Cacao (Theobroma cacao L.) is a tropical perennial crop which is of great economic importance to the confectionary industry and to the economies of many countries of the humid tropics where it is grown. Some recent studies have suggested that climate change could severely impact cacao production in West Africa. It is essential to incorporate our understanding of the physiology and genetic variation within cacao germplasm when discussing the implications of climate change on cacao productivity and developing strategies for climate resilience in cacao production. Here, we review the current research on the physiological responses of cacao to various climate factors. Our main findings are as follows: (1) water limitation causes significant yield reduction in cacao, but genotypic variation in sensitivity is evident; (2) in the field, cacao experiences higher temperatures than is often reported in the literature; (3) the complexity of the cacao/shade tree interaction can lead to contradictory results; (4) elevated CO 2 may alleviate some negative effects of climate change; (5) implementation of mitigation strategies can help reduce environmental stress; and (6) significant gaps in the research need addressing to accelerate the development of climate resilience. Harnessing the significant genetic variation apparent within cacao germplasm is essential to develop modern varieties capable of high yields in non-optimal conditions. Mitigation strategies will also be essential, but to use shading to best effect shade tree selection is crucial to avoid resource competition. Cacao is often described as being sensitive to climate change, but genetic variation, adaptive responses, appropriate mitigation strategies and interactive climate effects should all be considered when predicting the future of cacao production. Incorporating these physiological responses to various environmental conditions and developing a deeper understanding of the processes underlying these responses will help to accelerate the development of a more resource use efficient tree ensuring sustainable production into the future.

Traits such as transpiration efficiency (TE) that are influenced by plant water use can be used to characterise the adaptability of crops to specific growth environments. TE is defined as the amount of biomass produced per unit of water used, and can ensure continued crop production in drought-prone regions. Where TE is associated with reduced use of soil water during the vegetative growth phase, water availability during grain filling may be greater, which can delay the onset of drought stress and increase grain yield under water-limited conditions. This may become even more pertinent with predicted increases in severity and frequency of droughts with climate change. The aims of this study were to firstly dissect TE into its leaf-level physiological components to better understand the effects of genetic variation in these components on TE in sorghum. Secondly, to examine whether TE responses observed under well-watered conditions were preserved under drought, and whether transpiration response was an adaptive response to drought. Twenty-seven genotypes were screened for TE under well-watered conditions using a fully automated lysimetry platform to obtain accurate plant water use data. To determine whether variation in TE among these genotypes was associated with differences in maximum photosynthesis (Amax) or leaf conductance (g) we measured the net carbon assimilation rate of the second last fully expanded leaf at high light intensity, using an infrared gas analyser, and leaf water flux was measured using a porometer, as a proxy for conductance. Genotypic variation in TE among the sorghum germplasm used was mainly associated with differences in the response of transpiration rates to vapour pressure deficit (VPD). Genotypes with low transpiration rates per unit green leaf area (T/GLA) tended to have high TE. Variation in Amax explained some of the differences in TE that could not be explained by T/GLA and may have been a result of mechanisms associated with differences in biochemical pathways that affect the efficiency of conversion of CO2 into photosynthate. While drought tended to increase TE, genotypic variation in TE was largely conserved. However, the response of transpiration rates to drought stress differed across genotypes, with some genotypes showing reduced T/GLA under drought when VPD was high, whereas others did not. These contrasting responses were associated with differences in stomatal responses to drought stress, such that some genotypes were better able to conserve water under drought stress than others. This adaptive response was not related to TE per se and may have important implications for adaptation to drought stress. Hence, the phenotyping of sorghum lines using the associated physiological traits underpinning TE differences is beneficial in identifying traits that may support growth in certain environments and can optimise grain yield production under water-limited conditions. III

The effects of temperature and light integral on fruit growth and development of five cacao genotypes (Amelonado, AMAZ 15/15, SCA 6, SPEC 54/1 and UF 676) were studied in semi-controlled environment glasshouses in which the thermal regimes of cacao-growing regions of Brazil, Ghana and Malaysia were simulated. Fruit losses because of physiological wilt (cherelle wilt) were greater at higher temperatures and also differed significantly between genotypes, reflecting genetic differences in competition for assimilates between vegetative and reproductive components. Short-term measurements of fruit growth indicated faster growth rates at higher temperatures. In addition, a significant negative linear relationship between temperature and development time was observed. There was an effect of genotype on this relationship, such that time to fruit maturation at a given temperature was greatest for the clone UF 676 and least for AMAZ 15/15. Analysis of base temperatures, derived from these relationships indicated genetic variability in sensitivity of cacao fruit growth to temperature (base temperatures ranged from 7.5°C for Amelonado and AMAZ 15/15 to 12.9 for SPEC 54/1). Final fruit size was a positive function of bean number for all genotypes and a positive function of light integral for Amelonado in the Malaysia simulated environment (where the temperature was almost constant). In simulated environments where temperature was the main variable (Brazil and Ghana) increases in temperature resulted in a significant decrease in final pod size for one genotype (Amelonado) in Brazil and for two genotypes (SPEC 54/1 and UF 676) in Ghana. It was hypothesised that pod growth duration (mediated by temperature), assimilation and bean number are all determinants of final pod size but that under specific conditions one of these factors may override the others. There was variability between genotypes in the response of bean size and bean lipid content to temperature. Negative relationships between temperature and bean size were found for Amelonado and UF 676. Lipid concentration was a curvilinear function of temperature for Amelonado and UF 676, with optimal temperatures of 23°C and 24°C, respectively. The variability observed here of different cacao genotypes to temperature highlights the need and opportunities for appropriate matching of planting material with local environments. the reciprocal of the time to reach 95% of full size against corresponding mean temperature (t) in the Brazil simulated environment.

Atmospheric CO2 concentration continues to rise and is predicted to reach approximately 700 ppm by 2100. Some predictions suggest that the dry season in West Africa could be extended with climate change. This study examined the effects of elevated CO 2 concentration and water deficit on growth and photosynthesis of juvenile cacao. Light-saturated photosynthesis (P max ), quantum efficiency, and intrinsic water-use efficiency increased significantly in response to elevated CO 2 , as did a range of growth and development responses (e.g. leaf area and leaf number), but the magnitude of the increase was dependent on the water treatment. Stomatal index was significantly greater in the elevated CO 2 treatment; an atypical response which may be a reflection of the environment in which cacao evolved. This study shows a positive effect of elevated CO 2 on juvenile cacao which may help to alleviate some of the negative impacts of water deficit stress.

The effect of temperature on early vegetative growth, leaf chlorophyll fluorescence and chlorophyll content was examined on four genotypes of cacao (Amelonado, AMAZ 15-15, SCA 6 and SPEC 54/1). A controlled environment glasshouse was used to simulate the temperature conditions of three cacao-growing regions (Bahia, Brazil; Tafo, Ghana and Lower Perak, Malaysia) over the course of a year. Base temperatures calculated from increments in main stem growth varied from 18.6°C for AMAZ 15/15 to 20.8°C for SPEC 54/1. Temporal variation in Fv/Fm observed for two of the clones (SCA 6 and SPEC 54/1) in two of the compartments were correlated with temperature differences over time. Significant differences were also recorded between genotypes in leaf chlorophyll content. It was shown that variation over time in leaf chlorophyll content could be quantified accurately as a function of temperature and light integral. The results imply that genetic variability exists in cacao in response to temperature stress.