Food Security

, Volume 10, Issue 4, pp 897–910 | Cite as

Is dry soil planting an adaptation strategy for maize cultivation in semi-arid Tanzania?

  • Marcos A. LanaEmail author
  • Ana Carolina F. Vasconcelos
  • Christoph Gornott
  • Angela Schaffert
  • Michelle Bonatti
  • Johanna Volk
  • Frieder Graef
  • Kurt Christian Kersebaum
  • Stefan Sieber
Open Access
Original Paper


Agriculture has the greatest potential to lift the African continent out of poverty and alleviate hunger. Among the countries in sub-Saharan Africa, Tanzania has an abundance of natural resources and major agricultural potential. However, one of the most important constraints facing Tanzania’s agricultural sector is the dependence on unreliable and irregular weather, including rainfall. A strategy to cope with climate uncertainty in semi-arid regions is to proceed with the sowing of the crop before the onset of the rainy season. The advantage is that when the rains start, seeds are already in the soil and can begin immediately the process of germination. The objective of this paper was to assess the effectiveness of dry-soil planting for maize as an adaptation strategy in the context of a changing climate in Dodoma, a semi-arid region in Tanzania. For this assessment, the DSSAT crop model was used in combination with climate scenarios based on representative concentration pathways. A probability of crop failure of more than 80% can be expected when sowing occurs during the planting window (of 21 days) starting on 1st November. The next planting window we assessed, starting on 23rd November (which was still before the onset of rain), presented significantly lower probabilities of crop failure, indicating that sowing before the onset of the rainy season is a suitable adaptation strategy. Results also indicated that, despite not reaching the highest maize grain yields, fields prepared for dry-soil planting still produced adequate yields. The cultivation of several fields using the dry planting method is a strategy farmers can use to cope with low rainfall conditions, since it increases the chances of harvesting at least some of the cultivated fields. We conclude that dry-soil planting is a feasible and valid technique, even in scenarios of climate change, in order to provide acceptable maize yields in semi-arid Tanzania.


DSSAT Sub-Saharan region Maize yield Seed germination Sowing date Food security 

1 Introduction

Agriculture has the greatest potential to lift Africa out of poverty and to alleviate hunger. Given the nature of African agriculture, where a large proportion of farmers are smallholders and subsistence-based, it is essential to invest in and develop accessibility to quality inputs, markets for produce, good soils and soil management techniques, innovative finance tools and other resources that are needed for sustained agricultural production (Graef et al. 2017).

The agricultural sector in Tanzania is hampered by low productivity, poor infrastructure, and a lack of technology, being dominated by smallholder farmers cultivating average farm sizes of between 0.9 ha and 3.0 ha. About 70% of Tanzania’s crop area is cultivated by hand hoe, 20% by ox plough and 10% by tractor, and almost all is rain fed. The production of food crops dominates the agricultural economy (KPMG 2015).

This background was confirmed in an extensive survey carried out in Tanzania by the Trans-SEC Project (Innovating Strategies to Safeguard Food Security using Technology and Knowledge Transfer: A People-Centred Approach) (Below et al. 2015; Graef et al. 2014, 2017; Löhr et al. 2016; Mutabazi et al. 2015). The major constraints facing the Tanzanian agricultural sector are the reduction in the available labor force, low land productivity due to the application of poor technologies, and dependence on unreliable and irregular weather (Mmbaga and Lyamchai 2002). Both crops and livestock are adversely affected by periodic droughts. Irrigation holds the key to stabilizing agricultural production in Tanzania, to improving food security, increasing the productivity and incomes of farmers, and producing higher valued crops such as vegetables and flowers (Evans et al. 2012). However, some other barriers to improve local food production and security are not directly tied to a lack of irrigation, but are related to the availability of farm credit, access to seeds and fertilizers, quality of seeds, farm implements and market reliability (Mdemu et al. 2017).

One of the major uncertainties with the future trajectory of agricultural productivity in Africa is the likely impact of a changing climate (IPCC et al. 2007). Several studies (Kurukulasuriya et al. 2006; Lesch et al. 2005; Nelson et al. 2010; Seo et al. 2009; Thornton et al. 2006) provide strong evidence that predicted changes in temperature and rainfall caused by global warming may impose additional serious constraints on agriculture in Africa (Benin et al. 2016). Climate change will have a direct impact on the availability of water for irrigated crops. In addition to changes with precipitation, climate-change-induced higher temperatures increase the water requirements of crops (Nelson et al. 2009).

Climate change is likely to make matters worse with further increases in rainfall variability being predicted (Agrawala et al. 2003; Schlenker and Lobell 2010). Climate change is also expected to alter pest and disease outbreaks, increase the frequency and severity of droughts and floods, and increase the likelihood of poor yields, crop failure and livestock mortality (Harvey et al. 2014). The ability of agricultural communities and other agricultural stakeholders in sub-Saharan Africa to cope with the constraints and opportunities of current climate variability must first be enhanced in order to be able to adapt to climate change and the predicted future increase in climatic variability. Tools and approaches are now available that allow for a better understanding, characterization and mapping of the agricultural implications of climate variability and the development of climate risk management strategies specifically tailored to the needs of stakeholders (Cooper et al. 2008).

Specific concerns about climate variability include variability in the onset and cessation of rainfall, rainfall amount, and frequency and duration of periods of soil water deficits. This variability greatly affects crop yield in rainfed systems, and is a major disincentive to the adoption of yield-improving practices, which then challenges researchers to develop adoptable practices and varieties. Therefore, knowledge of the seasonal climate variability and its associated risks is important to the improvement of crop management (Liben et al. 2015).

The shifting of planting date based on cultivar properties could prevent critical stages of crop development from coinciding with periods of extreme high temperature and water deficit, consequently helping to reduce yield loss from climate change (Tao and Zhang 2010). Generally, crop yields may suffer substantially with either a late onset or early cessation of the growing season. Shifting of planting date ensures that the seed, once in the soil, has sufficient moisture to trigger the germination process and further crop development. However, in order to be successful and effective this strategy needs the farmers to be available all the time before the expected onset of rainfall (Mugalavai et al. 2008).

Under the conditions observed in rural communities of Tanzania, especially in the semi-arid regions, farmers usually farm several field areas, sometimes many kilometers apart; the principle behind this is to take advantage of better soils and reduce the risk of total crop failure associated with the small-scale spatial distribution of precipitation. To be able to use fields in different areas, waiting for the onset of rainfall is not an adequate strategy. Therefore, farmers often practice so-called ‘dry soil planting’, when they seed the crop into dry soil a few days before the expected start of the rainy season (KPMG 2015). The advantage of this practice is that when the rainfall starts seeds are already in the soil and can immediately start to germinate. This is especially important in regions where the rainy season is not long enough to provide adequate moisture during the whole cropping cycle.

