure to propagate the influence of the cold Benguela current inland. Precipitation tends to be overestimated in high altitude areas, and most of southern Mozambique. This could be attributed to challenges in handling complex topography and capturing large-scale circulation patterns. While MERRA driven WRF exhibits slightly less bias in temperature especially for La Nina years, ERA-I driven simulations are on average superior in terms of RMSE. When considering multiple variables and metrics, ERA-I is found to produce the best simulation of the climate over the domain. The influence of the regional model appears to be large enough to overcome the small difference in relative errors present in the lateral boundary conditions derived from these two reanalyses.

Arid and semi-arid environments have been identified with locations prone to impacts of climate variability and change. Investigating long-term trends is one way of tracing climate change impacts. This study investigates variability through annual and seasonal meteorological time series. Possible inhomogeneities and years of intervention are analysed using four absolute homogeneity tests. Trends in the climatic variables were determined using Mann–Kendall and Sen’s Slope estimator statistics. Association of El Nino Southern Oscillation (ENSO) with local climate is also investigated through multivariate analysis. Results from the study show that rainfall time series are fully homogeneous with 78.6 and 50% of the stations for maximum and minimum temperature, respectively, showing homogeneity. Trends also indicate a general decrease of 5.8, 7.4 and 18.1% in annual, summer and winter rainfall, respectively. Warming trends are observed in annual and winter temperature at 0.3 and 1.5% for maximum temperature and 1.7 and 6.5% for minimum temperature, respectively. Rainfall reported a positive correlation with Southern Oscillation Index (SOI) and at the same time negative association with Sea Surface Temperatures (SSTs). Strong relationships between SSTs and maximum temperature are observed during the El Nino and La Nina years. These study findings could facilitate planning and management of agricultural and water resources in Botswana.

Regional climate modelling studies often begin by downscaling a reanalysis dataset in order to simulate the observed climate, allowing the investigation of regional climate processes and quantification of the errors associated with the regional model. To date choice of reanalysis to perform such downscaling has been made based either on convenience or on performance of the reanalyses within the regional domain for relevant variables such as near-surface air temperature and precipitation. However, the only information passed from the reanalysis to the regional model are the atmospheric temperature, moisture and winds at the location of the boundaries of the regional domain. Here we present a methodology to evaluate reanalyses derived lateral boundary conditions for an example domain over southern Africa using satellite data. This study focusses on atmospheric temperature and moisture which are easily available. Five commonly used global reanalyses (NCEP1, NCEP2, ERA-I, 20CRv2, and MERRA) are evaluated against the Atmospheric Infrared Sounder satellite temperature and relative humidity over boundaries of two domains centred on southern Africa for the years 2003–2012 inclusive. The study reveals that MERRA is the most suitable for climate mean with NCEP1 the next most suitable. For climate variability, ERA-I is the best followed by MERRA. Overall, MERRA is preferred for generating lateral boundary conditions for this domain, followed by ERA-I. While a “better” LBC specification is not the sole precursor to an improved downscaling outcome, any reduction in uncertainty associated with the specification of LBCs is a step in the right direction.