A few previous studies have presented evidence for the spatially heterogeneous response of groundwater to meteorological droughts. To help develop an optimal monitoring network for groundwater resources under drought conditions, Chang and Teoh (1995) described the heterogeneous response of groundwater levels at 13 observation boreholes to meteorological droughts across a basin in Ohio, USA, although they did not investigate the hydrogeological causes of the heterogeneity. Van Lanen (2005) and Van Lanen and Tallaksen (2007) observed that drought characteristics derived from groundwater levels have “spatial effects” and noted that these spatial effects on groundwater drought are an important consideration when monitoring droughts using groundwater levels. Van Lanen and Tallaksen (2007) compared modelled groundwater recharge and discharge for a humid continental climate (Missouri, USA) and a tropical savannah climate (Guinea) for quick- and slow-responding catchments and showed that both climatology and the responsiveness of the catchment as defined by the aquifer characteristics have an influence on drought generation. Peters et al. (2006) investigated the propagation and spatial distribution of aspects of modelled groundwater drought, including recharge, groundwater level and groundwater discharge in the Pang catchment in the UK. They found that short droughts in groundwater levels were most severe near streams and were attenuated with distance from the streams, that longer periods of below-average recharge had more effect on suppressing groundwater levels on interfluves near groundwater divides, and that droughts in groundwater discharge are more attenuated upstream and less so downstream in the catchment. Tallaksen et al. (2009) also modelled the spatio-temporal response of the Pang catchment to drought events and found large differences between the spatio-temporal response of groundwater recharge, level and discharge and the driving meteorological droughts, where droughts in groundwater recharge and levels were found to cover relatively small areas but last longer than the meteorological droughts.

Mendicino et al. (2008) developed a groundwater resource index for drought monitoring and forecasting based on a simple distributed run-off/water balance model, and they evaluated the use of the index in three catchments in southern Italy. They found that the groundwater resource index was highly spatially variable and related it to variations in hydraulic conductivity across the catchments. Using a newly developed groundwater drought index, the standardised groundwater level index (SGI), Bloomfield and Marchant (2013) also investigated hydrogeological controls on groundwater drought. Based on 14 observation boreholes in different catchments across England, UK, they showed that groundwater drought duration depended on the autocorrelation structure of SGI time series. This was in turn inferred to be a function both of spatially varying recharge processes and of saturated flow processes within the local aquifer systems.

There has been significant work on the regional analysis of meteorological and other hydrological droughts. Cluster analysis (CA), principal component analysis (PCA) or some combination of both techniques has been used extensively by meteorologists and hydrologists to investigate the spatio-temporal distribution of hydrological variables, including drought indices (e.g. Klugman, 1978; Karl and Koscienly, 1982; Eder et al., 1987; Stahl and Demuth, 1999, 2001; Lana et al., 2001; Bonaccorso et al., 2003; Vincente-Serrano, 2006; Vicente-Serrano and Cuadrat-Prats, 2007; Raziel et al., 2008; Santos et al., 2010; Fleig et al., 2011; Hannaford et al., 2011; Lorenzo-Lacruz et al., 2013).

Although not previously applied to groundwater drought, CA and/or PCA techniques have been used to classify groundwater level hydrographs for a range of purposes. Winter et al. (2000) classified groundwater hydrographs from three small lake-dominated catchments to investigate groundwater recharge and differences in the hydrographs as a function of the geology of the catchments. Similarly, Moon et al. (2004) applied PCA to 66 groundwater level hydrographs from South Korea to characterise the spatial variability in groundwater recharge. Upton and Jackson (2011) used CA and PCA (following a methodology developed by Hannah et al., 2000) with 52 groundwater level hydrographs from the Pang and Lambourn catchments in the UK to produce regional or “master” hydrographs for modelling the spatial distribution of groundwater flooding.