As an example, according to Liwenga (2008), in Mvumi, Tanzania, planting activities are normally carried out some two to three weeks before the expected time for the onset of the rains; in that study villagers explained that early planting is done prior to the onset of the rains so that crops could benefit from all of the moisture provided during the short rainy season. The same author, citing Holtland (1994), explains that seedlings make optimal use of the natural nitrogen (N)-flush that occurs after the first rain-showers. As N availability is, after water, the greatest constraint to cultivation in the Dodoma region, the difference between dry planting and planting a week after the onset of the rains can be considerable.

Nevertheless, despite the advantage of a faster start, planting into dry soil also poses the risk that germination is initiated by a precipitation event that is not the start of a rainy season. In this situation, the crop can start to germinate but then die during subsequent drying of the soil and seedling. In other situations there is also a risk that if the seed stays for a long period in the soil without sufficient moisture to trigger germination, high temperatures can cause loss of vigor, or it can be damaged or eaten by insects or other animals (Benin et al. 2016; Cooper et al. 2008).

In addition to planting dry seeds, seeds can be ‘primed’ with water, an enhancement method that might improve seed performance under stress conditions such as drought or when freshly harvested or aged seeds are used which might fail to germinate (Lutts et al. 2016). Harris et al. (1999) demonstrated how simple soaking seeds in water before sowing can increase the speed of germination and emergence, leading to better crop stands, and allow seedlings to grow much more vigorously. Some farmers in Zimbabwe already have experience of soaking maize and sorghum seeds in water before sowing, in an attempt to improve establishment, but the practice appears to be neither widespread nor regularly followed, probably because farmers need the opportunity to experiment for themselves; to do their own research and development (Harris et al. 2001).

Maize is the staple food for the majority of Tanzanians, providing about 60% of Tanzanian’s dietary calories and 50% of the protein. With about five million ha, Tanzania has the largest planted area of maize in all Southern and East Africa. Maize production has significantly increased over the past 10 years, largely through expansion of area planted rather than increased grain yields. Over the past 50 years, maize production has kept pace with the increase in population (FAOSTAT 2016).

Most maize (80%) is produced by small-scale farmers and is grown both for subsistence and as a cash crop. Between 65% and 80% of all maize is consumed within the producing households: only 20% to 35% enters commercial channels. Maize comprises an average of 16% of national household food expenditures, though there are big regional variations. The choice to grow maize, even in areas of insufficient rainfall, is driven by a strong dietary preference for maize over the more drought-adapted traditional cereals such as sorghum and millets. Efforts are being made to develop more drought tolerant cultivars of maize available to Tanzanian farmers and to increase irrigation (Doebley et al. 2010).

Most maize production in Tanzania is under low-input rainfed conditions. Simple hand hoes, farm-recycled seed, little use of synthetic fertilizers or agrochemicals and minimal weeding are the usual technologies and inputs employed. Maize crops frequently fail because of insufficient soil moisture. Irrigation is not usually available or selected as an option for maize, and on-farm water harvesting techniques are not yet well known or used (Wilson and Lewis 2015).

Drought is a major threat to maize in many parts of Tanzania. Maize production can be a risky and unreliable business because of erratic rainfall and the high susceptibility of maize to drought, while the performance of local drought-tolerant cultivars is poor (Mmbaga and Lyamchai 2002). Erratic rainfall is making maize farmers in Tanzania vulnerable to low yields, which translates to food insecurity. Tanzania suffered the effects of a prolonged drought in the recent years (Doebley et al. 2010). These observations agree with the information gathered from farmers in the region during a comprehensive survey about many aspects of food security in the region (see Trans-SEC: Innovating Strategies to Safeguard Food Security using Technology and Knowledge Transfer: A People-Centred Approach -

In order to assess and understand the effects of climate change, crop models can be a useful tool to assess the influence of climatic and other environmental or management factors on crop development and yield (Reidsma et al. 2010). The Decision Support System for Agrotechnology Transfer – DSSAT v. 4.5 contains the Crop System Model CERES – Maize model (Jones et al. 2003) and can help to a) determine best planting dates, b) define fertilizer timing and amounts, c) support precision agriculture and d) detect/investigate potential impacts of climate change on agriculture. In the embedded CERES – Maize model, the development and growth of the crop is simulated on a daily basis from planting until physiological maturity. The model calculations are based on environmental and physiological processes that control the phenology and dry matter accumulation in different organs of the plant. The DSSAT also has other embedded models that can simulate the flow of nutrients and water balance in the soil. Despite the complex array of processes simulated by the crop model, some important processes such as effects of pests and diseases are still not well depicted (Donatelli et al. 2017), as are the impact of extreme events such as floods and hail (Alexandrov and Hoogenboom 2000) or high temperatures (above 35 °C) during anthesis, which reduces pollen viability (Porter and Semenov 2005).

Considering that smallholder farmers in Tanzania cultivate several fields at the same time, of which only some may be under a dry-soil planting management system, it can be expected that dry-soil planting is and will still be part of the strategies to reduce the risk of crop losses and help ensure food security in the context of a changing climate. Based on that, the objective of this paper was to assess the effectiveness of dry-soil maize planting as an adaptation strategy in a semi-arid region of Tanzania. This assessment was done using a calibrated CERES – Maize crop model and tested different climate scenarios for the 2020–2059 and 2060–2099 periods.

2 Methodology

2.1 Location

Our assessment was done for the Dodoma region of Tanzania, as part of the activities of the Trans-SEC Project (see Graef et al. 2017). The study region lies between latitudes 5°50′S to 6°10′S and between longitudes 35°40′E to 36°05′E, at an elevation of 1020 m above sea level. The dominant soil of the target region is defined as a ferralic Cambisol (FAO), which is low in fertility and seasonally waterlogged or flooded (Msongaleli et al. 2015). Detailed information about the soil used in this assessment is presented in Table 1. The area is characterized by low and erratic rainfall with a unimodal rainfall regime. The long-term mean annual rainfall is about 511 mm with average temperatures of 22.7 °C (Fig. 1). The onset of rainfall usually occurs in early-mid December, and the rainy season extends until April. As the rainfall pattern in the study region is variable within the rainy season, maize cultivation is restricted to areas where water availability is higher and the soil has a higher water holding capacity.
Table 1

Soil profile parameters of the study region in Dodoma, Tanzania, indicating physical and chemical parameters for different soil depths

Layer depth



Organic carbon

pH in water

Cation exchange capacity

Total nitrogen

Lower limit (LL)

Drained upper (DUL)

Saturation (SAT)







































































Fig. 1

Averages of (a) maximal and minimal monthly temperatures (°C) and (b) monthly precipitation (mm) for the study site in Dodoma, Tanzania, during the crop season (data refer to the 2000–2015 period). Bars indicate the monthly standard deviation of each parameter

2.2 Climate scenarios

Data for future climate scenarios are available from the Inter-Sectoral Impact Model Intercomparison Project – ISI-MIP (Warszawski et al. 2014) as a bias-corrected dataset with daily values for temperature (Tmax, Tavg, Tmin), precipitation, relative humidity, and solar radiation. These data are well accepted and also used for the Agricultural Model Intercomparison and Improvement Project – AgMIP (Rosenzweig et al. 2013). The selected Representative Concentration Pathways (RCPs) for our study were the high-emission scenario RCP8.5, the medium-low-emission scenario RCP4.5 and the low-emission scenario RCP2.6 scenario (Moss et al. 2010). For each RCP scenario, five different projections (model-derived estimates of future climate) were used, as follows: GFDL-ESM2M (Dunne et al. 2013), HadGEM2-ES (Johns et al. 2006), IPSL-CM5A-LR (Hourdin et al. 2013), MIROC-ESM.CHEM (Watanabe et al. 2011), and NorESM1-M (Iversen et al. 2012), totaling 15 different projections of three RCPs. All the models have the added capability of explicitly representing biogeochemical processes that interact with physical-climate processes. To capture the fertilization effect of increasing levels of atmospheric CO2 (Long et al. 2006; Tubiello et al. 2007), the concentration of CO2 was adjusted yearly according to the respective RCP (Meinshausen et al. 2011).

