Abstract

In this study, the impact of inter-seasonal climate variability on rainfed maize (Zea mays) production over the Wami-Ruvu basin of Tanzania is evaluated. Daily high-resolution climate simulations from the Coordinated Regional Climate Downscaling Experiment_Regional Climate Models (CORDEX_RCMs) are used to drive the Decision Support System for Agro-technological Transfer (DSSAT) to simulate maize yields. Climate simulations for the base period of 35 years (1971–2005) are used to drive DSSAT to simulate maize yields during the historical climate. On the other hand, climate projections for the period 2010–2039 (current), 2040–2069 (mid), and 2070–2099 centuries for two Representative Concentration Pathway (RCP45 and 85) emission scenarios are used to drive DSSAT to simulate maize yields in respective centuries. Statistical approaches based on Pearson correlation coefficient and the coefficients of determination are used in the analysis. Results show that rainfall, maximum temperature, and solar radiation are the most important climate variables that determine variation in rainfed maize yields over the Wami-Ruvu basin of Tanzania. They explain the variability in maize yields in historical climate condition (1971–2005), present century under RCP 4.5, and mid and end centuries under both RCP 4.5 and RCP 8.5.

INTRODUCTION

The agriculture sector is sensitive to climate variability (Khan et al. 2009), especially the inter-annual variability of precipitation, temperature patterns, and extreme weather events (droughts and floods). These climatic events are predicted to increase in future and are expected to have many and significant consequences to the agriculture sector (IPCC 2007, 2012, 2013; Sarker et al. 2012; Akram 2012). This would have a negative influence on food prices, food security, and land use decisions (Lobell & Field 2007; van Wart et al. 2013). African countries are particularly vulnerable to climate variability. This is due to their dependence on weather-sensitive agriculture (Rowhani et al. 2011). Moreover, the agriculture sector in Africa is dominated by smallholder farmers with limited access to technology and the resources to adapt (Amikuzuno & Donkoh 2012). The Intergovernmental Panel on Climate Change (IPCC) consistently predicts millions of people in Africa will face increased water stress due to climate variability (IPCC 2007, 2012, 2013). Yields from rainfed agriculture in some African countries could be reduced by up to 50% by 2020 (IPCC 2007).

In order to prevent the future destructive impact of climate variability on food production, it is crucial to adjust or suggest possible policies to cope with increased climate variability. African countries need to build a national legal framework to manage food resources in accordance with the anticipated climate variability. However, before devising a policy to cope with the impacts of climate variability especially to the agriculture sector, it is critical to have a clear understanding of how climate variability affects different food crops.

To date, most studies have focused on analyzing the impact of mean climate on crop yields (Mwandosya et al. 1998; Agrawala et al. 2003; Ehrhart & Twena 2006; IPCC 2007; Enfors & Gordon 2008; Thornton et al. 2009, 2010; Arndt et al. 2011; Müller et al. 2011; Luhunga et al. 2016). Studies that examine the influence of climate variability on crop production have received less attention (Rowhani et al. 2011). It is crucial to understand how climatic variables and crop yields are linked over time. Moreover, concentrating only on the analysis of the impacts of mean climate on crop production seriously undermines the analysis of the full impacts of climate change on cropping systems (Thornton et al. 2010).

Climate variability will have a more adverse impact on crop production than changes in mean climate alone (Morton 2007). Studies by Case (2006) and Mubaya et al. (2014) have reported increased temperature and decreased rainfall in east African regions that affected water availability, food security, and human health. In 2011, the Horn of Africa experienced a serious drought that affected food and water availability (Klein et al. 2016). This drought was followed with two consecutive years of heavy rainfall that triggered flooding events in parts of Kenya and Somalia (Klein et al. 2016).

Other studies by Kijazi & Reason (2009), Shemsanga et al. (2010) and Tumbo et al. (2010) have suggested that increased frequency of extreme weather events (floods and droughts) affected food production in many areas in Tanzania. In 2012, strong winds and hailstorms damaged food crops and forests in southwestern Tanzania (Kruger et al. 2012). On 16 March 2012, the Iringa region in the south of Tanzania experienced unusually heavy snow that affected crops and forests (Kruger et al. 2012).

However, few studies, if any, have analyzed the impacts of present and future climate variability on crop yields, especially on maize crop, which is the most important cereal crop in perspectives on financial value and food security in Tanzania (Rowhani et al. 2011; Washington & Pearce 2012). Rowhani et al. (2011) used observed monthly climate variables to analyze the influence of historical climate variability on crop production in Tanzania using statistical approaches to create regression equations between yields and climate variables for comparison. Msongaleli et al. (2015) used climate data from the general circulation models (GCMs) to analyze the impacts of climate variability and change on rainfed sorghum and maize over central Tanzania. However, the GCMs have coarse space resolution and are designed to simulate global climate characteristics like global temperature. The coarse space resolution of the GCMs severely limits the direct application of their output in regional and sub-regional decision-making (Masson & Knutti 2011; Ramirez-Villegas & Challinor 2012). This study evaluates the impacts of present and future climate variability on rainfed maize production over the Wami-Ruvu basin of Tanzania using high resolution climate data from the Coordinated Regional Climate Downscaling Experiment Regional Climate Models (CORDEX_RCMs). Moreover, this study analyzes the impacts of inter-seasonal fluctuation of temperature solar radiation and rainfall on maize yields.

DATA AND METHODOLOGY

Study area

Based on the land morphology, Tanzania is divided into nine basins: Lake Victoria, Wami-Ruvu, Lake Tanganyika, internal drainage, Pangani, Rufiji, Lake Nyasa, Rukwa, and Ruvuma basins (URT 2013). The Wami-Ruvu basin is located in the eastern part of Tanzania (Figure 1). It lies between latitudes 5°–7°S and longitudes 36°–39°E. The basin covers an area of about 66,820 km2, and is located in six regions: Dar es Salaam, parts of Coast, Morogoro, Dodoma, Tanga, and Manyara. The topographical features of the Wami-Ruvu basin are detailed as described in URT (2013), that is, it is covered by low-lying and mountain landscape. The dominant mountain landscape includes the Uluguru Mountains with an altitude of 400 m to 2,500 m, Nguru Mountains with an altitude of 400 to 2,000 m, Rubeho Mountains with an altitude of 500 to 1,000 m, Ukaguru Mountains with an altitude of 400 to 1,000 m, and Nguu Mountains located in the western part of Wami River with an altitude of 400 to 2,000 m above mean sea level. The low-lying areas include Mkata plains with an altitude of 400–800 m, Lower Wami with an altitude of 200–400 m, Kisaki located south-east of Uluguru mountain with an altitude of 140 to 200 m, and Kimbiji and Mbezi located to the southern coastal area of Dar es Salaam with an altitude of 50 to 100 m above mean sea level.

Figure 1

A map of Africa and Tanzania showing the location of the Wami-Ruvu basin.

Figure 1

A map of Africa and Tanzania showing the location of the Wami-Ruvu basin.

The Wami-Ruvu basin is divided into three sub-basins, namely, Wami, Ruvu, and Coastal sub-basins. The basin experiences both unimodal and bimodal rainfall patterns. The former is mainly experienced in the Wami sub-basin, while the latter is mainly experienced in Ruvu and Coastal sub-basins.

