Transition from the Southern to the Northern Hemisphere summer of zones of active convection over the Congo Basin

Summary This study used monthly rainfall totals for the period 1961 to 1988 and pentad OLR values for the period 1974 to 1991 to study the structure and transition of active convention across the Congo Basin (10°S–5°N, 15°E–35°E) from the southern to the northern hemisphere summer. This involved the examination of map patterns and cross-sections of monthly rainfall and pentad OLR data.The results from the study indicated that there were two seasons observed over the Congo Basin; one is the wet season lasting from September to April and the other a dry season covering the rest of the year. The onset of the wet season takes place rapidly with active convection spreading very quickly to the south near latitude 20°S. This is due to the formation of the meridional (north–south) branch of the ITCZ over this region.This study has confirmed that the annual rainfall over the Congo Basin is reliable with the coefficient of variation of less than 30%. The wet seasons (e.g., SON and DJF) also show reliable rainfall occurrence but the dry season (e.g., JJA) has low reliability.The anomalously wet seasons are characterised by a relatively slow transition rate (1° latitude per pentad) of zones of active convection resulting in a late withdrawal of the rainy season while the dry seasons show a rapid transition rate with an early withdrawal of zones of active convection.High-rainfall months (>200mm) are concentrated within the Southern Hemisphere summer months. These high-rainfall months progress from the equator to the southern latitude following generally the movement of the overhead sun.The results further revealed that the years 1987/1984 had the lowest/highest mean OLR values over the Congo Basin within the period 1979 to 1991. The rates of transition of the zones of low OLR values were 0.9/5.0 degrees of latitude during 1987/1984, respectively.Received June 18, 2002; revised September 30, 2002; accepted November 21, 2002 Published online: June 12, 2003     

Recent summer precipitation trends in the Greater Horn of Africa and the emerging role of Indian Ocean sea surface temperature

We utilize a variety of climate datasets to examine impacts of two mechanisms on precipitation in the Greater Horn of Africa (GHA) during northern-hemisphere summer. First, surface-pressure gradients draw moist air toward the GHA from the tropical Atlantic Ocean and Congo Basin. Variability of the strength of these gradients strongly influences GHA precipitation totals and accounts for important phenomena such as the 1960s–1980s rainfall decline and devastating 1984 drought. Following the 1980s, precipitation variability became increasingly influenced by the southern tropical Indian Ocean (STIO) region. Within this region, increases in sea-surface temperature, evaporation, and precipitation are linked with increased exports of dry mid-tropospheric air from the STIO region toward the GHA. Convergence of dry air above the GHA reduces local convection and precipitation. It also produces a clockwise circulation response near the ground that reduces moisture transports from the Congo Basin. Because precipitation originating in the Congo Basin has a unique isotopic signature, records of moisture transports from the Congo Basin may be preserved in the isotopic composition of annual tree rings in the Ethiopian Highlands. A negative trend in tree-ring oxygen-18 during the past half century suggests a decline in the proportion of precipitation originating from the Congo Basin. This trend may not be part of a natural cycle that will soon rebound because climate models characterize Indian Ocean warming as a principal signature of greenhouse-gas induced climate change. We therefore expect surface warming in the STIO region to continue to negatively impact GHA precipitation during northern-hemisphere summer.     

The resilience of annual vegetation primary production subjected to different climate change scenarios

We examined if climate change in two dry ecosystems—Mediterranean (DME) and Semiarid (SAE)—would cause substantial reduction in the production of annual vegetation. Field measurements and computer simulations were used to examine the following six climate change scenarios: (1) rainfall amount reduction; (2) increases of 10 % in annual evaporation rate and 5 % in annual temperature; (3) increase in magnitude of rainfall events, accompanied by reductions in frequency and seasonal variation; (4) postponement of the beginning of the first rainfall event of the growing season; (5) long dry spells during the growing season; and (6) early ending of the growing season. The results revealed the following outcomes. a) Reduction by 5–35 % in annual rainfall amount did not significantly affect productivity in the DME, but a large (25–35 %) decrease in rainfall would change vegetation productivity in the SAE and lead to a patchier environment. b) Similar results were observed: when temperature and evaporation rate were increased; when the magnitude of rainfall events increased but their frequency decreased; and during a long mid-season dry spell. c) In both ecosystems, changes in the temporal distribution of rainfall, especially at the beginning of the season, caused the largest reduction in productivity, accompanied by increased patchiness. d) Long-term data gathered during the last three decades indicated that both environments exhibited high resilience of productivity under rainfall variability. These results imply that the response of dry ecosystems to climate change is not characterized by a dramatic decrease in productivity. Moreover, these ecosystems are more resilient than expected, and their herbaceous productivity might undergo drastic changes only under more severe scenarios than those currently predicted in the literature.     