2.3 Crop modeling

To generate projections of maize yield and assess the impact of weather scenarios and dry-seed planting management, a crop simulation model was employed. The crop modeling process was done using DSSAT (Jones et al. 2003), an established crop model already used in several impact assessment studies (Rivington and Koo 2011).

The maize cultivar used in this study was a locally adapted, open pollinated cultivar called Situka, with a yield potential at around 4 t ha−1 in Dodoma (according to reports from local technicians), and commonly cultivated by local farmers because of its early maturity and tolerance to low levels of soil nitrogen (N), a typical condition of the study area (Msongaleli et al. 2015). This cultivar was already calibrated and validated for DSSAT (Mourice et al. 2014a, b) for Morogoro, situated 250 km east from Dodoma, where it can reach yields of 6.6 t ha−1. However, for this study, a further validation step was done using three years of field data including soil, weather, agronomic management and yield (Kimaro et al. 2008, 2009) from research experiments conducted in nearby villages to confirm the ability of the model to simulate past observations in the study region. Difference-based indices such as mean bias error (MBE) (Addiscott and Whitmore 1987), mean percentage error (MPE), root mean squared error (RMSE) (Fox 1981) and relative root mean squared error (RRMSE) (Jørgensen et al. 1986) were used to evaluate the simulation outputs. The final maize plant population density was defined at 33000 plants per ha, planted at 7 cm depth in rows with 75 cm spacing between rows. This low plant population density, the space between rows and planting depth represent the common practices in the region. Fertilizer use was set to 2000 kg ha−1 of dry cattle manure, equivalent to 40 kg ha−1 N, to be applied on the day of planting. No irrigation was applied. DSSAT was instructed to start the simulations on 1 January each year to provide a more realistic soil water balance at sowing. For planting dates taking place after 1 January, the simulation started on 1 January of the previous year. Harvest was set to take place two weeks after physiological maturity, as calculated by the model. The format of climate scenario files was adjusted for DSSAT structure using the Weatherman software (Wilkens et al. 2004). Simulations were run for each projection and RCP. For the analysis of results, yields generated with different projections were merged in ensembles (according RCP) in order to provide more robust information (Randall and Wood 2007).

2.4 Adaptation strategies

Initially, in order to establish a yield baseline, simulations using all the RCPs were run for the 2000–2015 period using the different planting windows (described in the next paragraph), starting on 1st of November and at every 21 days, until 1st of February, totaling seven planting windows. The objective was to verify if the simulations could mimic the best planting window (mid-December) and crop management as informed by the local farmers and technicians involved in the Trans-SEC project. Yields from the same planting window were averaged to identify the one with the highest yields (best planting date). Once the best planting window was defined for the baseline period, the next step was to run simulations using this planting window for the 2020–2059 and the 2060–2099 periods.

In order to mimic the actual practice in the region of dry soil planting (that means that farmers plant the seeds into dry soil and wait for rainfall onset within 21 days), five planting windows (each of 21 days) were defined, starting on 1 November. The assumptions were that 1) in the dry soil planting systems the seed would remain viable for 21 days and initiate the germination process only if the soil moisture of the top 20 cm reached at least 30% of the soil field capacity, and that 2) based on soil texture and soil bulk density the soil module embedded in the crop model was able to correctly simulate the soil moisture. Although maize seed can absorb moisture from the soil even when soil moisture conditions are far below permanent wilting point (Muthukuda Arachchi et al. 1999), the minimum external water potential permitting maize germination observed by Hunter and Erickson (1952) was shown to be −1.25 MPa. The same authors reported that maize seed started to germinate only when seed moisture exceeded the critical value of 30.5% of field capacity. If during the planting window the critical soil moisture was not reached, it is assumed that the seeds wereno longer in a condition to germinate, and the yield was set to zero. If germination starts, but a dry spell kills the seedling (crop abortion) or the crop produces less than 400 kg ha−1 of grain, then the crop season was considered lost (crop failure).

3 Results

3.1 Model and cultivar validation

As the crop model and the Situka cultivar were already validated for a different region of Tanzania (Mourice et al. 2014a, b), but not for a nearby location, a second validation was considered necessary. Simulations using the same conditions (real data for weather, soil, agronomic management) of the field experiments done by Kimaro et al. (2008, 2009) including a nearby region, were done. The results of the validation were satisfactory, indicating that the model could mimic maize grain yields from field observations of three available crop seasons (2004–2006), as presented in Table 2. As the MBE term describes the direction of the error bias, the positive value of 160 kg ha−1 indicates that the predictions were overestimated (in absolute values) compared to the observations. If all simulated and observed values are the same, then MBE, MPE, RMSE and RRMSE should be equal to zero. Overall, the results certified that the model could adequately mimic field observations for maize yield. Nevertheless, it is important to emphasize that the experiments used to generate these observations were village-level research experiments, and not data gathered from maize crops on smallholder farm fields, where yields are probably often lower. Data related to phenology and above ground biomass were not available for comparison in this exercise.
Table 2

Results of the validation process for the DSSAT crop simulation model using difference-based indices comparing simulated and field observations of grain yields from maize cv. Situka in three different seasons (2004–2006) in Tanzania. MBE: mean bias error; MPE: mean percentage error; RMSE: root mean squared error; RRMSE: relative root mean squared error


Yield (Kg.ha−1)
























MBE, Mean Bias Error; MPE, Mean percentage error; RMSE, Root Mean Squared Error; RRMSE, Relative Root Mean Squared Error

3.2 Definition of best actual planting dates

Simulations run for the 2000–2015 period using the data for the RCPs identified the planting windows beginning on the dates 12 December and 6 January as the ones with the highest grain yields (Fig. 2). This coincided with the onset of the rainy season as reported by the local farmers from the study sites (Fig. 1).
Fig. 2

Assessment of best maize planting dates according to grain yield using the Ceres maize model for the 2000–2015 period (referred to as baseline) in the study site in Dodoma, Tanzania. Bars represent the standard deviation of yield