Rainfall patterns in the Wami-Ruvu basin are mainly driven by the seasonal migration of the Inter Tropical Convergence Zone. However, the variation of climatic rainfall within the basin is possibly influenced by orographic features. For instance, in the plains, annual rainfall ranges from 800 mm to 1,000 mm near the coast and 500–600 mm inland towards Dodoma and north of Wami sub-basin (Luhunga et al. 2016). Over the high grounds, in the Uluguru Mountains, the annual rainfall exceeds 1,500 mm (Mwandosya et al. 1998). The eastern parts of Uluguru Mountain receive annual rainfall in the range of 2,500–4,000 mm while the western slopes receive an annual rainfall of 1,200–3,100 mm (Mbwanga 2005). The Nguru-Rubeho Mountains receive rainfall between 800 and 1,200 mm, and the Ukaguru Mountains 1,000–1,800 mm annually (Kashaigili 2011). The average monthly minimum and maximum temperatures are almost the same throughout the basin; the coldest month is August, with a temperature of about 18°C, and the hottest month is February, with a temperature of about 32° C (Ngana et al. 2010). The annual average maximum temperature in the plains ranges from 28.9 to 30.7° C, while minimum temperature ranges from 13.7 to 23.8° C. Potential evaporation in the Wami-Ruvu basin is 2,000 mm per year and between 1,600 and 1,800 mm per year in the Uluguru Mountains (Mwandosya et al. 1998).

The Wami-Ruvu basin is characterized by 12 soil types, namely, Cambisols, Ferralsols, Acrisols, Fluvisols, Luvisols, Lixisols, Arenosols, Leptosols, Nitisols, Vertisols, Planosols, and Haplic Phaeozems (URT 2013; Luhunga et al. 2016). The dominant soil types are Cambisols, which cover parts of Bagamoyo, Kisarawe, Mkuranga, Morogoro Rural, Dodoma Urban, Bahi, and Chamwino. These types of soil make good agricultural land and are intensively used for agriculture production.

As a result of favorable climate condition and fertile soils, recently, the Wami-Ruvu basin has witnessed many agriculture projects (e.g., Southern Agriculture Growth Corridor of Tanzania (SAGCOT), Feed the Future, and Rural Livelihoods Development Programme (RLDP)). These projects aim to develop southern Tanzania, Wami-Ruvu basin included, into a major regional food producer and engine of national economic development to dramatically reduce poverty among its nine million residents and sustain the region's ecosystems as the productive base of future well-being (Milder et al. 2012).

Despite the high agricultural potential and a large number of investments in the Wami-Ruvu basin, few studies, if any, have analyzed the impacts of climate variability on rainfed crop production using high-resolution climate datasets. Therefore, the study area is best placed in this kind of research to provide reliable information about the impacts of climate variability on rainfed maize production that can be used to prepare adaptation strategies by policy- and decision-makers.

Data

Data from RCM simulations

Daily climate variables (solar radiation, rainfall, minimum and maximum temperatures) from the CORDEX_RCMs are used to drive the crop model. The CORDEX program archives output from RCMs forced by different GCMs over different countries in the world. All CORDEX_RCMs have the spatial resolution of longitude 0.44° and latitude 0.44°, which is approximately 50 km by 50 km. The detailed descriptions of CORDEX_RCMs and their dynamics and physical parameterization are found in Nikulin et al. (2012).

Three CORDEX_RCMs driven by three GCMs for two Representative Concentration Pathway (RCP 4.5 and RCP 8.5) scenarios (Figure 2) are used. Table 1 lists the RCMs and the driving GCMs used in this study.

Table 1

CORDEX_RCMs and the driving GCMs

No. Regional climate model Model center Short name of RCM GCM 
DMI HIRHAM5 Danmarks Meteorologiske Institut (DMI), Denmark HIRHAM5 1. ICHEC-EC-EARTH 
SMHI Rossby Center Regional Atmospheric Model (RCA4) Sveriges Meteorologiska och Hydrologiska Institut (SMHI), Sweden RCA4 1. MPI-M-MPI-ESM-LR; 2. ICHEC-EC-EARTH; 3. CNRM-CERFACS-CNRM-CM5 
KNMI Regional Atmospheric Climate Model, version 2.2 (RACMO2.2T) Koninklijk Nederlands Meteorologisch Instituut (KNMI), The Netherlands RACMO22T 1. ICHEC-EC-EARTH 
No. Regional climate model Model center Short name of RCM GCM 
DMI HIRHAM5 Danmarks Meteorologiske Institut (DMI), Denmark HIRHAM5 1. ICHEC-EC-EARTH 
SMHI Rossby Center Regional Atmospheric Model (RCA4) Sveriges Meteorologiska och Hydrologiska Institut (SMHI), Sweden RCA4 1. MPI-M-MPI-ESM-LR; 2. ICHEC-EC-EARTH; 3. CNRM-CERFACS-CNRM-CM5 
KNMI Regional Atmospheric Climate Model, version 2.2 (RACMO2.2T) Koninklijk Nederlands Meteorologisch Instituut (KNMI), The Netherlands RACMO22T 1. ICHEC-EC-EARTH 
Figure 2

Emission of greenhouse gases across the RCPs (figure adopted from van Vuuren et al. (2011)).

Figure 2

Emission of greenhouse gases across the RCPs (figure adopted from van Vuuren et al. (2011)).

The RCMs simulate climate variables at grids, and the interpolation technique of inverse distance weighting is used to transfer model grid climate simulations to the location where experimental farming is carried out. For a detailed description about the inverse interpolation weighting, the reader may consult Hartkamp et al. (1999) and Ly et al. (2013). Interpolated climate data of maximum and minimum temperature, precipitation, and solar radiation from individual CORDEX_RCMs are used as input into the crop model for simulation of maize yield under present and future climate.

Soil and crop and management practices' data

Twenty soil profiles' data were used in this study; 12 were obtained from the Africa soils database and eight were excavated in the study regions. The soil water properties calculator was used to estimate different hydrological characteristics of soil profiles (Saxton & Rawls 2009). Information about management practices and previous maize yields was obtained from a household panel survey database from the National Bureau of Statistics (2012). The planting date and type of maize cultivate were obtained from a field survey conducted across the study area. For detailed descriptions about the computation of the hydrological characteristics of layers for each soil profile and information from different farms used to create crop model input files, the reader may consult Luhunga et al. (2016).