Ecological Implications of Changes in Drought Patterns: Shifts in Forest Composition in Panama

CO₂ concentration is increasing, temperature is likely to rise, and precipitation patterns might change. Of these potential climatic shifts, it is precipitation that will have the most impact on tropical forests, and seasonal patterns of rainfall and drought will probably be more important than the total quantity of precipitation. Many tree species are limited in distribution by their inability to survive drought. In a 50 ha forest plot at Barro Colorado Island in Panama (BCI), nearly all tree and shrub species associated with moist microhabitats are declining in abundance due to a decline in rainfall and lengthening dry seasons. This information forms the basis for a simple, general prediction: drying trends can rapidly remove drought-sensitive species from a forest. If the drying trend continues at BCI, the invasion of drought-tolerant species would be anticipated, but computer models predict that it could take 500 or more years for tree species to invade and become established. Predicting climate-induced changes in tropical forest also requires geographic information on tree distribution relative to precipitation patterns. In central Panama, species with the most restricted ranges are those from areas with a short dry season (10–14 weeks): 26–39% of the tree species in these wet regions do not occur where it is drier. In comparison, just 11–19% of species from the drier side of Panama (18 week dry season) are restricted to the dry region. From this information, I predict that a four-week extension of the dry season could eliminate 25% of the species locally; a nine-week extension in very wet regions could cause 40% extinction. Since drier forests are more deciduous than wetter forests, satellite images that monitor deciduousness might provide a way to assess long-term forest changes caused by changes in drought patterns. I predict that increasing rainfall and shorter dry seasons would not cause major extinction in tropical forest, but that drying trends are a much greater concern. Longer dry seasons may cause considerable local extinction of tree species and rapid forest change, and they will also tend to exacerbate direct human damage, which tends to favor drought-adapted and invasive tree species in favor of moisture-demanding ones.     

Changes in Fire and Climate in the Eastern Iberian Peninsula (Mediterranean Basin)

Fire is a dominant ecological factor in Mediterranean ecosystems, and changes in the fire regime can have important consequences for the stability of our landscapes. In this framework I asked firstly, what is the trend in fire number and area burned in the eastern Iberian Peninsula, and then, to what extent is the inter-annual variability of fires determined by climatic factors. To answer these questions I analysed the meteorological data (temperature and precipitation) from 350 stations covering the eastern Iberian Peninsula (1950–2000), and the fire records for the same area (historical data, 1874–1968, and data from recent decades, 1968–2000). The results suggested a slight tendency towards decreasing summer rainfall and a clear pattern of increasing annual and summer temperatures (on average, annual temperatures increased 0.35 °C per decade from 1950 to 2000). The analysis of fire records suggested a clear increase in the annual number of fires and area burned during the last century; however, in the last three decades the number of fires also increased but the area burned did not show a clear trend. For this period the inter-annual variability in area burned was significantly related to the summer rainfall, that is, in wet summers the area burned was lower that in dry summers. Furthermore, summer rainfall was significantly cross-correlated with summer area burned for a time-lag of 2 years, suggesting that high rainfall may increase fuel loads that burn 2 years later.     

Impact of Climate Change on Agricultural Production in the Sahel – Part 2. Case Study for Groundnut and Cowpea in Niger

