Growth and development dynamic model of watermelon in gravel-mulched field based on physiological development time

Abstract: Studies on growth and development dynamics of watermelon in gravel-mulched field are helpful in improving utilization efficiency of nutrient, reducing influence of ageing gravel-mulched field on watermelon growth and development and scientifically regulating and controlling watermelon growth and development. This article took into consideration of soil moisture, nutrient, temperature and illumination and established watermelon growth and development dynamic model and leaf area index (LAI) model of watermelon in gravel-mulched field. In 2006, a field experiment was performed in Yinxi Group, Hongquanzi Village, Zhongwei City, Ningxia (37°0'N, 105°12'E). In 2009, two different barrel experiments were conducted under rain shelter. Since the development dynamic model was built based on experience and previous published results. Data from the field and barrel experiments were used to validate the model. The LAI model was established by logistic model, a regression model. The model establishment was based on the results from one barrel experiment in 2009 and the model validation was based on those from the other barrel experiment in 2009. All the experiment simulated dry year, normal year and wet year with volumetric soil moisture of 11-20, 13-24 and 16-27 cm3/cm3, respectively. In the field experiment, organic fertilizer was 150 kg/hm2 for all types of years. In the barrel experiment for LAI model establishment, organic fertilizer was 0, 600, 1 200 and 1 800 kg/hm2. In the barrel experiment for LAI model validation, in addition to application amount above, organic fertilizer amount expanded to 1 560 kg/hm2 and urea was also used with different amount of 48, 61.5 and 33 kg/hm2. The experiment was under drip irrigation under film mulching. Model evaluation was based on root mean square of error(RMSE), normalized RMSE (nRMSE), and correlation coefficient (r). The results showed that: 1) the growth and development dynamic model had the RMSE of 0.95 and 1.78 d, nRMSE of 9.6% and 5.45, and r about 1. It suggested that the model was reliable in simulating growth and development dynamics of watermelon in gravel-mulched field; 2) the LAI model had RMSE of 0.00074-0.0033 d, nRMSE of 0.34%-2.3%, and r of 0.98 under all the type of year. The simulation results of the logistic model were consistent with the measured values and it could fit well the peaks of LAI curves. The LAI model had high accuracy of simulation of watermelon LAI in gravel-mulched field. The models here could be used for understanding watermelon growth dynamics in gravel-mulched field, and provide technical support for improving the utilization efficiency of water and fertilizer, and regulating and controlling watermelon growth process effectively.