Crop models

Crop simulation models are mathematical representations of crop growth. They simulate crop growth by numerical integration of constituent processes with the aid of computers (Graves et al. 2002). Crop models can be broadly categorized into two types: descriptive and explanatory. A descriptive model is constructed using experimental field data, where mathematical equations are derived to explain the behavior of the systems in a simple way (Miglietta & Bindi 1993). A system in agronomical terms may refer to crop with its elements (plant organs: leaf, stem, roots, and processes: growth, photosynthesis, transpiration, etc.). Descriptive models are suitable for quick examination of crop behavior under field conditions where environmental conditions remain relatively stable (Luhunga et al. 2016). The explanatory model consists of a quantitative description of the processes and mechanisms that influence the behavior of the system. The formulations of this type of model require detailed analysis of a system, where its processes and mechanisms are quantified separately. The model is formulated using the detailed description of a system that includes processes – leaf area expansion, tiller production, etc. An example of explanatory crop growth model is the Decision Support System for Agro-technological Transfer (DSSAT) (Jones et al. 2003). This model has been used to simulate crop growth for more than 20 years by different researchers, educators, consultants, extension agents, growers, and policy- and decision-makers worldwide (Hoogenboom et al. 2010). In this study, we used DSSAT v4.5 to simulate maize yields over the Wami-Ruvu basin. This version comprises 28 different crop growth models. It has new tools that facilitate the creation and management of experimental, soil, and weather data files. DSSAT v4.5 includes improved application programs for seasonal, spatial, sequence and crop rotation analyses that assess the economic risks and environmental impacts associated with irrigation, fertilizer and nutrient management, climate variability, climate change, soil carbon sequestration, and precision management (Luhunga et al. 2016). The minimum dataset required to run DSSAT is: site weather data for the duration of the growing season, soil profile and soil surface data, crop management data from the experiment, and observed experimental data from the experiment. Other information includes (1) latitude and longitude of the weather station, (2) daily values of incoming solar radiation (MJ/m2-day), (3) maximum and minimum daily air temperature (°C), and (4) daily total rainfall (mm). The dry and wet bulb temperatures and wind speed data may also be included as optional data.

Methods

Climate data from three individual RCMs forced by three GCMs are used as input into DSSAT to simulate maize yields for historical (1971–2005) climate conditions over Wami-Ruvu basin of Tanzania. Moreover, climate data from the three RCMs forced by three GCMs for two concentration pathway (RCP4.5 and RCP 8.5) scenarios are used as input into DSSAT to simulate maize yields for the present (2010–2039), mid (2040–2069), and end (2070–2099) centuries.

In order to minimize the uncertainties associated with the individual RCM-GCM combination, the ensemble average climate data from six RCM-GCM combinations were created and used to drive the crop model to simulate maize yields under historical (1971–2005) and future (2010–2099) climate conditions. Statistical analysis such as Pearson correlation coefficient and coefficient of determination was performed to assess the relationship between the simulated maize yields and climate variables (rainfall, solar radiation, minimum and maximum temperatures). Furthermore, the influence of climate variables on the duration of growing seasons is analyzed.

RESULTS

This section has two sub-sections that analyze the results. The first sub-section presents the statistical analyses using the Pearson correlation coefficient that look to obtain the association between climate variables and maize yields. The inter-seasons variation of the climate variables (rainfall, solar radiation, minimum and maximum temperatures), length of growing seasons, and maize yields are presented in the second sub-section.

The Pearson correlation coefficient between climate variables and maize yields

The Pearson correlation coefficient, which is the measure of statistical relationship between two variables, was performed to measure the strength of the relationship between climate variables (as independent variables) and maize yield (as dependent variables). Table 2 indicates the Pearson correlation coefficient and the coefficient of determination between inter-seasonal variation of maize yield and climate variables during the historical climate condition (1971–2005). Results reveal that there was a significant strong positive relationship between rainfall variability and maize yield (r = 0.7, p = 0.0001). Seasonal variation in maximum temperature and solar radiation exhibited a significant strong negative relationship with maize yield, with coefficient of (r = −0.6, p = 0.002) and (r = −0.7, p = 0.0001), respectively (Table 2). There was a negligible and non-significant negative relationship between seasonal variation in minimum temperature and maize yield (r = −0.1, p = 0.496). Moreover, there was a strong and statistically significant positive relationship between variation in the length of growing season and maize yield (r = 0.6, p = 0.001). The length of growing season revealed a strong and statistically significant relationship with rainfall (r = 0.7, p = 0.0001) and significant strong negative relationships with solar radiation and minimum and maximum temperatures. In general, results from the present study agree with a prior study by Omoyo et al. (2015), who indicated strong and statistically significant relationships between climate variables and maize yields in the arid and semi-arid lands of lower eastern Kenya, where maximum temperature correlated negatively with maize yields. The presented results also agree with the study by Adamgbe & Ujoh (2013), who indicated a strong and statistically significant positive relationship between rainfall and maize yields in Gboko, Nigeria.

Table 2

The correlation coefficient (r), the p-values, and the coefficient of determination (R2) between simulated inter-season maize yields and inter-season climate variables over the Wami-Ruvu basin of Tanzania in the base period (1971–2005)

Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001 0.7 0.5             
SR 0.0001 − 0.7 0.4 0.001 − 0.6 0.3          
TN 0.496 −0.1 0.02 0.118 −0.3 0.1 0.745 −0.1 0.004       
TX 0.002 − 0.6 0.3 0.0001 − 0.6 0.4 0.0001 0.6 0.388 0.0001 0.7 0.44    
LGS 0.001 0.6 0.4 0.0001 0.7 0.4 0.006 −0.5 0.238 0.0001 −0.8 0.56 <0.0001 − 0.9 0.83 
Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001 0.7 0.5             
SR 0.0001 − 0.7 0.4 0.001 − 0.6 0.3          
TN 0.496 −0.1 0.02 0.118 −0.3 0.1 0.745 −0.1 0.004       
TX 0.002 − 0.6 0.3 0.0001 − 0.6 0.4 0.0001 0.6 0.388 0.0001 0.7 0.44    
LGS 0.001 0.6 0.4 0.0001 0.7 0.4 0.006 −0.5 0.238 0.0001 −0.8 0.56 <0.0001 − 0.9 0.83 

The values in bold are statistically significant at p < 0.05.

RF, rainfall; SR, solar radiation; TN, minimum temperature; TX, maximum temperature; LGS, length of growing season.

Results from the coefficient of determination indicated that during historical climate condition, seasonal rainfall and solar radiation, respectively, explained 50% and 40% of maize yield variability. Maximum and minimum temperatures, respectively, explained 30% and 2% of maize yield variability. These findings are in agreement with a prior study by Huang et al. (2015), who indicated rainfall as a most important factor to explain variability in maize yields in the eastern United States. The presented results also agree with the study by Matsui (2016), who indicated rainfall variability as the dominant factor controlling variability in maize yields in four northern districts of Malawi in historical climate condition.

Table 3 shows the Pearson correlation coefficient and the coefficient of determination between inter-seasonal variations of maize yield and climate variables during the present century (2010–2039). Results reveal that there will be a significant moderate negative relationship between maximum temperature and maize yield in the present century under RCP 8.5 emission scenario (r = −0.5, p = 0.012). Rainfall is expected to have a weak and non-significant positive relationship with maize yield in the present century under RCP 8.5 (r = 0.2, p = 0.228), whereas minimum temperature and solar radiation are expected to have a weak and non-significant negative relationship with maize yield. However, under RCP 4.5, rainfall is expected to have a significant and strong positive relationship with maize yield (r = 0.7, p < 0.0001) and maximum temperature is expected to have a significant and weak negative relationship with maize yields (r = −0.4, p = 0.031).