Here we present the first systematic regional analysis of groundwater droughts using a case study from Lincolnshire, UK. The case study consists of 74 groundwater hydrographs from an area of approximately 8000 km2 that includes three regionally important aquifers, the Lincolnshire Limestone, the Chalk and the Spilsby Sandstone aquifers, each with contrasting aquifer characteristics (Sect. 2). The groundwater hydrographs have been normalised using the SGI technique of Bloomfield and Marchant (2013), and groups or clusters of similar groundwater hydrographs have been identified using CA, where hydrogeologically meaningful clusters are identified by explicitly searching for groups of hydrographs that can be explained by a posteriori knowledge of the groundwater system (Sect. 4.2). The drought characteristics of the clusters have been quantified in terms of drought event duration, magnitude and intensity, and the impact of three major, multi-annual droughts on the SGI time series has been investigated (Sect. 4.4). Controls on the groundwater drought response in each of the clusters have been explored and the results briefly discussed in terms of the implications for monitoring and managing groundwater droughts (Sect. 5).

As a first-order approximation, it is assumed that the broad meteorological drought history of the study area is spatially homogeneous. This assumption means that any relative differences in drought histories between sites or clusters need to be explained in terms of catchment or hydrogeological factors, rather than differences in the drought climatology. This assumption is tested as part of the analysis of correlations between precipitation and regional groundwater levels (see Sect. 4.2). It is also supported by the observations that the whole study area is governed by the same broad climatic patterns, i.e. rain-bearing low-pressure systems from the Atlantic and high-pressure systems leading to a lack of rainfall, with only small variation in annual precipitation across the region (Marsh and Hannaford, 2008). The assumption is also consistent with the previously documented spatial coherence of major hydrological (surface water) droughts in the UK (Hannaford et al., 2011; Fleig et al., 2011; Folland et al., 2015) where the current study area falls within a homogeneous drought region (“region 4” of Hannaford et al. (2011), “region GB4” of Fleig et al. (2012) and Kingston et al. (2013), and the “English Lowlands” of Folland et al. (2015)), although it is noted that the effects of landscape processes can cause heterogeneous meteorological signals to become attenuated (Van Loon, 2015).

Mean annual rainfall varies across the study area from about 600 to 700 mm (Marsh and Hannaford, 2008). The groundwater hydrographs used in the study have been analysed from 1983 to 2012. During this period, three multi-annual episodes of drought have previously been documented by Marsh et al. (2007, 2013), Kendon et al. (2013), Parry and Marsh (2013) and Folland et al. (2015) as follows: 1988 to 1992, 1995 to 1997 and 2010 to 2012 respectively. All are known to have been major drought events causing reduced surface flows and suppressed groundwater levels throughout large areas of central, eastern and southern UK as well as over parts of northwestern Europe (Lloyd-Hughes and Saunders, 2002; Lloyd-Hughes et al., 2010; Hannaford et al., 2011; Fleig et al., 2012; Kingston et al., 2013).

The study area consists of a sequence of Jurassic and Cretaceous aquifers separated by low-permeability clay and shale units. The whole sequence generally dips gently eastwards, and where each of the aquifer units passes under an overlying low-permeability formation they typically become confined. The whole sequence is unconformably overlain by Quaternary superficial deposits. Figure 1 shows the distribution of the three main aquifers in the region – the Jurassic Lincolnshire Limestone; the Lower Cretaceous–Upper Jurassic Spilsby Sandstone; and the Upper Cretaceous Chalk – and includes a schematic cross section of the hydrostratigraphy of the study area. These aquifers are hydrogeologically distinct from each other, and two of them, the Lincolnshire Limestone and the Chalk, have previously documented spatial variability. Below we summarise these features as they inform the heuristic rules used in Sect. 3.2.2 to guide the selection of clusters as part of the CA.

