---

作 者 ：李文峰^1,2;孟亚利^1,2;赵新华^1,2;陈兵林^1,2;许乃银^1,2;周治国^1,2

期 刊 ：应用生态学报 2009年 20卷 04期 页码：879-886

关键词：棉花；

摘 要 ：基于不同熟性棉花品种的异地分期播种试验，综合量化品种特性、主要气象条件（温度、太阳辐射）和栽培措施（施氮量）对棉花铃期与棉籽干物质积累的影响，基于生理发育时间，建立棉花铃期模拟模型，并基于棉籽生长的“库限制”假设，建立棉籽干物质积累模拟模型.通过量化棉铃对位叶氮浓度的变化，为模型构建氮素效应函数.利用不同生态点的品种、播期和施氮量田间试验资料对模型进行检验，结果表明：德夏棉1号、科棉1号和美棉33B的铃期预测值与实测值的根均方差（RMSE）分别为2.25 d、2.61 d和2.75 d，科棉1号和美棉33B的棉籽干物质模拟值与实测值的RMSE分别为9.5 mg·seed^-1和8.2 mg·seed^-1.表明该模型预测精度较高.

Abstract:Field experiments with different maturity cotton cultivars and sowing
dates were conducted at different sites to quantitatively study the effects of c
ultivar characteristics, weather conditions (air temperature and solar radiation
), and crop management variable (N application rate) on the cotton boll maturation period and cottonseed biomass accumulation. The cotton boll maturation period was simulated by using the scale of physiological development time. Based on t
he hypothesis of sink-determined, the cottonseed biomass accumulation model was
then developed. The subtending leaf N concentration of cotton boll was simulate
d with a semi-empirical equation, and used as the direct indicator of the N nut
rition effect on cottonseed growth and development. The model was tested by inde
pendent field data obtained in the Yellow River Valley (Xuzhou and Anyang) and t
he lower reaches of Yangtze River Valley (Huaian) in 2005. The simulated values
of boll maturation period showed reasonable agreement with observed values, with
a root mean square error (RMSE) of 225 days for cultivar DSC-1, of 261 day
s for cultivar KC-1, and of 275 days for cultivar AC-33B. The RMSE of cotton
seed dry mass prediction was 95 mg·seed-1 for KC-1 and 82 mg·seed
-1 for AC-33B, indicating that the model had a good prediction precision.

---