3.3 Climate change scenarios

For Tanzania, and in particular for the Dodoma region, the RCP’s showed slight changes in the precipitation amount during the crop season (Fig. 3). No significant shift in the onset of the rainy season was identified for the study region. For the 2020–2059 period, the scenarios indicated a stability or reduction in accumulated precipitation until February–March (using the 2000–2015 period as baseline). For the rest of the crop season, the RCPs tended to deviate in terms of accumulated precipitation. For the second period (2060–2099), the models projected a slight reduction in accumulated rainfall at the beginning of the cropping season, with later increments in precipitation in the 4.5 and 8.5 RCPs. In contrast to projections for other regions of the world, the precipitation pattern in Tanzania should not change its timing, especially for the first period (2020–2059). In terms of the amount of rainfall, the 4.5 and 8.5 RCPs indicated an increase in precipitation volume for the later period (2060–2099), when compared to the baseline (Fig. 3b), while the RCP 2.6 indicated a reduction in accumulated precipitation.
Fig. 3

Modelled accumulated precipitation along the cropping season in Dodoma, Tanzania represented by the 2000–2015 baseline (solid line) compared to different representative concentration pathways (RCPs) (dotted and dashed lines) for (a) 2020–2059 and (b) 2060–2099. Stacked bars represent the monthly accumulated precipitation departure from the baseline according different RCPs

For air temperature (Fig. 4), all RCP’s indicated an increment for Tmax and Tmin. For the first time slice, the three RCPs showed relatively similar increments in both Tmax and Tmin. For the 2060–2099 period, there was a clear distinction between the different RCPs, with the highest increase observed for the RCP 8.5, followed by the RCP 4.5 and finally the RCP 2.6.
Fig. 4

Maximal (Tmax) and minimal (Tmin) temperature changes in Dodoma, Tanzania according different representative concentration pathways (RCPs) for two time periods (2020–2059 and 2060–2099). The dashed line represents the baseline relating to the 2000–2015 period

3.4 Impact assessment with the use of dry soil planting

When using dry-soil planting as a strategy to make possible the cultivation of several different fields per farm, we assessed that the very early dry soil planting window (starting date 1 November) for both time periods (2020–2059 and 2060–2099) presented more than an 80% probability of failure for all RCPs, while the next planting windows reduced the probability of failure to less than 20% (Fig. 5a and b). In the best situation, when planting occurs near mid-December, the chances of losses from crop failure are very low. The dry soil planting practice in the selected region can only be said to occur when done before mid-December. After this planting window, with the very likely onset of the rainy season, sowing will take place into already wet soil.
Fig. 5

Frequency of maize crop failure (defined as grain yields lower than 400 kg ha−1) derived from all representative concentration pathways (RCPs) according different planting windows (each of 21 days) using dry soil planting as a strategy in Dodoma, Tanzania. Values for the 2020–2059 period are on the left (a), while the 2060–2099 period are on the right (b). The date on the x axis indicates the start of the planting window

Dry-seed planting is said to occur when the soil, at the time of sowing, has insufficient moisture to trigger germination, which as seen in Fig. 6a and b, are planting windows starting on 1st of November and 23rd November. In many years, the third planting window also does not meet the minimal soil moisture levels (30.5% of field capacity) at the beginning, as seen by the high dispersion of symbols in Fig. 6 indicating the start of the germination process. By advancing the planting window, the start of germination tends to concentrate on or very close to the beginning of each planting window, because of sufficient soil moisture to trigger germination. Especially for the two last planting windows, in both time periods, we observed that for almost all RCPs and years, the soil already had enough moisture for immediate seed imbibition and the start of germination. This higher soil moisture availability at the planting time, however, does not automatically ensure good crop performance, as yield will also be defined by climatic events that occur well after germination.
Fig. 6

Maize grain yields simulated by the DSSAT model using different scenario periods and different planting windows (dates in the X axis indicate the start of the planting window) using the dry soil planting adaptation strategy for Dodoma, Tanzania; (a) 2020–2059; (b) 2060–2099. The different representative concentration pathways (RCP’s) are represented by ○ (RCP 2.6), ● (RCP 4.5) and ▼ (RCP 8.5) and indicate when the critical threshold of soil moisture for seed imbibition was reached

For maize grain yield, the most successful planting windows for both time slices were between 23 November and 6 January, as seen in Fig. 7. As yield is defined by several parameters, it is assumed that the above mentioned planting windows were the ones combining the best emergence conditions for the crop (soil moisture) as well as adequate water supply during the crop’s growth cycle. Late planting date windows, despite the high values of initial soil moisture that allow for immediate germination, failed to provide an adequate amount of rainfall during the vegetative stages and especially during the reproductive stages of maize development, which occur towards the end of the rainy season. As seen in Fig. 3, precipitation ceases in April, when the plants that germinated during the latest planting times are expected to be in the grain filling phase. Comparing the baseline yields (for 2000–2015) and projected yields for 2020–2099 from both time periods (slices) (Fig. 7), no large changes in yield were expected to occur under both time slices.
Fig. 7

Maize grain yields generated for different planting windows and representative concentration pathways (RCPs) in two time periods (slices): 2020–2059 (a) and 2060–2099 (b) for Dodoma region, Tanzania. The baseline yield (2000–2015) is represented by the white circle symbol for each planting window in both time slices. Bars indicate the standard deviation

4 Discussion

In cropping systems reliant on rainfall as the sole source of moisture for crop growth, seasonal rainfall variability is inevitably mirrored in both highly variable crop production levels and in the risk-averse livelihood and coping strategies that have emerged over time in smallholder farming communities.

For this assessment, the Ceres crop model evaluation using the Situka maize cultivar was considered adequate since values of MPE and RRMSE are close to zero. This shows that, when input parameters are known, the model can properly mimic field observations such as the ones of Kimaro et al. (2008, 2009) used in this exercise for model evaluation. The RRMSE below 10% can be considered excellent, according Jamieson et al. (1991). The RMSE (204 kg ha−1) provides the square root of the average of squared differences between prediction and actual observations, while the RRMSE indicates as a percentage how heterogeneous the simulation results are in relation to the observations. The difference between the MPE and the RRMSE (1.7%) shows that, by using the RMSE as the base component, errors are magnified. Also a positive MBE is desirable, indicating that the model over-predicted yields, according Willmott (1982). Another aspect that supports the ability of the crop model to mimic field observations is that the simulations using weather scenarios for the period 2000–2015 matched the best planting dates reported by farmers in the study region.

For temperatures in East Africa, models project increments in surface air temperature over land between the end of the century (2070–2099) and the present day (1980–2010). Similarly, for annual precipitation, all models indicate an increase of 5% to 20% in precipitation (Warszawski et al. 2014). These different future climate scenarios allowed the ‘mapping’ of the uncertainty of impact assessments on crop yields and have already been used in several other assessments as listed by Müller et al. (2010), but with different levels of precision.