During the last 30 years, the climate of the West African Sahel has undergone various changes, especially in terms of rainfall. This has large consequences for the poor-resource farmers depending mainly on rainfed agriculture. This paper investigates the impacts of current climate variability and future climate change on groundnut and cowpea production in Niger for three major agricultural regions, including the groundnut basin.Niger was one of the largest West African groundnut producing and exporting countries. Groundnut production – as a cash crop – dropped fromabout 312,000 tons in the mid 1960s (about 68% exported) to as low as 13,000tons in 1988 and increased again to 110,000 tons in 2000. Cowpea, a food crop, showsa different tendency, going from 4,000 tons in the mid fifties to a maximum of 775,000 tons in 1997, and its cultivated area is still increasing. It is also a cash crop in local economies (especially for women).To highlight the impact of climate change on groundnut and cowpea production (significantly determined by rainfall in July, August and September), the following components of the rainfall regime were calculated for the period 1951–1998: mean annual and monthly rainfall, beginning, end and lengthof the rainy season, number of rainy days per month, amount of rainfall per rainy day and the maximum length of dry spell per month. Three sub-periods whose duration varied per region were defined: for Dosso 1951–1968,1969–1984 and 1985–1998; for Maradi 1951–1970, 1971–1987 and1988–1998; and for Zinder 1951–1966, 1967–1984 and 1985–1998. A change in rainfallregime components was observed between the three sub-periods, which were characterized in chronological order by wet, dry and intermediate conditions. To assess the impact of climate variability and change on groundnut and cowpea production, a statistical modeling approach has been followed, based on thirteen predictors as described and discussed in the preceding paper. Climate change is mimicked in terms of reduced total amount of rainfall for the three main rainfall months and an increased temperature, while maintaining other significant predictors at a constant level. In 2025,production of groundnut is estimated to be between 11 and 25% lower, while cowpeayield will fall maximally 30%. Various strategies to compensate thispotential loss are presented for the two crops.     

Long-term variations of annual and growing season rainfalls in Nigeria

Summary Evidence for changes in the annual and growing season rainfall series for the period 1919 to 1985 in Nigeria are examined on a regional basis, using power-spectral and lowpass filter techniques, and the Mann-Kendall rank statistic. Four regions, the Coastal Zone, the Guinea-Savanna Zone, the Midland area and the Sahel, are used in the investigation of rainfall variation from south to north across the country.Quasi-periodic oscillations in the annual and growing season rainfall series are found to be concentrated in four spectral bands: 2.0–2.4, 2.7–2.9, 3.2–3.6 and 5.6–6.3 years. The spatial coherence of the fluctuations in annual and growing season rainfall is found to be limited to Nigeria south of 11 degrees north latitude. Evidence also emerges of a progressive decline in annual and growing season rainfall for northern Nigeria, north of nine degrees north latitude, for the period 1939–1985.With 5 Figures     

Equatorial African climate teleconnections

Teleconnections between equatorial African climate and the surrounding circulation are examined using a convective index over the Congo River Basin in March to May (MAM) and July to September (JAS) seasons. Its influence on the wider region is determined through lag correlation and cross-wavelet analysis. During seasons of deeper convection, easterly winds weaken over the tropical Atlantic (anomalous flow toward Africa), whilst upper westerly winds weaken over southern Africa (in JAS). We view this as zonal overturning with ascent over the equatorial African lowlands and Congo River Basin that spreads moisture to the North African Sahel, with influence from the Pacific El Niño. Another facet of our study is the relationship between East African highlands rainfall and the Indian Ocean circulation. We find coupling between the Indian Ocean Rossby wave, a thermocline oscillation and Walker cell over the Indian Ocean that induces shifts in rainfall, particularly in the October to December season.     

Climate change and implications for agriculture in Niger

Five-year moving averages of annual rainfall for 21 locations in Niger showed a decline in the annual rainfall after 1960. Correlation coefficients of the moving averages of monthly rainfall with annual rainfall showed significant correlations between the decline in the annual rainfall with decreased rainfall in August. Analysis of daily rainfall data for rainy season parameters of interest to agriculture suggested that from 1965 there was a significant decrease in the amount of rainfall and in the number of rainy days in the months of July and August, resulting in a decreased volume of rainfall for each rainstorm. In comparison to the period 1945–64, major shifts have occurred in the average dates of onset and ending of rains during 1965–88. The length of the growing season was reduced by 5–20 days across different locations in Niger. The standard deviation for the onset and ending of the rains as well as the length of the growing season has increased, implying that cropping has become more risky. Water balance calculations also demonstrated that the probability of rainfall exceeding potential evapotranspiration decreased during the growing season. The implications of these changes for agriculture in Niger are discussed using field data.     

Deep Soil Moisture Storage and Transpiration in Forests and Pastures of Seasonally-Dry Amazonia

