高级搜索

国家级多源融合降水产品在汉江航道的适用性研究

Applicability study of national multi-source merged precipitation products in the Hanjiang waterway

  • 摘要: 为评估国家级多源融合降水产品(ART_5 km、ART_1 km,以下简称为两种产品)在汉江航道的适用性,本文采用二分类和连续变量指标,从降水强度、航段差异及典型降水过程等方面,系统检验2022—2023年两种产品在汉江航道的适用性。结果表明:(1) 两种产品对有无降水的识别能力有限,但在已发生降水的样本中,与气象观测站的相关性均较好,仅存在轻微系统性低估,整体稳定性良好。(2) 两种产品对降水强度依赖性明显。对弱降水识别和估测最优;小时尺度上,强降水系统性低估明显,极端降水易出现漏报;日尺度上,中雨及以上降水相关性仍较好。(3) 空间分异特征显著。上游定量估测误差最明显;中游航段二分类识别最优;下游在强降水下识别性能下降且误差集中,但对过程演变捕捉能力仍较好。(4) 时间差异明显。弱降水全年误差最小且稳定;中等强度降水在3—4月、9—10月识别准确率最佳且定量估测偏差较小;强降水6—8月识别较优,但存在空间离散大、午后误差大和极端降水部分漏报等问题。(5) 典型降水过程分析显示,两种产品均能够有效捕捉汉江航道降水的时空演变特征(雨带移动、强降水中心位置与结构),ART_5km对中等及以下降水的监测更优,ART_1km在强降水识别中更具优势。综上,ART_5 km和ART_1 km产品可为汉江航道不同风险等级航运预警提供数据支撑,强降水经阈值优化后仍具参考价值。

     

    Abstract: To evaluate the applicability of national multi-source merged precipitation products (ART_5 km and ART_1 km) over the Hanjiang waterway, this study employs dichotomous and continuous variable indices to systematically assess the performance of the two products during 2022-2023 from the perspectives of precipitation intensity, reach-specific spatial differences, and typical precipitation events. The results are as follows. (1) Both products exhibit limited capability in detecting precipitation occurrence, yet demonstrate favorable correlations with meteorological gauge observations for precipitating samples, with only slight systematic underestimation and good overall stability. (2) Marked precipitation intensity dependence is observed. Both products perform optimally for light precipitation, whereas significant systematic underestimation emerges for heavy precipitation and missed detections occur for extreme precipitation; at the daily scale, correlations remain favorable for moderate rain and above. (3) Notable spatial heterogeneity is evident. The middle reach shows the best dichotomous detection, the upper reach exhibits the largest quantitative estimation errors, and the lower reach suffers degraded detection and concentrated errors under heavy precipitation despite maintaining adequate capability in capturing temporal evolution. (4) Substantial temporal differences are found. Except during the dry season (December-February), both products maintain strong correlations with gauge observations. Light precipitation shows minimal and stable errors throughout the year; moderate precipitation achieves the highest detection accuracy during transitional periods (March-April and September-October) with small quantitative biases; heavy precipitation performs better in the rainy season (June-August) but features large spatial dispersion, pronounced afternoon errors, and occasional missed detections for extreme events. (5) Analysis of typical events reveals that both products effectively capture the spatiotemporal evolution of precipitation (including rainband movement and the position and structure of heavy precipitation centers). ART_5 km performs better for moderate and lighter precipitation, while ART_1 km shows greater advantage in detecting heavy precipitation. In summary, both products can provide data support for shipping warnings at different risk levels along the Hanjiang waterway, and heavy precipitation estimates remain valuable for high-risk warning decisions after threshold optimization.

     

/

返回文章
返回