A data fusion‐based drought index

Abstract

Drought and water stress monitoring plays an important role in the management of water resources, especially during periods of extreme climate conditions. Here, a data fusion‐based drought index (DFDI) has been developed and analyzed for three different locations of varying land use and climate regimes in Australia. The proposed index comprehensively considers all types of drought through a selection of indices and proxies associated with each drought type. In deriving the proposed index, weekly data from three different data sources (OzFlux Network, Asia‐Pacific Water Monitor, and MODIS‐Terra satellite) were employed to first derive commonly used individual standardized drought indices (SDIs), which were then grouped using an advanced clustering method. Next, three different multivariate methods (principal component analysis, factor analysis, and independent component analysis) were utilized to aggregate the SDIs located within each group. For the two clusters in which the grouped SDIs best reflected the water availability and vegetation conditions, the variables were aggregated based on an averaging between the standardized first principal components of the different multivariate methods. Then, considering those two aggregated indices as well as the classifications of months (dry/wet months and active/non‐active months), the proposed DFDI was developed. Finally, the symbolic regression method was used to derive mathematical equations for the proposed DFDI. The results presented here show that the proposed index has revealed new aspects in water stress monitoring which previous indices were not able to, by simultaneously considering both hydrometeorological and ecological concepts to define the real water stress of the study areas.

Prospects for Imaging Terrestrial Water Storage in South America Using Daily GPS Observations

“…Such a direct application of observed radial deformation series [19,21] does not require inverse modeling. However, one issue with such a use of GPS series is the time period of the observations, which, in most cases, covers only a few years, though it might be possible to overcome this problem by combining different datasets using a multi-fusion drought index approach [22]. For those interested in using individual GPS stations to assess TWS, the studies carried out by Ouellette et al [12] and Birhanu and Bendick [23] are recommended.…”

A Method for Objectively Integrating Soil Moisture Satellite Observations and Model Simulations Toward a Blended Drought Index

“…Agricultural drought is commonly defined as an event where root‐zone soil moisture (SM) deficits result in a reduction in crop yields, plant biomass and ecologic productivity (Anderson et al, ; Azmi et al, ; Bolten & Crow, ; McNally et al, ; Wilhite & Glantz ; Zhang et al, ). The SM status in various soil layers is an important indicator of agricultural drought, providing more information than the rainfall anomaly alone.…”

Evaluation of a combined drought indicator and its potential for agricultural drought prediction in southern Spain

“…SMAI values were calculated for each of the five selected agricultural regions, similar to those of the SPI. To obtain this index, we first calculated soil moisture dynamics by means of the simple water balance model of Brocca et al (2008). The long-term mean soil moisture was taken as the 10-year mean in the study period (2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013).…”