To assess the impacts of land-use changes on plant-available water (PAW) and evapotranspiration (ET), volumetric water content (VWC) was measured to 8 m beneath three, adjacent ecosystems for four years (1991–1994). Estimates of PAW, ET, and deep drainage were generated for mature evergreen forest, adjacent pasture, and capoeira (second-growth forest on abandoned pasture land). PAW between 0 and 8 m depth for forest, pasture, and capoeira ranged from a low of 56, 400, and 138 mm at the end of the 1992 dry season to a high of 941, 1116, and 1021 mm during the 1994 wet season. We found significant differences in deep (4–8 m) stocks of PAW when comparing pasture with both forest types. In contrast, mature forest and capoeira PAW were not significantly different from one another at any depth during the experiment. In all three ecosystems available soil moisture from 4–8 m was depleted during the 1991 dry season by plant water uptake and was not recharged to 1991 levels until 1994 due to an intervening 2-year, El Niño Southern Oscillation event. Water balance estimates (based on measurements to 8 m) showed an average 10% decrease in ET from pasture compared to mature forest. Less than 15 years after pasture abandonment, ET in second-growth forest recovered to rates nearly equaling the mature forest rate. In seasonally dry environments annual and interannual cycles of deep soil moisture recharge and depletion influence rates of transpiration and drainage. These deep cycles are not currently incorporated in models of regional and global moisture flux.     

Changes in rainfall regime over Burkina Faso under the climate change conditions simulated by 5 regional climate models

Sahelian rainfall has recorded a high variability during the last century with a significant decrease (more than 20 %) in the annual rainfall amount since 1970. Using a linear regression model, the fluctuations of the annual rainfall from the observations over Burkina Faso during 1961–2009 period are described through the changes in the characteristics of the rainy season. The methodology is then applied to simulated rainfall data produced by five regional climate models under A1B scenario over two periods: 1971–2000 as reference period and 2021–2050 as projection period. As found with other climate models, the projected change in annual rainfall for West Africa is very uncertain. However, the present study shows that some features of the impact of climate change on rainfall regime in the region are robust. The number of the low rainfall events (0.1–5 mm/d) is projected to decrease by 3 % and the number of strong rainfall events (>50 mm/d) is expected to increase by 15 % on average. In addition, the rainy season onset is projected by all models to be delayed by one week on average and a consensus exists on the lengthening of the dry spells at about 20 %. Furthermore, the simulated relationship between changed annual rainfall amounts and the number of rain days or their intensity varies strongly from one model to another and some changes do not correspond to what is observed for the rainfall variability over the last 50 years.     

Food security in the face of climate change: Adaptive capacity of small-scale social-ecological systems to environmental variability

Improving the adaptive capacity of small-scale irrigation systems to the impacts of climate change is crucial for food security in Asia. This study analyzes the capacity of small-scale irrigation systems dependent on the Asian monsoon to adapt to variability in river discharge caused by climate change. Our study is motivated by the Pumpa irrigation system, a small-scale irrigation system located in Nepal that is a model for this type of system. We developed an agent-based model in which we simulated the decisions farmers make about the irrigation strategy to use according to available water flow. Given the uncertainty associated with how climate change may affect the Asian monsoon, we simulated the performance of the system under different projections of climate change in the region (increase and decrease in rainfall, reduction and expansion of the monsoon season, and changes in the timing of the onset of the monsoon). Accordingly to our simulations, farmers might need to adapt to rainfall intensification and a late onset in the monsoon season. The demands for collective action among farmers (e.g. infrastructure repair, meetings, decisions, etc.) might increase considerably due to climate change. Although our model suggests that investment in new infrastructure might increase the performance of the system under some climate change scenarios, the high inequality among farmers when water availability is reduced might hinder the efficiency of these measures due to a reduction of farmers’ willingness to cooperate. Our modeling exercise helps to hypothesize about the most sensitive climate change scenarios for smallscale irrigation farming in Nepal and helps to frame a discussion of some possible solutions and fundamental trade-offs in the process of adaptation to improve for food and water security under climate change.     

The influence of the land surface on the transition from dry to wet season in Amazonia

Summary Analysis of the fifteen years of European Centre for Medium Range Weather Forecasts (ECMWF) reanalysis suggests that the transition from dry to wet season in Southern Amazonia is initially driven by increases of surface latent heat flux. These fluxes rapidly reduce Convective Inhibition Energy (CINE) and increase Convective Available Potential Energy (CAPE), consequently providing favourable conditions for increased rainfall even before the large-scale circulation has changed. The increase of rainfall presumably initiates the reversal of the cross-equatorial flow, leading to large-scale net moisture convergence over Southern Amazonia. An analysis of early and late wet season onsets on an interannual scale shows that a longer dry season with lower rainfall reduces surface latent heat flux in the dry and earlier transition periods compared to that of a normal wet season onset. These conditions result in a higher CINE and a lower CAPE, causing a delay in the increase of local rainfall in the initiating phase of the transition and consequently in the wet season onset. Conversely, a wetter dry season leads to a higher surface latent heat flux and weaker CINE, providing a necessary condition for an earlier increase of local rainfall and an earlier wet season onset. Our results imply that if land use change in Amazonia reduces rainfall during dry and transition seasons, it could significantly delay the wet season onset and prolong the dry season.     