The Lincolnshire Limestone Formation is an oolitic limestone with fine-grained, micritic and peloidal units (Allen et al., 1997), and it is up to 40 m thick at outcrop in the west. It dips and thins to the east, where it becomes confined and eventually pinches out down-dip. Maximum unsaturated zone thickness is up to about 45 m towards the southwest of the outcrop. Groundwater movement is almost entirely by fracture flow along well-developed bedding plane fractures and joints. Abstraction takes place mainly from the region immediately to the east of the outcrop. It has highly variable transmissivities and storage coefficients typical of a fractured limestone. Allen et al. (1997) have reported a wide range of transmissivity values for the Lincolnshire Limestone with an interquartile range of 260 to 2260 m2 day-1 and a geometric mean of 660 m2 day-1, with slightly higher transmissivities being reported from the south of the region, and a very wide range of storage coefficients from 2 × 10-7 to 0.58.

The Spilsby Sandstone aquifer is up to about 30 m thick, consisting of a variably, but often poorly cemented, pebbly quartz sandstone with alternating thin clays and marls (Whitehead and Lawrence, 2006). It outcrops along the foot of the Wolds escarpment (Fig. 1), where it is associated with springs and maximum unsaturated zone thickness is about 30m. It dips to the east and away from outcrop and is generally confined by clays above and below (Fig. 1). Jones et al. (2000) reported transmissivity values in the range 130 to 170 m2 day-1 and a geometric mean of 140 m2 day-1, with storage coefficients ranging from 1 × 10-4 to 1 × 10-3 and with a geometric mean of 4 × 10-4.

The Chalk is a microporous fractured limestone (Bloomfield et al., 1995). Storage and transmissivity are controlled by local sub-karstic development of the fracture network (Bloomfield, 1996; Maurice et al., 2006). The Chalk group reaches a thickness of over 250 m. Groundwater flows from the recharge areas in the west, eastward down-dip towards and into the confined Chalk to the east. The Chalk bedrock surface was significantly altered during the Ipswichian interglacial of the Quaternary. As a result of glacial activity a cliff line and wave-cut platform were eroded into the Chalk (Fig. 1). The Chalk to the east of the palaeo-cliff line is now buried beneath a covering of till, sand and gravel superficial deposits (Whitehead and Lawrence, 2006). Maximum unsaturated zone thickness occurs towards the northwest of the Chalk outcrop and is about 60 m, contrasting with the relatively thin unsaturated zone to the east of the palaeo-cliff line. Allen et al. (1997) and Whitehead and Lawrence (2006) have reported that transmissivity values differ between the northern and southern Chalk in Lincolnshire. In the northern part of the region, transmissivity has an interquartile range of 1020 to 6070 m2 day-1 with a geometric mean of 2350 m2 day-1, whereas in the southern area, in the region of the eroded Chalk, transmissivity is slightly reduced and has an interquartile range of 850 to 3010 m2 day-1 with a geometric mean of 1380 m2 day-1. Similarly, Allen et al. (1997) report storage coefficients with an interquartile range of 3.5 × 10-5 to 1.5 × 10-3 and with a geometric mean of 2 × 10-4 for the northern Chalk and 6.1 × 10-5 to 2.7 × 10-3 and with a geometric mean of 1.5 × 10-3 for the southern Chalk.

The Quaternary superficial deposits in the study area comprise glaciofluvial sand and gravels and tills; peat; tidal flat deposits; river terrace sands and gravels; and overlying alluvium. The Lincolnshire Limestone Formation and the western part of the Chalk outcrop are largely absent of superficial cover.