Despite the significant increases in temperature predicted for the later time period (slice) (2060–2099) (Fig. 4), these deleterious effects could be neutralized or partially reduced by the increment in precipitation (Fig. 3) observed in the middle of the cropping season. An increment in temperature will also accelerate the development of the crop, which may help to avoid the occurrence of the grain filling phase after the end of the rainy season. Additionally, the viability of maize pollen decreases with exposure to temperatures above 35 °C (Dupuis and Dumas 1990), and this effect can be enhanced under high vapor pressure deficits (Hatfield and Prueger 2015). It is important to note that there is a differentiation between the RCPs only in the 2060–2099 period (Fig. 7b), when RCP 4.5 projected the highest yields. For this RCP, the significant increment in temperature was neutralized by the increment in accumulated precipitation.

Data from a field survey carried out by Mourice et al. (2014a, b) for the Wami Basin (which includes the study region) shows that reported yields of maize on smallholder farms range from 50 kg ha−1 to 3600 kg h−1, with an average of only 860 kg ha−1. The discrepancy observed between the yields from Table 2 (which originated from field experiment observations and simulations) and the ones reported above can be attribute to many factors, ranging from agronomic management (sowing date, plant population, fertilizer use, cultivar, irrigation), biophysical factors (soil type, soil compaction, fertility, water or temperature stress, attack of pests, weeds, diseases) to other problems such as animal grazing, theft and lodging. Specifically with regard to crop models an important aspect to be considered is that under field conditions crops will often be subjected to stressors that are not included or not simulated in the model (e.g. pests, diseases, and many others). As stated by Donatelli et al. (2017), this capability to fully model the farm situation is still missing in many crop models, although developments in recent decades are moving towards the quantitative description of the impact of pests and diseases on yield. Differences in management (cultivar, plant population, planting date and soil and nutrient management), soil and weather also contribute to increase this gap. Luhunga et al. (2017), using a fixed CO2 concentration of 360 ppm, also assessed the impacts of climate change scenarios on maize production in the Wami river basin of Tanzania using a single planting window of 15 days (1–15 December), reaching slightly lower baseline yields for maize. Both studies used the same maize cultivar (Situka), but without conducting a further calibration to verify the plausibility of the model for other regions, as recommended by Grassini et al. (2015). Since the region where the crop model was applied does have diverse environmental conditions, it is important to test, adjust, and validate model parameters as necessary.

Regarding dry-soil planting as an adaptation strategy, Akinnagbe and Irohibe (2014) reported that in Tanzania, to avoid crop production risks due to rainfall variability and drought, staggered planting into dry soil before the onset of rains is very common for many farmers. The cultivation of several fields using the dry planting method is therefore regarded as a special feature of adapting to low rainfall conditions since it increases the chances of harvesting at least some of the cultivated fields.

Another aspect that favors dry-soil planting as an ex-ante drought risk adaptation strategy to help ensure food security is to shift and extend the cultivation area to different fields (Westengen and Brysting 2014). The spatial diversification of fields is an effective ex-ante drought risk coping strategy that reduces exposure to risk (Shiferaw et al. 2014). Under such conditions, the risk of income shortfall is reduced by growing maize crops on several fields, and reduced even more when different crops are cultivated. According to Pandey et al. (2007), this principle is used in different types of diversification common in rural societies. Examples include the spatial diversification of farms, diversification of agricultural enterprises and diversification from farm to non-farm activities.

Although ex-ante strategies can be costly in terms of foregone opportunities for income gains because of the choice of safer but low-return activities by farmers, they might help with reducing fluctuations of income (Pandey et al. 2007) and can reduce food insecurity. Moreover, since the availability of labor is one of the major factors affecting the performance of agriculture, the cultivation of some plots before the rainy season maximizes labor efficiency. In this situation, farmers prepare and cultivate the most distant fields before the expected start of the rainy season (Graef and Haigis 2001), leaving the cultivation of the closest fields for the beginning of the rainy season.

Despite the potential to reduce crop losses and therefore to affect food security, the employment of dry-soil planting needs to be followed by other initiatives to reduce mining of soil nutrients (Bekunda et al. 2004), improvement of soil water retention potential, the reduction of soil erosion, diversification of crops, among others. Practices to design climate-change-resilient farming systems are comprehensively described by Altieri et al. (2015) and can foster the local food security by diluting the risk of total crop loss, reducing the variability in yields and ensuring a source of income.

5 Conclusions

Planting into dry soil is an important strategy for farmers managing several fields under circumstances of low water availability. The main advantage of this practice is the planting before the onset of the rainy season, while the main drawback is that a delay in the onset of the rainy season will increase the risk of a false start or crop failure.

For this impact assessment, a crop model (the Ceres model in DSSAT) was satisfactorily validated for a maize cultivar (Situka) in the Dodoma region of Tanzania, indicating its ability to respond to different planting dates and climate change scenarios.

The climate change scenarios did not indicate a considerable shifting in the onset and end of the rainy season, but changes in the accumulated precipitation during the cropping season were predicted for the 2060–2099 period. For temperature, an increment in maximum temperature during the cropping season for the 2020–2059 period of about 1 °C was predicted for all representative concentration pathways (RCP’s), while for the 2060–2099 period increments ranged from <1 °C to almost 3 °C, depending on the RCP.

Regarding the probability of failure of the maize crop (yields <400 kg ha−1), early planting windows (beginning 1 November) presented up to an 85% chance of failure, with the lowest failure probability occurring with the 15 December planting window. The model simulation results indicated that fields prepared and sowed under the concept of dry-soil planting (no more than 21 days before the onset of the rainy season) had a considerably lower probability of crop failure for the two different tested periods (2020–2059 and 2060–2099).

We conclude that dry-soil planting is a feasible adaptation strategy for farmers to cultivate more fields with maize and therefore reduce the risks of food insecurity in the Dodoma region of Tanzania.