Drought analysis in the Tons River Basin,India during 1969-2008

The primary focus of this study is the analysis of droughts in the Tons River Basin during the period 1969–2008. Precipitation data observed at four gauging stations are used to identify drought over the study area. The event of drought is derived from the standardized precipitation index (SPI) on a 3-month scale. Our results indicated that severe drought occurred in the Allahabad, Rewa, and Satna stations in the years 1973 and 1979. The droughts in this region had occurred mainly due to erratic behavior in monsoons, especially due to long breaks between monsoons. During the drought years, the deficiency of the annual rainfall in the analysis of annual rainfall departure had varied from ?26% in 1976 to ?60% in 1973 at Allahabad station in the basin. The maximum deficiency of annual and seasonal rainfall recorded in the basin is 60%. The maximum seasonal rainfall departure observed in the basin is in the order of ?60% at Allahabad station in 1973, while maximum annual rainfall departure had been recorded as ?60% during 1979 at the Satna station. Extreme dry events (z score <?2) were detected during July, August, and September. Moreover, severe dry events were observed in August, September, and October. The drought conditions in the Tons River Basin are dominantly driven by total rainfall throughout the period between June and November.     

Characteristics of rainfall over Brazil: Annual variations and connections with the Southern Oscillation

Summary Annual and interannual variations of rainfall over Brazil are discussed. First, rainy and dry seasons for several stations of Brazil are determined using the data of 21 years (1958–1978). The progressive movement of the Intertropical Convergence Zone seems to be associated with the progresive variation of rainfall seasons in the equatorial eastern Brazil. The annual migration of deep tropical convection from Central and Southern Portion of the Amazon basin in austral summer to the northwestern sector of South America in austral winter seems to be responsible for the annual cycle of rainfall in the Amazon basin. The conncection between the interannual variation of rainfall over Brazil and the Southern Oscillation is also discussed. The correlation coefficient between the Southern Oscillation index and the rainfall is generally small over most of Brazil except over Rio Grande do Sul. The correlation between the spring rainfall of Rio Grande do Sul and the Southern Oscillation index of the same or of the previous season is significantly high and shows prospects for seasonal rainfall prediction.With 5 Figures     

Satellite-based rainfall data reveal a recent drying trend in central equatorial Africa

West-central Uganda, a biodiversity hotspot on the eastern edge of central equatorial Africa (CEA), is a region coping with balancing food security needs of a rapidly growing human population dependent on subsistence agriculture with the conservation of critically endangered species. Documenting and understanding rainfall trends is thus of critical importance in west-central Uganda, but sparse information exists on rainfall trends in CEA during the past several decades. The recently created African Rainfall Climatology version 2 (ARC2) dataset has been shown to perform satisfactorily at identifying rainfall days and estimating seasonal rainfall totals in west-central Uganda. Therefore, we use ARC2 data to assess rainfall trends in west-central Uganda and other parts of equatorial Africa from 1983–2012. The core variables examined were three-month rainfall variables for west-central Uganda, and annual rainfall variables and seasonal rainfall totals for a transect that extended from northwestern Democratic Republic of the Congo to southern Somalia. Significant decreases in rainfall in west-central Uganda occurred for multiple three-month periods centered on boreal summer, and rainfall associated with the two growing seasons decreased by 20 % from 1983–2012. The drying trend in west-central Uganda extended westward into the Congo rainforest. Rainfall in CEA was significantly correlated with the Atlantic Multidecadal Oscillation (AMO) at the annual scale and during boreal summer and autumn. Two other possible causes of the decreasing rainfall in CEA besides North Atlantic Ocean sea-surface temperatures (e.g., AMO), are the warming of the Indian Ocean and increasing concentrations of carbonaceous aerosols over tropical Africa from biomass burning.     

