北部九州におけるソルガムの出穂に要する数種積算温度の一定性と重回帰式による出穂期の予測(収量予測・情報処理・環境) Comparison of the Constancy of Heat Unit Accumulation Needed for Heading of Sorghum and Prediction of Heading Date by the Multiple Regression Formula in Northern Kyushu District(Modelling, Information Technology and Environment)

北部九州におけるソルガムの計画的栽培利用計画に資するため,播種当年の気象条件を用いて早期かつ簡易な予測法について検討した.兼用型ソルガムの中生品種P956を用いて,まず出穂に要する積算温度の一定性を検討した.積算温度としては単純積算温度,生育下限温度10℃以上を種算した有勅積算温度I法,および播種後日数と出穂までの単純積算温度との関係から最小二乗法により求めた生育下限温度以上を積算する有効積算温度II法の3つを調べた.その結果,積算温度の一定性の程度を示す変動係数が9.7〜8.0%であり,有効積算温度を用いて予測する方法では精度が低いと判断された.つぎに,播種後の平均気温と日長を用いた重回帰式を作成し,出穂期の予測を行った.その結果,播種後10日間の平均気温と播種時の日長を変数とする重回帰式は,その予測値と観測値との差の標準偏差が2.8日であり,出穂期を精度良く予測できた.

The constancy of the heat unit accumulation needed for the heading of sorghum cultivar P956 was examined for the simple heat unit accumulation method, the effective heat unit accumulation Method I in which a temperature above 10℃ was accumulated, and the effective heat unit accumulation Method II in which the temperature above the minimum growth temperature was accumulated. The minimum growth temperature was calculated by the least squares method from the relation between the days after sowing and the simple heat unit accumulation until the heading date. The results showed that the coefficient of variation, showing the accuracy of the application, was 9.7 to 8.0 % for P956 and the constancy was inferior. In addition, the heading date was predicted by preparing the multiple regression formula using the average temperature and day length after sowing. The results showed that the standard deviation of differences in days between the observed and the estimated by the multiple regression formula, using the temperature for ten days after sowing and the day length at sowing time as variables, was 2.8 days for P956 and the heading date was accurately predicted by this formula.