報告書番号

385049

タイトル（和文）

降雨流出解析手法に関する研究（その４）－雨量観測所の最適配置に関する考察－

タイトル（英文）

STUDIES ON THE RAINFALL-RUNOFF ANALYSIS (4)-A FEW REMARKS ON THE OPTIMUM STATIONING OF RAINGAGES-

概要 （図表や脚注は「報告書全文」に掲載しております）

本報告書で提案した流出予測の精度向上を図るための雨量計最適配置手法は，流域平均降雨量について，仮想真値ＸＭと観測降雨量による推定値Ｘに関する２乗誤差の期待値Ｅ（ＸＭ－Ｘ）２を評価基準とし，これが最小となる観測所位置の組合せを最適配置と定義した。この評価式は降雨分布の距離相関係数のみに依存し，空間的定常性を仮定したものである。試験流域への適用結果は以下の通りである。（１）１０～３時間降雨量の範囲内では，距離相関係数は距離とともに指数関数的に低下する。また，本評価式の既設観測所への適用結果は良好であった。（２）本手法を用いて流域内の観測所数とその推定誤差を検討したところ，ダム・サイトへの設置は推定誤差を大きくするだけで意味がない。また，３箇所以上設置すれば精度良く降雨データが得られる。（３）既往の代表的手法である標本計画法との比較から，本手法の方が相対誤差が低く，本最適配置法

概要 （英文）

IT IS NECESSARY TO ESTIMATE THE RAINFALL IN A RIVER BASIN AS ACCURATELY AS POSSIBLE FOR PREDICTION OF THE RUN OFF DISCHARGE. RAINGAGES MUST BE STATIONED PROPERLY TO ENHANCE ACCURACY OF PREDICTIONS. THIS REPORT IS CONCERNED WITH THE OPTIMUM STATIONING OF RAINGAGES FOR ESTIMATION OF AREAL MEAN RAINFALL. WE INTRODUCE THE EQUATIONOF ESTIMATED ERROR DEFINED BY THE EXPECTATION OF SQUARE DEVIATION BETWEEN :ASSUMED REAL VALUE,AND ESTIMATOR BASED ON OBSERVED RAINFALL. AND,THE OPTIMUM STATIONING OF RAINGAGES IS DEFINED BY THE COMBINATIONS OF THE RELATIVE POSITIONS WHICH HAVE THE MINIMUM EXPECTATION. TRANSFORMING THE EQUATION OF ESTIMATED ERROR ASSUMING SPATIAL STATIONARITY IT BECOMESDEPENDENT ONLY ON THE SPATIAL CORRELATION COEFFICIENT OF RAINFALL. THEPROPOSED METHOD IN THIS REPORT WAS APPLIED TO THE SASAMA EXPERIMENTAL BASIN (68KM2). THE MAIN RESULTS ARE AS FOLLOWS. (1)THE SPATIAL CORRELATION COEFFICIENT DECREASED IN EXPONENTIAL FASHION WITH THE INCREASING DISTANCE FOR THE RAINFALL PERIODS IN THE RANGE FROM 10 MINUTES TO 3 HOURS. AND,APPLYING THIS EQUATION OF ESTIMATED ERROR TO THE EXISTING RAINGAGES IN THE BASIN,THE RESULTS SHOW FAIRLY GOOD AGREEMENT. (2)WE EXAMINED THE CORRELATION BETWEEN THE NUMBER OF RAINGAGES AND ESTIMATED ERROR IN THE FOLLOWING TWO CASES. ONE IS THE CASE WHEN A RAINGAGE IN THE NETWORKS IS ALWAYS STATIONED AT THE DAM SITE,AND THE OTHER IS THE CASE WHEN ONE IS NOT ALWAYS STATIONED. THE ERROR INCREASES MARKEDLY FOR THE FORMER CASE. AND,BY STATIONING MORE THREE RAINGAGES,WE CAN OBTAIN SUFFICIENT INFORMATION CONCERNING RAINFALL IN THE BASIN. (3)IN COMPARISON WITH THE PREVIOUS SAMPLE DESIGN METHOD,IT WAS VERIFIED THAT THE RELATIVE ERROR CALCULATED BY OUR PROPOSED METHOD WAS LOWER,AND THUS OUR METHOD IS USEFUL FOR OPTIMUM STATIONING COMPARED WITH THE PREVIOUS SAMPLE DESIGN METHOD.

報告者

キーワード