Table 3

The correlation coefficient (r), the p-values, and the coefficient of determination (R2) between simulated inter-season maize yields and inter-season climate variables over the Wami-Ruvu basin of Tanzania in the present century (2010–2039) under RCP 8.5 and RCP 4.5

Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.228a, <0.0001b 0.2a, 0.7b 0.05a, 0.45b             
SR 0.197a, <0.0001b −0.2a, − 0.7b 0.06a, 0.44b 0.090a, 0.003b −0.3a, − 0.5b 0.10a, 0.27b          
TN 0.104a, 0.113b −0.3a, 0.3b 0.09a, 0.09b 0.438a, 0.570b −0.2a, 0.1b 0.02a, 0.01b 0.196a, 0.110b −0.2a, −0.3b 0.06a, 0.1b       
TX 0.012a, 0.031b − 0.5a, −0.4b 0.20a, 0.16b 0.010a, 0.008b − 0.5a, −0.5b 0.21a, 0.23b 0.124a, 0.007b 0.3a, 0.5b 0.08a, 0.23b 0.0001a, 0.0001b 0.8a, 0.6b 0.67a, 0.38b    
LGS 0.038a, 0.152b 0.4a, 0.3b 0.15a, 0.07b 0.141a, 0.052b 0.3a, 0.4b 0.08a, 0.13b 0.764a, 0.142b −0.1a, −0.3b 0.003a, 0.1b 0.0001a, 0.0001b − 0.9a, −0.7b 0.86a, 0.55b < 0.0001a, <0.0001b 0.9a, −0.9b 0.89a, 0.83b 
Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.228a, <0.0001b 0.2a, 0.7b 0.05a, 0.45b             
SR 0.197a, <0.0001b −0.2a, − 0.7b 0.06a, 0.44b 0.090a, 0.003b −0.3a, − 0.5b 0.10a, 0.27b          
TN 0.104a, 0.113b −0.3a, 0.3b 0.09a, 0.09b 0.438a, 0.570b −0.2a, 0.1b 0.02a, 0.01b 0.196a, 0.110b −0.2a, −0.3b 0.06a, 0.1b       
TX 0.012a, 0.031b − 0.5a, −0.4b 0.20a, 0.16b 0.010a, 0.008b − 0.5a, −0.5b 0.21a, 0.23b 0.124a, 0.007b 0.3a, 0.5b 0.08a, 0.23b 0.0001a, 0.0001b 0.8a, 0.6b 0.67a, 0.38b    
LGS 0.038a, 0.152b 0.4a, 0.3b 0.15a, 0.07b 0.141a, 0.052b 0.3a, 0.4b 0.08a, 0.13b 0.764a, 0.142b −0.1a, −0.3b 0.003a, 0.1b 0.0001a, 0.0001b − 0.9a, −0.7b 0.86a, 0.55b < 0.0001a, <0.0001b 0.9a, −0.9b 0.89a, 0.83b 

The values in bold are statistically significant at p < 0.05.

LGS, length of growing season; RF, rainfall; SR, solar radiation; TN, minimum temperature; TX, maximum temperature.

aRCP 8.5.

bRCP 4.5.

The coefficient of determination shows that, under business as usual emission scenario (RCP 8.5), rainfall will no longer dominate to explain variability in maize yield as was the case in the historical climate condition, but variability in maximum temperature will be the dominant factor to explain variability in maize yield by 20%. However, under controlled emission scenario (RCP 4.5), rainfall will continue to be the dominant factor in explaining variability in maize yield by 45%.

Table 4 indicates Pearson correlation coefficients and the coefficient of determination between inter-seasonal variation of maize yields and climate variables in the mid-century under RCP 8.5 and RCP 4.5. Results reveal that there will be a significant and strong positive relationship between rainfall and maize yield in mid century under RCP 4.5 and RCP 8.5. Solar radiation is expected to have a significant negative relationship with maize yield of (r = −0.7, p = 0.0001) under RCP 8.5 and (r = −0.6, p = 0.002) under RCP 4.5. The coefficient of determination reveals that rainfall will be the dominant factor to explain the variability in maize yield in mid century under RCP 8.5 by 60%. However, under RCP 4.5 maximum temperature will be the dominant climate variable to explain the variability in maize yield by 34%.

Table 4

The correlation coefficient (r), the p-values, and the coefficient of determination (R2) between simulated inter-season maize yields and inter-season climate variables over the Wami-Ruvu basin of Tanzania in the mid century (2040–2069) under RCP 8.5 and RCP 4.5

Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001a, 0.002b 0.8a, 0.5b 0.60a, 0.28b             
SR 0.0001a, 0.002b − 0.7a, −0.6b 0.53a, 0.30b 0.0001a, 0.089b − 0.8a, −0.3b 0.65a, 0.10b          
TN 0.624a, 0.027b −0.1a, − 0.4b 0.01a, 0.16b 0.642a, 0.005b −0.1a, − 0.5b 0.01a, 0.25b 0.971a, 0.727b 0.01a, −0.07b 0.00a, 0.004b       
TX 0.026a, 0.001b − 0.4a, −0.6b 0.17a, 0.34b 0.010a, 0.0001b − 0.5a, −0.7b 0.21a, 0.47b 0.023a, 0.010b 0.4a, 0.5b 0.17a, 0.22b 0.0001a, <0.0001b 0.9a, 0.8b 0.8a, 0.6b    
LGS 0.035a, 0.006b 0.4a, 0.5b 0.15a, 0.24b 0.049a, 0.000b 0.4a, 0.6b 0.13a, 0.37b 0.092a, 0.167b −0.3a, −0.3b 0.09a, 0.07b 0.0001a, <0.0001b − 0.9a, −0.9b 0.9a, 0.8b 0.0001a, <0.0001b − 0.9a, −0.9b 0.95a, 0.85b 
Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001a, 0.002b 0.8a, 0.5b 0.60a, 0.28b             
SR 0.0001a, 0.002b − 0.7a, −0.6b 0.53a, 0.30b 0.0001a, 0.089b − 0.8a, −0.3b 0.65a, 0.10b          
TN 0.624a, 0.027b −0.1a, − 0.4b 0.01a, 0.16b 0.642a, 0.005b −0.1a, − 0.5b 0.01a, 0.25b 0.971a, 0.727b 0.01a, −0.07b 0.00a, 0.004b       
TX 0.026a, 0.001b − 0.4a, −0.6b 0.17a, 0.34b 0.010a, 0.0001b − 0.5a, −0.7b 0.21a, 0.47b 0.023a, 0.010b 0.4a, 0.5b 0.17a, 0.22b 0.0001a, <0.0001b 0.9a, 0.8b 0.8a, 0.6b    
LGS 0.035a, 0.006b 0.4a, 0.5b 0.15a, 0.24b 0.049a, 0.000b 0.4a, 0.6b 0.13a, 0.37b 0.092a, 0.167b −0.3a, −0.3b 0.09a, 0.07b 0.0001a, <0.0001b − 0.9a, −0.9b 0.9a, 0.8b 0.0001a, <0.0001b − 0.9a, −0.9b 0.95a, 0.85b 

The values in bold are statistically significant at p< 0.05.

