New variants of the Palmer drought scheme capable of integrated utility

• A hybrid model of standardized Palmer drought index (SPDI) is introduced.
• Copulas are employed to propose a new SPDI-based joint drought index (SPDI-JDI).
• SPDI-JDI combines multi-temporal SPDIs and reflects a joint drought status.
• SPDI-JDI performs well against multiple Palmer indices and U.S. Drought Monitor.

SummaryThe Palmer drought model that was developed half a century ago has often been used for monitoring and decision making. It has gained wide-spread familiarity and acceptance, together with numerous complaints, criticisms and improvements. This study first introduces a recent variation of Palmer’s model, i.e. the standardized Palmer drought index (SPDI) as a hybrid of a Palmer drought index (PI) and a standardized drought index (SI). Then, a more complicated methodology is employed to propose an SPDI-based joint drought index (SPDI-JDI), which is also physically based on the Palmer drought model and is a multivariate SI in nature. Using meteorological observations from twelve WMO gauges from all around the globe, empirical and parametric copula-based approaches are adopted and compared for the construction of SPDI-JDI. The findings in this study indicate that the SPDI-JDI derived from parametric copulas can provide an overall measure of the joint probability-based drought status by combining the probabilistic properties and dependence structure of multi-temporal scale marginals of SPDI. Meanwhile, the SPDI-JDI shows high coherence and correlation with individual PIs and performs well in drought detection when compared to the integration of Palmer drought severity index, Palmer modified drought index, Palmer hydrologic drought index and Palmer Z index as well as to the U.S. Drought Monitor observations. Taken together, new variations of the SPDI-JDI and/or SPDI are expected to bring new vigor and vitality into Palmer’s drought model, making it more up-to-date for drought monitoring and prediction.