All the activities related to this work were done with support of the Trans-SEC Project, sponsored by the German Federal Ministry of Education and Research (BMBF) and co-financed by the German Ministry for Economic Cooperation and Development (BMZ). The first author also expresses his gratitude to the MACSUR Knowledge HUB (CropM, C1).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. Addiscott, T. M., & Whitmore, A. P. (1987). Computer simulation of changes in soil mineral nitrogen and crop nitrogen during autumn, winter and spring. Journal of Agricultural Science, UK, 109, 141–157.CrossRefGoogle Scholar
  2. Agrawala, S., Moehner, A., Hemp, A., van Aalst, M., Hitz, S., Meena, H., et al. (2003). Development and climate change in Tanzania: focus on Mount Kilimanjaro. Paris, France.Google Scholar
  3. Akinnagbe, O., & Irohibe, I. J. (2014). Agricultural adaptation strategies to climate change impacts in Africa: a review. Bangladesh Journal of Agricultural Research, 39(3), 407–418.CrossRefGoogle Scholar
  4. Alexandrov, V. A., & Hoogenboom, G. (2000). The impact of climate variability and change on crop yield in Bulgaria. Agricultural and Forest Meteorology 104(4):315–327CrossRefGoogle Scholar
  5. Altieri, M. A., Nicholls, C. I., Henao, A., & Lana, M. A. (2015). Agroecology and the design of climate change-resilient farming systems. Agronomy for Sustainable Development, 35(3), 869–890.CrossRefGoogle Scholar
  6. Bekunda, M. A., Nkonya, E., Mugendi, D., & Msaky, J. J. (2004). Soil fertility status, management, and research in East Africa. East African Journal of Rural Development, 20(1), 94–112.Google Scholar
  7. Below, T. B., Schmid, J. C., & Sieber, S. (2015). Farmers’ knowledge and perception of climatic risks and options for climate change adaptation: a case study from two Tanzanian villages. Regional Environmental Change, 15(7), 1169–1180.CrossRefGoogle Scholar
  8. Benin, S., Wood, S., & Nin-Pratt, A. (2016). Introduction. In Agricultural productivity in Africa: Trends, patterns, and determinants. In S. Benin (Ed.), Chapter 1. (pp. 1–23). Washington, DC: International Food Policy Research Institute (IFPRI).
  9. Cooper, P. J. M., Dimes, J., Rao, K. P. C., Shapiro, B., Shiferaw, B., & Twomlow, S. (2008). Coping better with current climatic variability in the rain-fed farming systems of sub-Saharan Africa: an essential first step in adapting to future climate change? Agriculture, Ecosystems and Environment, 126(1–2), 24–35.CrossRefGoogle Scholar
  10. Doebley, J. F., Iltis, H., Gates, R. N., Quarin, C. L., Stephen, L., Zubeda, M., et al. (2010). Mitigating the impact of drought in Tanzania : the WEMA intervention. African Journal of Agricultural Research, 9, 12.Google Scholar
  11. Donatelli, M., Magarey, R. D., Bregaglio, S., Willocquet, L., Whish, J. P. M., & Savary, S. (2017). Modelling the impacts of pests and diseases on agricultural systems. Agricultural Systems, 155, 213–224.CrossRefGoogle Scholar
  12. Dunne, J. P., John, J. G., Shevliakova, S., Stouffer, R. J., Krasting, J. P., Malyshev, S. L., et al. (2013). GFDL’s ESM2 global coupled climate-carbon earth system models. Part II: carbon system formulation and baseline simulation characteristics. Journal of Climate, 26(7), 2247–2267.CrossRefGoogle Scholar
  13. Dupuis, I., & Dumas, C. (1990). Influence of temperature stress on in vitro fertilization and heat shock protein synthesis in maize (Zea mays L.) reproductive tissues. Plant Physiology, 94(2), 665–670.CrossRefGoogle Scholar
  14. Evans, A. E. V, Giordano, M., & Clayton, T. (2012). Investing in Agricultural Water Management to Benefit Smallholder Farmers in Ghana. IWMI Working Papers.
  15. FAOSTAT. (2016). FAOSTAT. Food and Agriculture Organisation of the United Nations. Accessed 20 Dec 2015.
  16. Fox, D. G. (1981). Judging air quality model performance. Bulletin of the American Meteorological Society.<0599:JAQMP>2.0.CO;2.
  17. Graef, F., & Haigis, J. (2001). Spatial and temporal rainfall variability in the Sahel and its effects on farmers’ management strategies. Journal of Arid Environments, 48(2), 221–231.CrossRefGoogle Scholar
  18. Graef, F., Sieber, S., Mutabazi, K., Asch, F., Biesalski, H. K., Bitegeko, J., et al. (2014). Framework for participatory food security research in rural food value chains. Global Food Security, 3(1), 8–15.CrossRefGoogle Scholar
  19. Graef, F., Uckert, G., Schindler, J., König, H. J., Mbwana, H. A., et al. (2017). Expert-based ex-ante assessments of potential social, ecological, and economic impacts of upgrading strategies for improving food security in rural Tanzania using the ScalA-FS approach. Food Security. Scholar
  20. Grassini, P., van Bussel, L. G. J., Van Wart, J., Wolf, J., Claessens, L., Yang, H., et al. (2015). How good is good enough? Data requirements for reliable crop yield simulations and yield-gap analysis. Field Crops Research, 177, 49–63.CrossRefGoogle Scholar
  21. Harris, D., Joshi, A., Khan, P. A., Gothkar, P., & Sodhi, P. S. (1999). On-farm seed priming in semi-arid agriculture: Development and evaluation in maize, rice and chickpea in India using participatory methods. Experimental Agriculture, 35(1), 15–29.CrossRefGoogle Scholar
  22. Harris, D., Raghuwanshi, B. S., Gangwar, J. S., Singh, S. C., Joshi, K. D., Rashid, A., & Hollington, P. A. (2001). Participatory evaluation by farmers of “on-farm” seed priming in wheat in India. Experimental Agriculture, 37(3), 403–415.CrossRefGoogle Scholar
  23. Harvey, C. A., Rakotobe, Z. L., Rao, N. S., Dave, R., Razafimahatratra, H., Rabarijohn, R. H., et al. (2014). Extreme vulnerability of smallholder farmers to agricultural risks and climate change in Madagascar. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 369(1639). CrossRefGoogle Scholar
  24. Hatfield, J. L., & Prueger, J. H. (2015). Temperature extremes: effect on plant growth and development. Weather and Climate Extremes, 10, 4–10.CrossRefGoogle Scholar
  25. Hotland, G. (1994) A farming systems analysis of Mvumi Division, Dodoma Region, Tanzania: A case study on intensifying agriculture in semi-arid Africa (pp. 131). Dodoma: Diocese of Central Tanganika.Google Scholar
  26. Hourdin, F., Foujols, M. A., Codron, F., Guemas, V., Dufresne, J. L., Bony, S., et al. (2013). Impact of the LMDZ atmospheric grid configuration on the climate and sensitivity of the IPSL-CM5A coupled model. Climate Dynamics, 40(9–10), 2167–2192.CrossRefGoogle Scholar
