ORIGINAL PAPER Warm nights drive Coffea arabica ripening in Tanzania A. C. W. Craparo 1,2 & P. J. A. Van Asten 3 & P. Läderach 4 & L. T. P. Jassogne 5 & S. W. Grab 2 Received: 18 February 2020 /Revised: 30 August 2020 /Accepted: 2 September 2020 # ISB 2020 Abstract Studies have demonstrated that plant phenophases (e.g. budburst, flowering, ripening) are occurring increasingly earlier in the season across diverse ecologies globally. Despite much interest that climate change impacts have on coffee (Coffea arabica), relatively little is known about the driving factors determining its phenophases. Using high-resolution microclimatic data, this study provides initial insights on how climate change is impacting C. arabica phenophases in Tanzania. In particular, we use generalized additive models to show how warming nocturnal temperatures (T night ), as opposed to day-time or maximum tem- peratures, have a superseding effect on the ripening of coffee and subsequent timing of harvest. A warm night index (WNI), generated from mean nocturnal temperature, permits accurate prediction of the start of the harvest season, which is superior to existing methods using growing degree days (GDD). The non-linear function indicates that a WNI of 15 °C is associated with the latest ripening coffee cherries (adjusted R 2 = 0.95). As the WNI increases past the inflection point of ~ 16 °C, ripening occurs earlier and progresses more or less linearly at a rate of ~ 17 ± 1.95 days for every 1 °C increase in WNI. Using the WNI will thus not only allow farmers to more accurately predict their harvest start date, but also assist with identifying the most suitable adaptation strategies which may reduce harvest-related costs and buffer potential losses in quality and production. Keywords Phenology . Nocturnal temperature . Climate change . Coffea arabica Introduction The timing of biological events in plants is strongly controlled by climate, with plant phenotypic plasticity being the focus of numerous climate-phenology studies. Temperature and pre- cipitation are widely regarded as the most consistent and dom- inant controllers in the timing of flowering phenology (e.g. Grab and Craparo 2011; Fitchett et al. 2014; Zhang et al. 2015; Crabbe et al. 2016; Chen et al. 2018). Optimal temper- ature ranges allow for correct functioning of plant metabolic processes and photochemistry. For many species, it also di- rectly controls the dynamics of plant development through the process of chilling and forcing requirements during dormancy (Luedeling et al. 2013). Other environmental parameters such as photoperiodic control and soil moisture also contribute to phenological processes (Gordo and Sanz 2010; Jones 2014). Coffee occurs naturally in Ethiopia and South Sudan, with the African continent still hosting more coffee-producing countries (25) than any other (Davis et al. 2019). There are an estimated 12 million coffee farmers globally, with approx- imately half this population in East Africa and ~ 30% in Asia and Oceania (ICO 2015). In Tanzania, coffee accounts for ~ 5% of total export value and provides direct economic support to ~ 450,000 smallholder livelihoods (USDA 2016). Several highland areas in Tanzania produce specialty-grade coffee and therefore obtain considerable premiums worldwide (Tanzania Coffee Industry Development Strategy 2012). Recently, the future of coffee production and security of smallholder farmer livelihoods has gained immediate attention as several studies have demonstrated the vulnerability of C. arabica to climate change (e.g. Davis et al. 2012; Bunn et al. 2014, 2015; Craparo et al. 2015; Moat et al. 2017). Supra-optimal temper- atures and drought stress are shown to be major constraints to current and future coffee production (Gay et al. 2006; Bunn * A. C. W. Craparo a.craparo@cgiar.org 1 International Center for Tropical Agriculture (CIAT), Hanoi, Vietnam 2 School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, P/Bag3, WITS, Johannesburg 2050, South Africa 3 Olam International, Kampala, Uganda 4 International Center for Tropical Agriculture (CIAT), Rome, Italy 5 International Institute of Tropical Agriculture (IITA), Kampala, Uganda International Journal of Biometeorology https://doi.org/10.1007/s00484-020-02016-6