LGS, length of growing season; RF, rainfall; SR, solar radiation; TN, minimum temperature; TX, maximum temperature.

aRCP 8.5.

bRCP 4.5.

The correlation coefficient and the coefficient of determination between the inter-seasonal climate variables and maize yields in the end century under RCP 8.5 and RCP 4.5 are presented in Table 5. This table reveals that there will be a significant strong positive relationship of r = 0.6, p = 0.0001 between rainfall and maize yields in the end century under both RCP 4.5 and RCP 8.5. Maximum temperature and solar radiation are expected to have a significant negative relationship with maize yield. The coefficient of determination indicates that rainfall will be the dominant factor to explain variability in maize yields in the end century under both RCP 4.5 and RCP 8.5.

Table 5

The correlation coefficient (r), the p-values, and the coefficient of determination (R2) between simulated inter-season maize yields and inter-season climate variables over the Wami-Ruvu basin of Tanzania in the end century (2070–2099) under RCP 8.5 and RCP 4.5

Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001a, 0.0001b 0.6a, 0.6b 0.40a, 0.41b             
SR 0.0001a, 0.005b − 0.6a, −0.5b 0.40a, 0.25b 0.0001a, 0.002b − 0.7a, −0.5b 0.55a, 0.29b          
TN 0.551a, 0.956b −0.1a, 0.01b 0.01a, 0.00b 0.161a, 0.658b 0.3a, −0.1 0.07a, 0.01b 0.070a, 0.175b −0.3a, −0.3b 0.11a, 0.07b       
TX 0.026, 0.002b − 0.4a, −0.5b 0.16a, 0.30b 0.647a, 0.0001b −0.1a, − 0.7b 0.01a, 0.51b 0.556a, 0.0001b 0.1a, 0.7b 0.01a, 0.43b 0.0001a, 0.016b 0.88a, 0.44b 0.78a, 0.19b    
LGS 0.176a, 0.011b 0.3a, 0.5b 0.06a, 0.21b 0.647a, 0.002b −0.1a, 0.6b 0.01a, 0.30b 0.525a, 0.111b 0.1a, −0.3b 0.02a, 0.09b 0.0001a, 0.0001b − 0.97a, −0.73b 0.94a, 0.53b 0.0001a, <0.0001b − 0.96a, −0.84b 0.93a, 0.70b 
Variables Maize yields
 
RF
 
SR
 
TN
 
TX
 
p-values R2 p-values R2 p-values R2 p-values R2 p-values R2 
RF 0.0001a, 0.0001b 0.6a, 0.6b 0.40a, 0.41b             
SR 0.0001a, 0.005b − 0.6a, −0.5b 0.40a, 0.25b 0.0001a, 0.002b − 0.7a, −0.5b 0.55a, 0.29b          
TN 0.551a, 0.956b −0.1a, 0.01b 0.01a, 0.00b 0.161a, 0.658b 0.3a, −0.1 0.07a, 0.01b 0.070a, 0.175b −0.3a, −0.3b 0.11a, 0.07b       
TX 0.026, 0.002b − 0.4a, −0.5b 0.16a, 0.30b 0.647a, 0.0001b −0.1a, − 0.7b 0.01a, 0.51b 0.556a, 0.0001b 0.1a, 0.7b 0.01a, 0.43b 0.0001a, 0.016b 0.88a, 0.44b 0.78a, 0.19b    
LGS 0.176a, 0.011b 0.3a, 0.5b 0.06a, 0.21b 0.647a, 0.002b −0.1a, 0.6b 0.01a, 0.30b 0.525a, 0.111b 0.1a, −0.3b 0.02a, 0.09b 0.0001a, 0.0001b − 0.97a, −0.73b 0.94a, 0.53b 0.0001a, <0.0001b − 0.96a, −0.84b 0.93a, 0.70b 

The values in bold are statistically significant at p < 0.05.

LGS, length of growing season; RF, rainfall; SR, solar radiation; TN, minimum temperature; TX, maximum temperature.

aRCP 8.5.

bRCP 4.5.

Inter-seasonal variations of maize yields and climate variables

Figure 3 presents the inter-seasonal variation of maize yields in present (2010–2039), mid (2040–2069), and end (2070–2099) century under RCP 4.5 and RCP 8.5. This figure indicates maize yields simulated by five ensemble members and the ensemble mean of yields. Results reveal spread of maize yields among ensemble members. The spread in maize yields among ensemble members is associated with the RCM-GCM combination used to drive the crop model. For instance, high spread of maize yield was observed when RACMO22-ICHEC and RCA4-CNRM were used to drive DSSAT to simulate maize yields. The spread of maize yields for RCA4-CNRM driven simulations is associated with the driving GCM (CNRM). This is due to the fact that maize yields when DSSAT was forced with the same RCM (RCA4) but different GCMs (MPI and ICHEC) are related to the ensemble of the yields.

Figure 3

Simulated maize yields over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 3

Simulated maize yields over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

The deviation of simulated maize yields when DSSAT was forced by climate variables from RACMO22-ICHEC is associated with a difference in RCM formulation. This is due to the fact that simulated maize yields when DSSAT was forced by HIRHAM5 and RCA4 both forced by ICHEC are related to the ensemble mean of the yields. The deviation of simulated maize yields can also be explained based on the performance of the climate model to simulate seasonal variation of maximum temperature. For instance, high maize yields are simulated by DSSAT forced by lower seasonal maximum temperature from RACMO22T-ICHEC, and lower maize yields are simulated by DSSAT forced by high seasonal maximum temperature from the RCA4-CNRM model.

The inter-seasonal variations of maize yields simulated by ensemble members, especially for the RCA4-CNRM and RACMO22T-ICHEC driven simulations, indicates higher variations compared with the ensemble average of the yields. The variation of simulated maize yields by individual ensemble members is possibly due to high variations in seasonal rainfall, solar radiation, and minimum and maximum temperatures (Figures 48). It should be noted that maximum temperature varies inversely with the simulated maize yields (Tables 35) and the minimum temperature varies inversely with the length of growing season (Tables 35).

Figure 4

Inter-seasonal variation of rainfall over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 4

Inter-seasonal variation of rainfall over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 5

Inter-seasonal variation of maximum temperature over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 5

Inter-seasonal variation of maximum temperature over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 6

Inter-seasonal variation of minimum temperature over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 6

Inter-seasonal variation of minimum temperature over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 7

Inter-seasonal variation of solar radiation over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 7

Inter-seasonal variation of solar radiation over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 8

Inter-seasonal variation of length of growing season over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

Figure 8

Inter-seasonal variation of length of growing season over the Wami-Ruvu basin under RCP 4.5 and RCP 8.5. (a) RCP 4.5, present century; (b) RCP 4.5, mid century; (c) RCP 4.5, end century; (d) RCP 8.5, present century; (e) RCP 8.5, mid century; (f) RCP 8.5 end century.

The inter-annual variation of the yields shows quite regular cycles of maize yields with alternating peaks and troughs during present, mid, and end centuries. It is important to note that from 2014 to 2016, the Wami-Ruvu basin experienced lower maize yields. This was due to decreased seasonal rainfall, increased temperatures, and solar radiation that reduced the length of the growing season.