  27. Hunter, J. R., & Erickson, A. E. (1952). Relation of seed germination to soil moisture tension. Agronomy Journal, 44(3), 107–109.CrossRefGoogle Scholar
  28. IPCC, 2007: Climate Change 2007: Synthesis Report. In Core Writing Team, Pachauri, R. K & Reisinger A. (Eds.), Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 104). Geneva: IPCC.Google Scholar
  29. Iversen, T., Bentsen, M., Bethke, I., Debernard, J. B., Kirkevåg, a., Seland, Ø., et al. (2012). The Norwegian Earth System Model, NorESM1-M – Part 2: Climate response and scenario projections. Geoscientific Model Development Discussions, 5(3), 2933–2998.CrossRefGoogle Scholar
  30. Jamieson, P. D., Porter, J. R., & Wilson, D. R. (1991). A test of the computer simulation model ARCWHEAT1 on wheat crops grown in New Zealand. Field Crops Research, 27(4), 337–350.CrossRefGoogle Scholar
  31. Johns, T. C., Durman, C. F., Banks, H. T., Roberts, M. J., McLaren, A. J., Ridley, J. K., et al. (2006). The new Hadley Centre Climate Model (HadGEM1): evaluation of coupled simulations. Journal of Climate, 19(7), 1327–1353.CrossRefGoogle Scholar
  32. Jones, J. W., Hoogenboom, G., Porter, C. H., Boote, K. J., Batchelor, W. D., Hunt, L. A., et al. (2003). The DSSAT cropping system model. European Journal of Agronomy, 18(3–4), 235–265.CrossRefGoogle Scholar
  33. Jørgensen, S. E., Kamp-Nielsen, L., & Jørgensen, L. A. (1986). Examination of the generality of eutrophication models. Ecological Modelling, 32(4), 251–266.CrossRefGoogle Scholar
  34. Kimaro, A. A., Timmer, V. R., Chamshama, S. A. O., Mugasha, A. G., & Kimaro, D. A. (2008). Differential response to tree fallows in rotational woodlot systems in semi-arid Tanzania: Post-fallow maize yield, nutrient uptake, and soil nutrients. Agriculture, Ecosystems and Environment, 125(1–4), 73–83.CrossRefGoogle Scholar
  35. Kimaro, A. A., Timmer, V. R., Chamshama, S. A. O., Ngaga, Y. N., & Kimaro, D. A. (2009). Competition between maize and pigeonpea in semi-arid Tanzania: Effect on yields and nutrition of crops. Agriculture, Ecosystems and Environment, 134(1–2), 115–125.CrossRefGoogle Scholar
  36. KPMG. (2015). Agriculture in Africa Sector Report. Kpmg, 376(9745), 1–18.Google Scholar
  37. Kurukulasuriya, P., Mendelsohn, R., Hassan, R., Benhin, J., Deressa, T., Diop, M., et al. (2006). Will African agriculture survive climate change? World Bank Economic Review, 20(3), 367–388.CrossRefGoogle Scholar
  38. Lesch, S. M., Corwin, D. L., & Robinson, D. A. (2005). Apparent soil electrical conductivity mapping as an agricultural management tool in arid zone soils. Computers and Electronics in Agriculture, 46(1–3), 351–378.CrossRefGoogle Scholar
  39. Liben, F. M., Wortmann, C. S., & Tesfaye, K. (2015). Dry soil planting of maize for variable onset of rainfall in Ethiopia. Agronomy Journal, 107(4), 1618–1625.CrossRefGoogle Scholar
  40. Liwenga, E. T. (2008). Adaptive livelihood strategies for coping with water scarcity in the drylands of central Tanzania. Physics and Chemistry of the Earth, 33(8–13), 775–779.CrossRefGoogle Scholar
  41. Löhr, K., Hochmuth, C., Graef, F., Wambura, J., & Sieber, S. (2016). Conflict management programs in trans-disciplinary research projects: the case of a food security project in Tanzania. Food Security. Scholar
  42. Long, S. P., Ainsworth, E. A., Leakey, A. D., Nosberger, J., & Ort, D. R. (2006). Food for thought: lower-than-expected crop yield stimulation with rising CO2 concentrations. Science, 312(5782), 1918–1921.CrossRefGoogle Scholar
  43. Luhunga, P., Chang’a, L., & Djolov, G. (2017). Assessment of the impacts of climate change on maize production in the wami ruvu Basin of Tanzania. Journal of Water and Climate Change, 8(1), 142–164.CrossRefGoogle Scholar
  44. Lutts, S., Benincasa, P., Wojtyla, L., Kubala, S., Pace, R., Lechowska, K., et al. (2016). Seed Priming: New Comprehensive Approaches for an Old Empirical Technique. In S. Araujo & A. Balestrazzi (Eds.), New Challenges in Seed Biology - Basic and Translational Research Driving Seed Technology (pp. 1–46). InTech. Google Scholar
  45. Mdemu, M. V., Mziray, N., Bjornlund, H., & Kashaigili, J. J. (2017). Barriers to and opportunities for improving productivity and profitability of the Kiwere and Magozi irrigation schemes in Tanzania. International Journal of Water Resources Development, 33(5), 725–739.CrossRefGoogle Scholar
  46. Meinshausen, M., Smith, S. J., Calvin, K., Daniel, J. S., Kainuma, M. L. T., Lamarque, J., et al. (2011). The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change, 109(1), 213–241.CrossRefGoogle Scholar
  47. Mmbaga, T. E., & Lyamchai, C. Y. (2002). Drought management options in maize production in northern Tanzania. In D. K. Friesen & A. F. E. Palmer (Eds.), Proceedings of the Seventh Eastern and Southern Africa Regional Maize Conference (pp. 281–287). Nairobi: CIMMYT (International Maize and Wheat Improvement Center) and KARI (Kenya Agricultural Research Institute).Google Scholar
  48. Moss, R. H. R. H., Edmonds, J. A. J. A., Hibbard, K. A. K. A., Manning, M. R., Rose, S. K. S. K., van Vuuren, D. P. D. P., et al. (2010). The next generation of scenarios for climate change research and assessment. Nature, 463(7282), 747–756.CrossRefGoogle Scholar
  49. Mourice, S. K., Rweyemamu, C. L., Tumbo, S. D., & Amuri, N. (2014a). Maize cultivar specific parameters for decision support system for agrotechnology transfer (DSSAT) application in Tanzania. American Journal of Plant Sciences, 5(6), 821–833.CrossRefGoogle Scholar
  50. Mourice, S. K., Tumbo, S. D., Nyambilila, A., & Rweyemamu, C. L. (2014b). Modeling potential rain-fed maize productivity and yield gaps in the Wami River sub-basin, Tanzania. Acta Agriculturae Scandinavica, Section B — Soil & Plant Science, 65(2), 132–140.Google Scholar
  51. 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.CrossRefGoogle Scholar
  52. Mugalavai, E. M., Kipkorir, E. C., Raes, D., & Rao, M. S. (2008). Analysis of rainfall onset, cessation and length of growing season for western Kenya. Agricultural and Forest Meteorology, 148(6–7), 1123–1135.CrossRefGoogle Scholar
  53. Müller, C., Bondeau, A., Popp, A., Waha, K., & Fader, M. (2010). Climate Change Impacts on Agricultural Yields. In IATRC Public Trade Policy Research and Analysis Symposium 2010 ‘Climate Change in World Agriculture Mitigation Adaptation Trade and Food Security (pp. 1–28).
