A modification of temperature functions in SUBSTOR-Potato model for improving crop growth and yield simulation under high temperature condition

Abstract

Potato phenology, growth, and yield is projected to be highly affected by global warming in the future. Therefore, assessing the climate change impacts and establishing adaptation strategies are needed. The objectives of this study were to evaluate and calibrate the potato growth model (SUBSTOR-Potato) under elevated temperature condition before applying the model to climate change impact assessments. Planting date experiments under open field condition were conducted at the experimental farm of Seoul National University (37.27oN, 126.99oE), Korea in 2014 and 2015. For the spring season, three cultivars differing in maturity group (Irish Cobbler as extremely early

Superior as early

Atlantic as mid-late maturing) were grown at three different planting dates. Superior and Atlantic were planted at two planting dates in the fall of 2014. Tuber initiation onset dates varied from 11 to 22 days after emergence, depending on cultivars and planting dates. Tuber initiation onset was hastened curve-linearly with increasing temperature, showing optimum temperature at around 22-24oC in all the cultivars tested. and also longer photoperiod and lower solar radiation delayed tuber initiation in Superior and Atlantic. Those factors exerted interactive effects on tuber initiation multiplicatively. The most important determinant of tuber yield was growth duration, which is limited by the beginning of rainy season in summer and frost in the late fall. Yield increased along with delayed tuber initiation. Within the optimum temperature range (17-22oC), larger diurnal temperature range increased the tuber yield. Elevated temperature experiment was conducted in four plastic houses controlled to target temperatures of ambient temperature (AT), AT+1.5oC, AT+3oC, and AT+5oC. Superior was planted at April 29 and September 17 in 2015. For the latter experiments, only emergence and tuber initiation onset was observed. Tuber initiation onset was delayed in spring season relative to fall season due to photoperiod effect. As affected by high temperature, low irradiance and long daylength, plants in AT+5.0oC failed to form tubers at spring season planting. Yield and harvest index tended to decrease with elevated temperature above ambient (22oC) and drop to almost nil at AT+5.0oC. Tuber number at early stage was reduced by higher temperature, resulting in the decrease of assimilates allocated to tuber and average weight of tubers at harvest. Stem growth was enhanced by elevated temperature at the expense of tuber growth. The simulation performance of SUBSTOR-Potato model was evaluated using the above experimental data. The model simulated tuber initiation onset later than the actual as the model determines the tuber initiation date by extrapolating the linear tuber bulking rate to the time axis, and also showed poor performance in simulating tuber initiation onset under low solar radiation, long days, and high temperature condition. We modified the original function for determining the tuber initiation onset and also added a new function of solar radiation effects on tuber initiation under high temperature and long days. The modified model performed better than the original one in predicting not only tuber initiation but also tuber yield under both field and plastic house conditions. In addition, the original model could not explain tuber bulking rate changes as caused by the reduction of tuber number under elevated temperature conditions at early growth stage. The response of rate/duration of tuber formation to temperature was added to the model, resulting in the better accuracy but similar precision for estimating the temporal changes of tuber bulking and tuber yield. According to our data, potato yield is expected to decrease under warmer climate than the current. However, the physiology of tuber initiation and growth are still not well understood and models are too simplified for predicting the heat stress. So further detailed studies are needed to grasp the knowledge of physiological responses of potato growth to high temperature and to add it to the models. Afterwards, validating the models under various environments should be preceded before applications.