The ensemble average of the yields shows that from 2020 to 2026 the Wami-Ruvu basin will experience decreased maize yields. This decrease in maize yields will be due to a decrease in seasonal rainfall, increase in seasonal temperatures and solar radiation that will reduce the length of the growing season (Figures 48). Moreover, from 2043 to 2044, all the ensemble members and the ensemble average of the yields predict decreased maize yields over Wami-Ruvu basin. This will be influenced by a sudden decrease in seasonal rainfall, increased minimum and maximum temperatures, and increased solar radiation. These climate variables will shorten the length of growing season, which, in turn, will affect maize yields.

DISCUSSION

The impact of climate variability on maize yields over the Wami-Ruvu basin of Tanzania was evaluated. High resolution climate simulations from three CORDEX_RCMs driven by three GCMs were used to drive DSSAT to simulate maize yields in historical (1971–2005) and future (present (2010–2039), mid (2040–2069), and end (2070–2099)) climate conditions under RCP 4.5 and RCP 8.5. Results from the Pearson correlation coefficient and coefficient of determination indicated that in historical climate condition, inter-seasonal variation of rainfall played a dominant role in explaining the inter-seasonal variation of maize yields in the Wami-Ruvu basin. However, in the present century (2010–2039) under RCP 8.5, maximum temperature will play a significant role in explaining the variation of maize yields in the Wami-Ruvu basin. Moreover, in the present century under RCP 4.5, rainfall will play a significant role in explaining variability in maize yields over the Wami-Ruvu basin.

In mid century under RCP 4.5 and RCP 8.5, there will be a significant and strong positive relationship between rainfall and maize yield. Solar radiation is expected to have a significant negative relationship with maize yield of r =−0.7, p = 0.0001 under RCP 8.5 and r = −0.6, p = 0.002 under RCP 4.5. The coefficient of determination reveals that rainfall will be the dominant factor to explain the variability in maize yield in mid century under RCP 8.5 by 60%. However, under RCP 4.5, maximum temperature will be the dominant climate variable to explain the variability in maize yield by 34%.

In the end century, under both RCP 4.5 and RCP 8.5, there will be a significant strong positive relationship of r = 0.6, p = 0.0001 between rainfall and maize yields. Maximum temperature and solar radiation are expected to have a significant negative relationship with maize yield. The coefficient of determination indicated that rainfall will be the dominant factor to explain variability in maize yields in the end century under both RCP 4.5 and RCP 8.5.

Figures 38 indicate the interpersonal variation of simulated maize yields, rainfall, maximum temperature, minimum temperature, solar radiation, and the length of growing season. Simulated inter-seasonal variation in maize yields shows spread between the ensemble members. This is an indication of the uncertainty associated with the projections when one opts to consider individual models for the analysis. The ensemble average of the yields shows regular cycles with patterns of peaks and troughs of maize yields. However, from 2020 to 2026 the ensemble average of the yields predicts decline in maize yields due to increased maximum temperature that reduces the length of growing season. In 2043–2044 the ensemble average of the yields also predicts a decline in maize yields due to an abrupt decrease in seasonal rainfall and an increase in seasonal temperatures and solar radiation. It is important to mention that the projected increase in seasonal temperatures and solar radiation and decreased seasonal rainfall will affect more the length of growing season, which, in turn, affects maize yields.

CONCLUSION

In this study, the impact of climate variability on maize yields over the Wami-Ruvu basin of Tanzania was carried out using high resolution climate simulations from CORDEX_RCMs. Daily solar radiation, rainfall, and minimum and maximum temperatures were used to drive DSSAT to simulate maize yields during the periods 1971–2039, 2040–2069, and 2070–2099. Maize simulations were carried out under two representative concentration pathways (RCP 4.5 and RCP 8.5). Pearson correlation coefficient, which is the measure of statistical relationship between two variables, was performed to measure the strength of the relationship between climate variables (as independent variables) and maize yield (as dependent variable). Results revealed that there was a significant and strong positive relationship between rainfall variability and maize yield (r = 0.7, p = 0.0001). Seasonal variation in maximum temperature and solar radiation exhibited a significant strong negative relationship with maize yield, with coefficient of r = −0.6, p = 0.002 and r = −0.7, p = 0.0001, respectively. There was a negligible and non-significant negative relationship between seasonal variation in minimum temperature and maize yield (r = −0.1, p = 0.496). Moreover, there was a strong and statistically significant positive relationship between variation in the length of growing season and maize yield (r = 0.6, p = 0.001). The length of growing season revealed a strong and statistically significant relationship with rainfall (r = 0.7, p = 0.0001) and a significant strong negative relationship with solar radiation and minimum and maximum temperatures.

Results from the coefficient of determination indicated that during historical climate condition, seasonal rainfall and solar radiation, respectively, explained 50% and 40% of maize yield variability. Maximum and minimum temperatures, respectively, explained 30% and 2% of maize yield variability. These findings are in agreement with a prior study by Huang et al. (2015), who indicated rainfall as a most important factor to explain variability in maize yields in the eastern United States.

In the present century (2010–2039) under RCP 8.5 and RCP 4.5, the inter-seasonal variation of maximum temperature has a correlation coefficient of r = −0.5, p = 0.012 with the inter-seasonal variations of maize yields. The bivariate regressions to analyze the dominant climate variable that determines, the inter-seasonal variation of maize yields in the Wami-Ruvu basin showed that the maximum temperature explained better the inter-seasonal variation of maize yields over Wami-Ruvu basin by 20%.

In mid century, under RCP 8.5, coefficient of determination revealed that rainfall will be the dominant factor to explain the variability in maize yield by 60%, whereas maximum temperature will be the dominant climate variable to explain the variability in maize yield by 34% under RCP 4.5. The correlation coefficient and the coefficient of determination between the inter-seasonal climate variables and maize yields in the end century under RCP 8.5 and RCP 4.5 revealed a significant strong positive relationship of r = 0.6, p = 0.0001 between rainfall and maize yields. Maximum temperature and solar radiation are expected to have a significant negative relationship with maize yield. The coefficient of determination indicates that rainfall will be the dominant factor to explain variability in maize yields in the end century under both RCP 4.5 and RCP 8.5.

The seasonal variation of ensemble average of the yields shows regular cycles with patterns of peaks and troughs of maize yields. However, from 2020 to 2026 the ensemble average of the yields predicts decline in maize yields due to increased maximum temperature that reduces the length of growing season. In 2043–2044, the ensemble average of the yields also predicts decline in maize yields due to an abrupt decrease in seasonal rainfall and an increase in seasonal temperatures and solar radiation. It is important to mention that the projected increase in seasonal temperatures and solar radiation and decreased seasonal rainfall will affect more the length of the growing season, which, in turn, affects maize yields.

It is important to note that in this study the management and farming practices do not vary with time. Therefore, more studies are recommended to test how the climate variability will affect maize yield under different management and farming practices. Furthermore, we recommend more studies that examine the possibility of reducing the projected impacts of climate variability on rainfed maize yields in the Wami-Ruvu basin.