  54. Mutabazi, K. D., Sieber, S., Maeda, C., & Tscherning, K. (2015). Assessing the determinants of poverty and vulnerability of smallholder farmers in a changing climate: the case of Morogoro region, Tanzania. Regional Environmental Change, 15(7), 1243–1258.CrossRefGoogle Scholar
  55. Muthukuda Arachchi, D. H., Naylor, R. E. L., & Bingham, I. J. (1999). A thermal time analysis of ageing of maize (Zea mays L.) seed can account for reduced germination in hot moist soil. Field Crops Research, 63(2), 159–167.CrossRefGoogle Scholar
  56. Nelson, G. C., Rosegrant, M. W., Koo, J., Robertson, R., Sulser, T., Zhu, T., et al. (2009). Climate change: Impact on agriculture and costs of adaptation. Food Policy.Washington, DC: International Food Policy Research Institute.
  57. Nelson, G. C., Rosegrant, M. W., Palazzo, A., Gray, I., Ingersoll, C., Robertson, R., et al. (2010). Food security, farming, and climate change to 2050: Scenarios, Results, Policy Options (p 155). Washington, DC: IFPRI.
  58. Pandey, S., Bhandari, H., Ding, S., Prapertchob, P., Sharan, R., Naik, D., et al. (2007). Coping with drought in rice farming in Asia: Insights from a cross-country comparative study. Agricultural Economics, 37, 213–224.CrossRefGoogle Scholar
  59. Porter, J. R., & Semenov, M. A. (2005). Crop responses to climatic variation. Philosophical Transactions of the Royal Society B: Biological Sciences, 360(1463), 2021–2035.CrossRefGoogle Scholar
  60. Randall, D. A., Wood, R. A., Bony, S., Colman, R., Fichefet, T., Fyfe, J., et al. (2007). Climate models and their evaluation. In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, et al. (Eds.), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (pp. 589–662). Cambridge: Cambridge University Press.
  61. Reidsma, P., Ewert, F., Lansink, A. O., & Leemans, R. (2010). Adaptation to climate change and climate variability in European agriculture: The importance of farm level responses. European Journal of Agronomy, 32(1), 91–102.CrossRefGoogle Scholar
  62. Rivington, M., & Koo, J. (2011). Report on the Meta-Analysis of Crop Modelling for Climate Change and Food Security Survey. Foreign Affairs. file:///Users/DURU/Documents/Mendeley Desktop/Rivington, Koo - Unknown - Report on the Meta-Analysis of Crop Modelling for Climate Change and Food Security Survey.pdfGoogle Scholar
  63. Rosenzweig, C., Jones, J. W., Hatfield, J. L., Ruane, A. C., Boote, K. J., Thorburn, P., et al. (2013). The Agricultural Model Intercomparison and Improvement Project (AgMIP): Protocols and pilot studies. Agricultural and Forest Meteorology, 170(0), 166–182.CrossRefGoogle Scholar
  64. Schlenker, W., & Lobell, D. B. (2010). Robust negative impacts of climate change on African agriculture. Environmental Research Letters, 5(1), 1–8.CrossRefGoogle Scholar
  65. Seo, S. N., Mendelsohn, R., Dinar, A., Hassan, R., & Kurukulasuriya, P. (2009). A ricardian analysis of the distribution of climate change impacts on agriculture across agro-ecological zones in Africa. Environmental and Resource Economics, 43(3), 313–332.CrossRefGoogle Scholar
  66. Shiferaw, B., Tesfaye, K., Kassie, M., Abate, T., Prasanna, B. M., & Menkir, A. (2014). Managing vulnerability to drought and enhancing livelihood resilience in sub-Saharan Africa: Technological, institutional and policy options. Weather and Climate Extremes, 3, 67–79.CrossRefGoogle Scholar
  67. Tao, F. L., & Zhang, Z. (2010). Adaptation of maize production to climate change in North China Plain: Quantify the relative contributions of adaptation options. European Journal of Agronomy, 33(2), 103–116.CrossRefGoogle Scholar
  68. Thornton PK, Jones PG, Owiyo T, Kruska RL, Herrero M, Kristjanson P, Notenbaert A, Bekele N and Omolo A, with contributions from Orindi V, Otiende B, Ochieng A, Bhadwal S, Anantram K, Nair S, Kumar V and Kulkar U (2006). Mapping climate vulnerability and poverty in Africa. Report to the Department for International Development (pp 171). Nairobi: ILRI. Source:
  69. Tubiello, F. N., Amthor, J. S., Boote, K. J., Donatelli, M., Easterling, W., Fischer, G., et al. (2007). Crop response to elevated CO2 and world food supply - A comment on “Food for Thought...” by Long et al., Science 312 : 1918–1921, 2006. European Journal of Agronomy, 26(3), 215–223.CrossRefGoogle Scholar
  70. Warszawski, L., Frieler, K., Huber, V., Piontek, F., Serdeczny, O., & Schewe, J. (2014). The Inter-Sectoral Impact Model Intercomparison Project (ISI–MIP): Project framework. Proceedings of the National Academy of Sciences, 111(9), 3228–3232.CrossRefGoogle Scholar
  71. Watanabe, S., Hajima, T., Sudo, K., Nagashima, T., Takemura, T., Okajima, H., et al. (2011). MIROC-ESM: model description and basic results of CMIP5-20c3m experiments. Geoscientific Model Development Discussions, 4(2), 1063–1128.CrossRefGoogle Scholar
  72. Westengen, O. T., & Brysting, A. K. (2014). Crop adaptation to climate change in the semi-arid zone in Tanzania: the role of genetic resources and seed systems. Agriculture & Food Security, 3(1), 3.CrossRefGoogle Scholar
  73. Wilkens, P. W., Hoogenboom, G., Porter, C. H., Jones, J. W., & Uryasev, O. (2004). DSSAT v4 Weather Data Editing Program (Weatherman). In Data Management and Analysis Tools - Decision Support System for Agrotechnology Transfer Version 4.0: DSSAT v4 :Data Management and Analysis Tools (Vol. 2, pp. 92–151). Honolulu: University of Hawaii.Google Scholar
  74. Willmott, C. J. (1982). Some comments on the evaluation of model performance. Bulletin - American Meteorological Society.
  75. Wilson, R. T., & Lewis, J. (2015). The MAIZE value chain in tanzania. Food and Agriculture Organization of the United Nations (p. 60). Rome, Italy.

Copyright information

© The Author(s) 2017

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  • Marcos A. Lana
    • 1
    • 2
    Email author
  • Ana Carolina F. Vasconcelos
    • 3
  • Christoph Gornott
    • 4
  • Angela Schaffert
    • 5
  • Michelle Bonatti
    • 2
  • Johanna Volk
    • 6
  • Frieder Graef
    • 2
  • Kurt Christian Kersebaum
    • 2
  • Stefan Sieber
    • 2
  1. 1.Swedish University of Agricultural Sciences (SLU)UppsalaSweden
  2. 2.Leibniz Centre for Agricultural Landscape Research (ZALF)MünchebergGermany
  3. 3.Federal University of Campina GrandeCampina GrandeBrazil
  4. 4.Potsdam Institute for Climate Impact Research (PIK)PotsdamGermany
  5. 5.Hohenheim UniversityHohenheimGermany
  6. 6.Freie Universität BerlinBerlinGermany

Personalised recommendations