Groundwater level data for the 74 observation boreholes (Fig. 1) has been provided by the Environment Agency from their groundwater level monitoring network database (Environment Agency, 2014). Prior to the study none of the sites were believed to be significantly impacted by abstraction, although all three regional aquifers are used for public water supply, abstractions for agricultural irrigation and industrial use (Allen et al., 1997; Whitehead and Lawrence, 2006). Where observation boreholes penetrate both the Chalk and underlying Spilsby Sandstone aquifer, the boreholes are completed with screens so that they monitor water levels in only one of the two aquifers. Groundwater levels have been recorded over a range of frequencies, but typically at weekly to monthly time steps. Based on the raw groundwater level data, mean monthly groundwater levels have been estimated. If no observations were available for a given month, then a linear interpolation was used to estimate the monthly groundwater levels following the method described by Bloomfield and Marchant (2013).

Precipitation data have been taken from the Centre for Ecology and Hydrology's Continuous Estimation of River Flows (CERF) 1 km gridded precipitation data set (Keller et al., 2005; Dore et al., 2012; Bloomfield and Marchant, 2013). CERF daily gridded precipitation data are generated from rain gauge data held in the UK Met Office national precipitation monitoring network. A triangular planes methodology is used to produce a daily 1 km2 grid based on a weighted average (inverse distance) of the three nearest rain gauges. Daily rainfall is then summed to give total monthly gridded rainfall. The precipitation data that are used with each groundwater level observation site are the monthly total for the CERF 1 km2 grid square that contains the given groundwater observation borehole.

Before undertaking the regional drought analysis, the correlation between mean SPI and SGI for the entire region, based on all 74 sites, was investigated and the large-scale drought history of the study area were defined.

Figure 2a is a heat map showing the correlation coefficient between SPI for precipitation accumulation periods q = 1 to 36 months and SGI for lags between SPI and SGI of 0 to 5 months based on average values of SPI and SGI for all 74 sites. Dark blue denotes zero correlation and dark red a perfect correlation. Figure 2a shows that there is a good correlation between SPI and SGI. The strongest correlation (0.84, denoted by the closed black circle in Fig. 2a) is for a precipitation accumulation period (qmax) of 12 months (SPI12) with no lag between the SGI and SPI time series. This is consistent with the observations of Bloomfield and Marchant (2013), who previously reported qmax for a variety of groundwater hydrographs from the UK with an average of 13 months, and Folland et al. (2015), who reported a qmax of 12 months for aggregated time series representing the English Lowlands. Figure 2b and c, the average SPI12 and SGI time series respectively, have similar features. For example, episodes of high groundwater levels in 1983, 1994, 2002, and 2008 correspond with high values of SPI12. Three episodes of regionally significant groundwater drought associated with prolonged low groundwater levels from October 1988 to November 1993, May 1995 to February 1998, and from August 2010 to August 2012 correspond closely with episodes of meteorological drought in the SPI12 time series and are consistent with those identified by previous studies (Lloyd-Hughes and Saunders, 2002; Marsh et al., 2007, 2013; Kendon et al., 2013; Hannaford et al., 2011; Parry and Marsh, 2013; Folland et al., 2015). It is inferred from these observations that the large-scale drought history of the study area is represented well by the average SPI12 and SGI time series.

CA has been used to analyse the heterogeneous response of groundwater to droughts across the study region. Clustering has been undertaken using both an agglomerative hierarchical complete-linkage algorithm and a non-hierarchical k-means clustering algorithm, and the resulting clusters searched for those that are hydrogeologically meaningful and that can be explained by known features of the catchment and groundwater systems. Figure 3a is a dendrogram that fully illustrates the level of similarity between individuals within the clusters formed by the hierarchical clustering. The number of clusters is controlled through the threshold on the distance between groups. For example, a threshold of 0.62 leads to the six clusters shown in Fig. 3b. Figure 3c is an equivalent map showing the distribution of sites by clusters formed by k-means clustering for k = 6.