REFERENCES

REFERENCES
Adamgbe
,
E. M.
&
Ujoh
,
F.
2013
Effect of variability in rainfall characteristics on maize yield in Gboko, Nigeria
.
Journal of Environmental Protection
4
,
881
887
.
http://dx.doi.org/10.4236/jep.2013.49103. Published online September 2013 (http://www.scirp.org/journal/jep
).
Agrawala
,
S.
,
Moehder
,
A.
,
Hemp
,
A.
,
van Aalst
,
M.
,
Hitz
,
S.
,
Smith
,
J.
,
Meena
,
H.
,
Mwakifamba
,
S.
,
Hyera
,
T.
&
Mwaipopo
,
O.
2003
Development and Climate Change in Tanzania: Focus on Mount Kilimanjaro
.
OECD
,
Paris
.
Akram
,
N.
2012
Is climate change hindering economic growth of Asian economies?
Asia-Pacific Development Journal
19
(
2
),
1
18
.
Amikuzuno
,
J.
&
Donkoh
,
S. A.
2012
Climate variability and yields of major staple food crops in Northern Ghana
.
African Crop Science Journal
20
(
Suppl. s2
),
349
360
.
Arndt
,
C.
,
Farmer
,
W.
,
Strzepek
,
K.
&
Thurlow
,
J.
2011
Climate Change, Agriculture, and Food Security in Tanzania
.
United Nations University, UNU-WIDER
,
Katajanokanlaituri, Helsinki, Finland
.
Case
,
M.
2006
Climate Change Impacts on East Africa
. .
Ehrhart
,
C.
&
Twena
,
M.
2006
Climate Change and Poverty in Tanzania: Realities and Response Options for CARE
.
Background Report
.
CARE International Poverty-Climate Change Initiative
. .
Graves
,
A. R.
,
Hess
,
T.
,
Matthews
,
R. B.
,
Stephens
,
W.
&
Middleton
,
T.
2002
Crop simulation models as tools in computer laboratory and classroom-based education
.
J. Natural Res. Life Sci. Educ.
31
,
48
58
.
Hartkamp
,
A. D.
,
De Beurs
,
K.
,
Stein
,
A.
&
White
,
J. W.
1999
Interpolation Techniques for Climate Variables
.
NRG-GIS Series 99-01
.
CIMMYT
,
Mexico, D.F
.
Hoogenboom
,
G.
,
Jones
,
J. W.
,
Wilkens
,
P. W.
,
Porter
,
C. H.
,
Boote
,
K. J.
,
Hunt
,
L. A.
,
Singh
,
U.
,
Lizaso
,
J. L.
,
White
,
J. W.
,
Uryasev
,
O.
,
Royce
,
F. S.
,
Ogoshi
,
R.
,
Gijsman
,
A. J.
&
Tsuji
,
G. Y.
2010
Decision Support System for Agrotechnology Transfer Version 4.5
[CD-ROM]
.
University of Hawaii
,
Honolulu
.
Huang
,
C.
,
Duiker
,
S. W.
,
Deng
,
L.
,
Fang
,
C.
&
Zeng
,
W.
2015
Influence of precipitation on maize yield in the Eastern United States
.
Sustainability
7
,
5996
6010
.
doi:10.3390/su7055996
.
IPCC
2007
Working Group II Fourth Assessment Report
.
Climate Change: Climate Change Impacts, Adaptation and Vulnerability
. .
IPCC
2012
Managing the risks of extreme events and disasters to advance climate change adaptation – summary for policy makers
. In:
A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change
(
Field
,
C. B.
,
Barros
,
V.
,
Stocker
,
T. F.
,
Qin
,
D.
,
Dokken
,
D. J.
,
Ebi
,
K. L.
,
Mastrandrea
,
M. D.
,
Mach
,
K. J.
,
Plattner
,
G.-K.
,
Allen
,
S. K.
,
Tignor
,
M.
&
Midgley
,
P. M.
, eds).
World Meteorological Organization
,
Geneva
,
Switzerland
,
24
pp.
IPCC
2013
Summary for policy makers
. In:
Climate Change (2013) The Physical Science Basis
(
Stocker
,
T. F.
,
Qin
,
D.
,
Plattner
,
G.-K.
,
Tignor
,
M.
,
Allen
,
S. K.
,
Boschung
,
J.
,
Nauels
,
A.
,
Xia
,
Y.
,
Bex
,
V.
&
Midgley
,
P. M.
, eds).
Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
.
Cambridge University Press
,
Cambridge
,
UK
&
New York
,
USA
.
Jones
,
J. W.
,
Hoogenboom
,
G.
,
Porter
,
C. H.
,
Boote
,
K. J.
,
Batchelor
,
W. D.
,
Hunt
,
L. A.
,
Wilkens
,
P. W.
,
Singh
,
U.
,
Gijsman
,
A. J.
&
Ritchie
,
J. T.
2003
The DSSAT cropping system model
.
European Journal of Agronomy
18
,
235
265
.
Kashaigili
,
J. J.
2011
Rapid Environmental Flow Assessment for the Ruvu River, A Consultancy Report Submitted to iWASH
. .
Khan
,
A. S.
,
Kumar
,
S.
,
Hussain
,
M. Z.
&
Kalra
,
N.
2009
Climate change, climate variability and Indian agriculture: Impacts vulnerability and adaptation strategies
. In:
Climate Change and Crops
(
Singh
,
S. N.
, ed.).
Springer-Verlag
,
Berlin
, pp.
19
38
. .
Klein
,
F.
,
Goosse
,
H.
,
Graham
,
N. E.
&
Verschuren
,
D.
2016
Comparison of simulated and reconstructed variations in East African hydroclimate over the last millennium
.
Climate of the Past
12
,
1499
1518
.
www.clim-past.net/12/1499/2016/doi:10.5194/cp-12-1499-2016
.
Kruger
,
A.
,
McBride
,
C.
,
Kijazi
,
A.
,
Chang'a
,
L.
,
Ng'ongolo
,
H.
,
Luhunga
,
P.
&
Malekela
,
C.
2012
[Southern Africa] State of Tanzania Climate in 2012 [in ‘State of the Climate in 2012’]
.
Bulletin of the American Meteorological Society
94
(
8
),
S168
S170
.
Lobell
,
D. B.
&
Field
,
C. B.
2007
Global scale climate–crop yield relationships and the impacts of recent warming
.
Environmental Research Letters
2
,
014002
.
Luhunga
,
P.
,
Chang'a
,
L.
&
Djolov
,
G.
2016
Assessment of the impacts of climate change on maize production in the Wami Ruvu basin of Tanzania
.
Journal of Water and Climate Change
7
(
2
),
142
164
.
doi:10.2166/wcc.2016.055
.
Ly
,
S.
,
Charles
,
C.
&
Degre
,
A.
2013
Geostatistical interpolation of daily rainfall at catchment scale: the use of several variogram models in the Ourthe and Ambleve catchments, Belgium
.
Hydrology and Earth System Sciences
15
,
2259
2274
.
www.hydrol-earth-syst-sci.net/15/2259/2011/doi:10.5194/hess-15-2259-2011
.
Masson
,
D.
&
Knutti
,
R.
2011
Climate model genealogy
