The standardized precipitation index (SPI) is a foundational tool for drought assessment. However, its application is constrained by complex probability distribution selection and mandatory normal transformation. To overcome these limitations, this study introduces a systematic framework that uses the Anderson–Darling test to optimize distribution selection. This approach refines drought evaluation through the actual precipitation index (API). By using untransformed precipitation data, the API enables a more direct climatic assessment. Results demonstrated strong consistency between the API and SPI across upper northern Thailand. Both indices successfully detected the 2011 regional floods and the 2015-2016 El Niño drought. Comparatively, the API demonstrated superior sensitivity to moisture saturation, particularly in Chiang Mai, Nan, and Phayao. Furthermore, spatial dynamic analysis using regular vine (R-vine) copulas identified Lampang as the primary regional hub. Lampang governs the central cluster (Chiang Mai, Lamphun, and Phrae) and mediates dependencies between Phayao and Chiang Rai. Nevertheless, localized geographic interactions create substantial concurrent extreme rainfall risks for the Chiang Mai–Lamphun pair. Because it preserves physical rainfall units, the API facilitates more actionable risk management than the SPI. Consequently, integrating R-vine copulas within the API framework is strongly recommended. This integration enhances spatial rainfall modeling, early warning systems, and adaptive water resource management, ultimately supporting climate resilience and sustainable development in vulnerable regions.

The runoff variability index evaluates the fluctuations in runoff levels. In this article, we outline a method for quantifying the runoff variability index using the length-biased Weibull-Rayleigh (LBWR) distribution and the selecting a suitable parameter estimation technique (the maximum likelihood estimators (MLE), method of moment (MOM), maximum product of spacings estimators (MPSE), Anderson-Darling minimum distance estimators (ADE), and Cramér-von Mises minimum distance estimators (CMVE) methods). Our simulations results indicated that most ADE methods showed enhanced efficiency compared to other estimation methods in terms of mean square error (MSE) and average relative bias (AvRB). This study represents the first investigation into the runoff variability index that integrates the LBWR distribution with ADE parameter estimation. Four stations were studied: Two in Nan province and two in Phrae province. The results indicated that Nan province experiences events more frequently than once every ten years, in contrast to Phrae province. Furthermore, the runoff variability index values are useful for classifying the runoff at the four study locations, which corresponds with the particular geographic conditions at each site. Local and regional authorities can use this runoff variability index values to formulate evidence-based water management strategies, improve flood preparedness, and support long-term water security. This directly contributes to the development of more sustainable and resilient infrastructure in the face of an increasingly variable climate.