Figure 3b and c show that the spatial distribution of sites as a function of the clusters formed by the hierarchical and non-hierarchical approaches are broadly similar, so the choice of clustering algorithm is based on a plot of RMSSD against number of clusters. Figure 4 shows that the RMSSD for the k-means clustering is systematically lower than that for the hierarchical clustering algorithm where there are three clusters or more, so we have chosen to use the non-hierarchical k-means clustering approach. Note also that both clustering algorithms are better than a clustering scheme based solely on the three classes of aquifer (e.g. Lincolnshire Limestone, Chalk and Spilsby Sandstone). However, an optimal number of k-mean clusters is not clearly evident in Fig. 4. After careful inspection of the clusters formed by a range of k-means clustering classes and a consideration of the study-specific clustering rules described in Sect. 3.2.2, k = 6 was selected. Based on k-means clustering where k = 6, Fig. 3c shows the distribution of sites between the six clusters (cluster 1 to cluster 6, or CL1–CL6).

It can be seen from Fig. 3c that the resulting k-means clusters have a degree of spatial coherency. We have previously assumed that such spatial correlations in the SGI time series are primarily a function of catchment and hydrogeological factors and not a consequence of heterogeneity in the driving meteorology. Here we test if this is the case, prior to further exploration of the features of each cluster, by investigating if precipitation associated with each cluster is substantially different from regional average precipitation. To do this, we first need to identify a representative accumulation period, qmax, for precipitation for each cluster.

Figure 5 is a set of heat maps, similar to Fig. 2a, showing the correlation between SPI for precipitation accumulation periods, q, 1 to 36 months, and SGI for lags between SPI and SGI time series of 0 to 5 months for each of the six clusters. Dark blue denotes zero correlation, and dark red a perfect correlation, with the strongest correlation for each cluster marked by the closed black circle. Table 1 gives qmax for each cluster and also gives the maximum associated correlation coefficient. In all cases except CL2, the maximum correlation between SPI and SGI is found where there is no lag between the two time series. For CL2 it is found at a lag of 1 month. The highest correlations are for CL2, CL4 and CL1 at 0.86, 0.82 and 0.74 respectively. The correlations for CL3 and CL5 are moderate (0.36 and 0.53), and for CL6 there is effectively no correlation (0.09). This is consistent with the observations made in Sect. 4.3 below that linear trends in CL3 and CL5 appear to affect the SGI time series and that the SGI hydrograph for CL6 appears to be anomalous, departing from the mean regional SGI and SPI signals. Values of qmax for CL1 to CL5 from Fig. 5 are 4, 16, 15, 9, and 17 months respectively. Based on these, Fig. 6 shows SPI time series for each cluster, where black lines are the mean SPI for the cluster and the red lines are average SPI across the study area based on the same cluster-specific qmax. Since Fig. 6 illustrates that the two SPI time series for each cluster are similar, we infer that heterogeneity in the driving meteorology across the study region, or at least between the clusters as defined here, does not play an important role in the clustering process and that membership of clusters is dominated by catchment or hydrogeological factors.

Figure 7 shows the mean SGI time series for each cluster. Two main qualitative observations can be made regarding the SGI hydrographs. Five of the six clusters have a similar overall form to the mean SGI hydrograph for the region (Fig. 2c) showing common patterns of low (and high) groundwater level stand. However, CL6 appears to be an exception with a different overall form to the SGI hydrograph – it also exhibits an anomalous step change in SGI from drought to high groundwater level stand over an 8-month period from May 1990 to December 1990. Secondly, two of the clusters, CL3 and CL5, appear to show declining linear trends in SGI, making direct comparison of drought histories between these and other clusters problematic.

Bloomfield and Marchant (2013) have previously shown that mmax, a measure of the significant autocorrelation length of SGI time series, relates to features of groundwater drought. A similar analysis of autocorrelation structure of SGI time series for each cluster is presented here. Figure 8 shows autocorrelation plots for SGI hydrographs for each of the six clusters. In each figure the pale grey lines are autocorrelation plots for individual sites and the solid black line is the autocorrelation plot for the mean SGI time series for the cluster, with the horizontal dashed line indicating the significant level of autocorrelation based on the record length. Based on these plots, values of mmax for the mean SGI time series for each cluster are given in Table 1. Values of mmax for CL3, CL5 and CL6 are anomalously large, consistent with the anomalous features of these SGI hydrographs described above. For the remaining clusters, Fig. 8 and Table 1 show that CL1 has the shortest autocorrelation of 15 months. In comparison, CL2 has an autocorrelation of 23 months and CL4 is intermediate at 18 months.