Assessing future climate changes and extreme indicators in east and south Asia using the RegCM4 regional climate model

This paper assesses future climate changes over East and South Asia using a regional climate model (RegCM4) with a 50?km spatial resolution. To evaluate the model performance, RegCM4 is driven with ??perfect boundary forcing?? from the reanalysis data during 1970?C1999 to simulate the present day climate. The model performs well in reproducing not only the mean climate and seasonality but also most of the chosen indicators of climate extremes. Future climate changes are evaluated based on two experiments driven with boundary forcing from the European-Hamburg general climate model (ECHAM5), one for the present (1970?C1999) and one for the SRES A1B future scenario (2070?C2099). The model predicts an annual temperature increase of about 3°?C5° (smaller over the ocean and larger over the land), and an increase of annual precipitation over most of China north of 30°N and a decrease or little change in the rest of China, India and Indochina. For temperature-related extreme indicators in the future, the model predicts a generally longer growing season, more hot days in summer, and less frost days in winter. For precipitation-related extremes, the number of days with more than 10?mm of rainfall is predicted to increase north of 30°N and decrease in the south, and the maximum five-day rainfall amount and daily intensity will increase across the whole model domain. In addition, the maximum number of consecutive dry days is predicted to increase over most of the model domain, south of 40°N. Most of the Yangtze River Basin in China stands out as ??hotspots?? of extreme precipitation changes, with the strongest increases of daily rain intensity, maximum five-day rain amount, and the number of consecutive dry days, suggesting increased risks of both floods and droughts.     

Possible climatic impacts of tropical deforestation

Large-scale conversion of tropical forests into pastures or annual crops will likely lead to changes in the local microclimate of those regions. Larger diurnal fluctuations of surface temperature and humidity deficit, increased surface runoff during rainy periods and decreased runoff during the dry season, and decreased soil moistrue are to be expected.It is likely that evapotranspiration will be reduced because of less available radiative energy at the canopy level since grass presents a higher albedo than forests, also because of the reduced availability of soil moisture at the rooting zone primarily during the dry season. Recent results from general circulation model (GCM) simulations of Amazonian deforestation seem to suggest that the equilibrium climate for a grassy vegetation in Amazonia would be one in which regional precipitation would be significantly reduced.Global climate changes probably will occur if there is a marked change in rainfall patterns in tropical forest regions as a result of deforestation. Besides that, biomass burning of tropical forests is likely adding CO₂ into the atmosphere, thus contributing to the enhanced greenhouse warming.     

洱海流域近50年气候变化特征及其对洱海水资源的影响

利用1961-2010年洱海流域的气候和洱海水资源等资料,统计分析了洱海流域气候变化特征及不同气候类型对洱海水资源量的影响,并建立了洱海水资源量与洱海流域降水量、气温的定量关系,对洱海水资源量进行定量估计.结果表明:近50年洱海流域气温呈波动上升趋势,气候变暖明显;21世纪的第一个10年是洱海流域近50年来最暖的10年.年降水量总体上呈减少趋势.洱海水资源量与年降水量之间有显著的正相关关系,而与气温呈明显的负相关关系.洱海流域气候类型在20世纪60和70年代以偏冷和偏湿为主,进入80年代后开始出现暖年,特别是21世纪的第一个10年,气候以偏暖和偏干为主,未出现过偏冷年.在偏干和偏暖的年份洱海水资源均为枯水年;而偏湿和偏冷的年份洱海水资源多为丰水年;气候正常的年份,洱海水资源多为正常.可根据洱海流域未来气候趋势的预测结果,分别通过气候类型及回归预测方程对洱海水资源的丰欠作定性的估计和定量的预测.     

Seasonal forecasts in the Sahel region: the use of rainfall-based predictive variables

In the Sahel region, seasonal predictions are crucial to alleviate the impacts of climate variability on populations' livelihoods. Agricultural planning (e.g., decisions about sowing date, fertilizer application date, and choice of crop or cultivar) is based on empirical predictive indices whose accuracy to date has not been scientifically proven. This paper attempts to statistically test whether the pattern of rainfall distribution over the May–July period contributes to predicting the real onset date and the nature (wet or dry) of the rainy season, as farmers believe. To that end, we considered historical records of daily rainfall from 51 stations spanning the period 1920–2008 and the different agro-climatic zones in Burkina Faso. We performed (1) principal component analysis to identify climatic zones, based on the patterns of intra-seasonal rainfall, (2) and linear discriminant analysis to find the best rainfall-based variables to distinguish between real and false onset dates of the rainy season, and between wet and dry seasons in each climatic zone. A total of nine climatic zones were identified in each of which, based on rainfall records from May to July, we derived linear discriminant functions to correctly predict the nature of a potential onset date of the rainy season (real or false) and that of the rainy season (dry or wet) in at least three cases out of five. These functions should contribute to alleviating the negative impacts of climate variability in the different climatic zones of Burkina Faso.