Nowcasting for hydrological applications is discussed. The tracking algorithm extrapolates radar images in space and time. It originates from the pattern recognition techniques TREC (Tracking Radar Echoes by Correlation, Rinehart and Garvey, J. Appl. Meteor.. 34 ( 1995) 1286) and COTREC (Continuity of TREC vectors, Li et al., Nature, 273 (1978) 287). To evaluate the quality of the extrapolation. a parameter scheme is introduced, able to distinguish between errors in the position and the intensity of the predicted precipitation. The parameters for the position are the absolute error, the relative error and the error of the forecasted direction. The parameters for the intensity are the ratio of the medians and the variations of the rain rate (ratio of two quantiles) between the actual and the forecasted image. To judge the overall quality of the forecast, the correlation coefficient between the forecasted and the actual radar image has been used. To improve the forecast, three aspects have been investigated: (a) Common meteorological attributes of convective cells, derived from a hail statistics, have been determined to optimize the parameters of the tracking algorithm. Using (a), the forecast procedure modifications (b) and (c) have been applied. (b) Small-scale features have been removed by using larger tracking areas and by applying a spatial and temporal smoothing, since problems with the tracking algorithm are mainly caused by small-scale/short-term variations of the echo pattern or because of limitations caused by the radar technique itself (erroneous vectors caused by clutter or shielding). (c) The searching area and the number of searched boxes have been restricted. This limits false detections, which is especially useful in stratiform precipitation and for stationary echoes. Whereas a larger scale and the removal of small-scale features improve the forecasted position for the convective precipitation, the forecast of the stratiform event is not influenced, but limiting the search area leads to a slightly better forecast. The forecast of the intensity is successful for both precipitation events. Forecasting the variation of the rain rate calls for further investigation. Applying COTREC improves the forecast of the convective precipitation, especially for extrapolation times exceeding 30 min. (C) 2000 Elsevier Science B.V. All rights reserved. References: 42
展开▼
机译：讨论了水文应用的临近预报。跟踪算法在空间和时间上推断雷达图像。它起源于模式识别技术TREC（Tracking Radar Echoes by Correlation， Rinehart and Garvey， J. Appl. Meteor.. 34 （ 1995） 1286）和COTREC （Continuity of TREC vectors， Li et al.， Nature， 273 （1978） 287）。评估外推的质量。引入了一种参数方案，能够区分位置误差和预测降水强度。位置的参数是绝对误差、相对误差和预测方向的误差。强度的参数是实际图像和预测图像之间的中位数比率和降雨率的变化（两个分位数的比率）。为了判断预报的整体质量，使用了预报图像与实际雷达图像之间的相关系数。为了改进预报，研究了三个方面：（a）确定了冰雹统计得出的对流单体的共同气象属性，以优化跟踪算法的参数。使用（a），应用了预测程序（b）和（c）的修改。（b） 由于跟踪算法的问题主要是由于回波模式的小尺度/短期变化或雷达技术本身造成的局限性（杂波或屏蔽造成的错误矢量），因此通过使用较大的跟踪区域和应用空间和时间平滑处理，已经消除了小比例尺的特征。（c） 搜查区域和搜查箱数目受到限制。这限制了错误检测，这在层状降水和静止回波中特别有用。虽然较大尺度和小尺度特征的移除改善了对流降水的预报位置，但层状事件的预报不受影响，但限制搜索区域会导致预报稍好。两种降水事件的强度预报都是成功的。预测降雨率的变化需要进一步调查。应用COTREC可以改善对流降水的预报，特别是对于超过30分钟的外推时间。 （C） 2000 Elsevier Science B.V.保留所有权利。[参考文献： 42]
展开▼