These contrasting characteristics between the clusters can be seen clearly in Fig. 9a, which illustrates SGI time series for all sites within each cluster, grouped in their respective clusters and presented in the form of a heat map where low values of SGI (associated with drought conditions) are in shades of green to red (increasing drought intensity) and episodes of high groundwater level stand are in shades of green to blue (increasingly high groundwater levels). The three major episodes of drought can be seen clearly in the heat maps for CL1, CL2 and CL4 but are obscured by the trends in CL3 and CL5 and absent in CL6. The degree of coherency of individual SGI time series within each cluster also appears to be consistent with differences in autocorrelation between the clusters. Figure 9b is a heat map of the cross-correlation coefficients for all the individual SGI time series ordered as a function of the six clusters, where dark red denotes high correlations and dark blue denotes low correlations. Sites within CL1 and CL4, i.e. clusters with moderate or short autocorrelation, show relatively low levels of internal coherency compared with sites in CL2 with relatively long autocorrelation that are highly correlated.

Based on the above, the following is a summary of the features of each cluster:

CL1 is dominated by sites from the northern parts of the Lincolnshire Limestone. The mean SGI time series of CL1 has a relatively short autocorrelation (mmax of 15 months), and within the cluster SGI hydrographs are relatively variable.

Clusters CL1, CL2 and CL4 consist of 61 of the 74 hydrographs analysed. Here the characteristics of groundwater droughts in these clusters are quantified, and the response of the clusters to three major drought episodes is investigated.

The duration, magnitude and mean intensity of groundwater drought events have been investigated based on an analysis of the SGI hydrographs where, following the convention of McKee et al. (1993), negative values of SGI denote drought conditions (note, however, that the current convention of the World Meteorological Organization for SPI refers to drought conditions where SPI is continuously negative and reaches and intensity of -1.0 or less and that negative values between 0 and -1 are classified as near normal and simply indicate less than a median precipitation; World Meteorological Organization, 2012). Groundwater drought duration, D, is taken to be the total number of consecutive months where SGI is negative. Groundwater drought magnitude, M, is taken to be the total cumulative value of monthly SGI for a given drought event, and mean drought intensity, I, is given by M / D. Summary drought statistics for CL1, CL2 and CL4 are given in Table 2.

Table 2 shows that there are differences in the character of the groundwater drought events in the SGI hydrographs for clusters CL1, CL2 and CL3. For example, CL1 has more than twice the number of drought episodes (39 episodes) as CL2 (15 episodes), and the average and maximum duration of droughts in CL1 (4.6 and 27 months respectively) are less than half those of CL2 (11.3 and 61 months). The mean drought event magnitude in CL1 (-2.9) is less than half that in CL2 (-7.9), and the mean drought event intensity in CL1 (-0.43) is almost twice that of CL2 (-0.28). In all cases, the drought event statistics for CL4 fall between those for CL1 and CL2. In summary, CL1 exhibits shorter but generally more intense drought episodes compared with CL2, with CL4 drought events being of intermediate character. These relative drought phenomena are a consequence of the degree of autocorrelation in the respective SGI time series, where CL1 has a relatively short autocorrelation compared with relatively long autocorrelation for CL2. This observation is consistent with previous site-specific and modelling studies that noted a similar relationship between the “flashiness” or responsiveness of the groundwater system to meteorological divers and the number of droughts, where quickly responding groundwater systems typically experience more droughts than more slowly responding catchments (Peters et al., 2003; Van Loon and Van Lanen, 2012; Van Lanen et al. 2013).