.
Geophysical Research Letters
38
,
L08703
.
doi:10.1029/2011GL046864
.
Matsui
,
N.
2016
11-year rainfall and maize yield variation (2003–2013) in four northern districts of Malawi
.
Journal of Agricultural Science
8
(
5
).
Mbwanga
,
M.
2005
On Farm Verification and Promotion of Green Manure for Enhancing Upland Rice Productivity on Striga Infested Fields in Tanzania
. .
Miglietta
,
F.
&
Bindi
,
M.
1993
Crop growth simulation models for research, farm management and agrometeorology EARSeL
.
Advanced Remote Sensing
2
,
148
157
.
Milder
,
J. C.
,
Buck
,
E. L.
,
Hart
,
K. A.
&
Scherr
,
J. S.
2012
A Green Growth Investment Framework for SAGCOT, the SAGCOT Greenprint
. .
Morton
,
J. F.
2007
The impact of climate change on smallholder and subsistence agriculture
.
Proceedings of the National Academy of Sciences of the United States of America
104
,
19680
19685
.
Msongaleli
,
B. M
,
Rwehumbiza
,
F.
,
Tumbo
,
S. D.
&
Kihupi
,
N.
2015
Impacts of Climate Variability and Change on Rainfed Sorghum and Maize: Implications for Food Security Policy in Tanzania
.
Journal of Agricultural Science
7
(
5
),
124
142
. doi:10.5539/jas.v7n5p124.
Mubaya
,
C. P.
,
Jalloh
,
A.
&
Mogaka
,
H.
2014
Review of Research and Policies for Climate Change Adaptation in Urban Areas in East Africa
. .
Müller
,
C.
,
Cramer
,
W.
,
Hare
,
W. L.
&
Lotze-Campen
,
H.
2011
Climate change risks for African agriculture
.
Proceedings of the National Academy of Sciences of the United States of America
108
,
4313
4315
.
Mwandosya
,
M. J.
,
Nyenzi
,
B. S.
&
Luhanga
,
M. L.
1998
The Assessment of Vulnerability and Adaptation to Climate Change Impacts in Tanzania
.
Centre for Energy, Environment, Science and Technology (CEEST)
,
Dar-es-Salaam
,
Tanzania
.
Ngana
,
J.
,
Mahay
,
F.
&
Cross
,
K.
2010
Wami Basin – A Situation Analysis
.
Report for the Wami/Ruvu basin Water Office
.
IUCN Eastern and Southern Africa Programme
.
Xviii + 92 pp
.
Nikulin
,
G.
,
Jones
,
C.
,
Giorgi
,
F.
,
Asrar
,
G.
,
Büchner
,
M.
,
Cerezo-Mota
,
R.
,
Christensen
,
O. B.
,
Déqué
,
M.
,
Fernandez
,
J.
,
Hänsler
,
A.
,
van Meijgaard
,
E.
,
Samuelsson
,
P.
,
Sylla
,
M. B.
&
Sushama
,
L.
2012
Precipitation climatology in an ensemble of CORDEX-Africa regional climate simulations
.
Journal of Climate
25
,
6057
6078
.
doi:10.1175/JCLI-D-11-00375.1
.
Omoyo
,
N. N.
,
Wakhungu
,
J.
&
Oteng'I
,
S.
2015
Effects of climate variability on maize yield in the arid and semi arid lands of lower eastern Kenya
.
Agriculture & Food Security
4
(
8
).
doi: 10.1186/s40066-015-0028-2
.
Ramirez-Villegas
,
J.
&
Challinor
,
A.
2012
Assessing relevant climate data for agricultural applications
.
Agricultural and Forest Meteorology
161
,
26
45
.
doi:10.1016/j.agrformet.2012.03.015
.
Rowhani
,
P.
,
Lobell
,
D. B.
,
Linderman
,
M.
&
Ramankutty
,
N.
2011
Climate variabilty and crop production in Tanzania
.
Agricultural and Forest Meteorology
151
,
449
460
.
Saxton
,
K. E.
&
Rawls
,
W. J.
2009
Soil Water Properties Calculator (Version 6.02)
. .
Shemsanga
,
C.
,
Omambia
,
A. N.
&
Gu
,
Y.
2010
The cost of climate change in Tanzania: impacts and adaptations
.
Journal of American Science
6
(
3
),
182
196
.
The National Bureau of Statistics
2012
National Panel Survey 2009–2010
. .
Thornton
,
P. K.
,
Jones
,
P. G.
,
Owiyo
,
T.
,
Kruska
,
R. L.
,
Herrero
,
M.
,
Orindi
,
V.
,
Bhadwal
,
S.
,
Kristjanson
,
P.
,
Notenbaert
,
A.
,
Bekele
,
N.
&
Omolo
,
A.
2009
Climate change and poverty in Africa: mapping hotspots of vulnerability
.
African Journal of Agricultural and Resource Economics
2
(
1
),
24
44
.
Thornton
,
P. K.
,
Jones
,
P. G.
,
Alagarswamy
,
G.
&
Andresen
,
J.
2010
Spatial variation of crop yield response to climate change in East Africa
.
Global Environmental Change
19
,
54
65
.
doi: 10.1016/j.gloenvcha.2008.08.005
.
Tumbo
,
S. D.
,
Mpeta
,
E.
,
Tadross
,
M.
,
Kahimba
,
F. C.
,
Mbillinyi
,
B. P.
&
Mahoo
,
H. F.
2010
Application of self-organizing maps technique in downscaling GCMs climate change projections for same, Tanzania
.
Journal of Physics and Chemistry of the Earth
35
,
608
617
.
URT
2013
Status of Gauging Station Wami/Ruvu Basin – DPG Tan, Annual Basin Hydrological Report
.
Wami/Ruvu Basin Water Board
,
Morogoro
,
Tanzania
.
Van Vuuren
,
D. P.
,
Edmonds
,
J.
,
Kainuma
,
M.
,
Riahi
,
K.
,
Thomson
,
A.
,
Hibbard
,
K.
,
Hurtt
,
G. C.
,
Kram
,
T.
,
Krey
,
V.
,
Lamarque
,
J.-F.
,
Matsui
,
T.
,
Meinshausen
,
M.
,
Nakicenovic
,
N.
,
Smith
,
S. J.
&
Rose
,
S. K.
2011
Representative concentration pathways: an overview
.
Climatic Change
109
(
5
).
doi:10.1007/s10584-011-0148-z
.
van Wart
,
J.
,
van Bussel
,
L. G. J.
,
Wolf
,
J.
,
Licker
,
R.
,
Grassini
,
P.
,
Nelson
,
A. D.
&
Sassman
,
K. G.
2013
Use of agro-climatic zones to upscale simulated crop yield potential
.
Field Crops Research
143
,
44
55
.
doi:10.1016/j.fcr.2012.11.023
.
Washington
,
R.
&
Pearce
,
H.
2012
Climate Change in East African Agriculture: Recent Trends, Current Projections, Crop-Climate Suitability, and Prospects for Improved Climate Model Information
.
CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)
,
Copenhagen
,
Denmark
. .