There is a strong relationship between drought duration and magnitude for all three clusters (Fig. 10), where longer episodes of groundwater drought are associated with droughts of greater magnitude. However, there is no such regular or simple relationship between drought duration and intensity. Maximum drought intensity is similar for all three clusters – for CL1, CL2 and CL4 it is -1.10, -1.05 and -1.13 respectively (Table 2 and Fig. 11) – and is associated with two of the major drought events, i.e. with the latter part of the 1988–1993 drought for CL2 and the 2010–2012 drought for CL1 and CL4. Figure 11 shows the empirical distribution of D, M and I for clusters CL1, CL2 and CL4. Drought duration (Fig. 11) in all three clusters is highly positively skewed with many short drought events and relatively few long drought events. As previously noted, the longest duration droughts are associated with CL2, the cluster with the longest autocorrelation in the SGI time series. These observations are consistent with those of Hisdal and Tallaksen (2003), Tallaksen et al. (2009) and Fleig et al. (2011), who have also described strongly skewed distributions of hydrological drought durations.

Three major, multi-annual droughts have already been described from the regional (Fig. 2) and the cluster-specific (Figs. 7 and 9a) SGI time series. Table 3 summarises differences in the relationships between the driving meteorology and the drought characteristics of each cluster for the three major droughts. Each of the major drought episodes has been quantified using drought characteristics as applied to SPI12 and SGI for each of the clusters.

The 1988–1993 event was the longest of the three major droughts and consequently had the greatest drought magnitude. The groundwater and meteorological droughts started approximately contemporaneously in the winter of 1988. In CL2 the drought was continuous with negative SGI from November 1988 to November 1993, whereas in CL4 there were two short breaks in the drought and numerous breaks in the drought in CL1. In CL2 there was a gradual intensification in the drought magnitude across the event, peaking in June 1992 at an SGI of -1.85 (4 months after the peak SPI12 meteorological drought). In contrast, not only were there short breaks in the drought in CL1 and CL4 but there were approximately annual cycles of drought intensification and decline over the 4-year period – these were particularly pronounced in CL4. This is seen in Fig. 9a, where between 1988 and 1993 the drought status of CL4 is designated by the red tones in the heat map, but these tones show a series of approximately annual variations giving the appearance of vertical stripes during that period and within that cluster. However, the most pronounced differences in response to major droughts between clusters CL1, CL2 and CL4 is in the timing of the end of drought. Groundwater drought conditions ended in CL1 and CL4 in May 1993, 7 months after the end of the meteorological drought, but this was still 6 months before the groundwater drought ended in CL2 (Fig. 9a).

The 1995–1997 drought, although shorter than the 1988–1993 drought, followed a similar pattern, with groundwater drought starting approximately contemporaneously with the meteorological drought. Although it was a continuous event for all three clusters (there were no breaks in the drought for CL1 and CL4), CL1 and CL4 again show approximately annual intensifications and declines in drought status during the episode. Such approximately annual changes in drought status are not seen in CL2. The 1995–1997 drought had the greatest magnitude in CL2 due to the prolonged end to the drought in this cluster, with groundwater drought in CL1 and CL4 finishing approximately contemporaneously with the meteorological drought but 6 months later in CL2. The 2011–2012 drought was much shorter than the other two multi-annual droughts, lasting just over a year starting relatively abruptly in early 2012 and finished abruptly in CL1 and CL4 in May 2012 in response to an unusual episode of spring recharge (Parry et al., 2013). The groundwater drought in CL2 again finished relatively late, this time about 3 months later, in August 2012. The relatively short delay in the breaking of the groundwater drought in CL2 compared with CL1 and CL4 probably reflects the relatively smaller groundwater drought deficit accumulated due to the shorter duration and lower magnitude of the drought compared with the 1988–1993 and 1995–1998 drought episodes.