OR/15/007 Introduction

From Earthwise
Jump to navigation Jump to search
Durant M. 2015. Description of groundwater droughts in the UK: 1890 to 2015. British Geological Survey Internal Report, OR/15/007.

The UK has experienced several, in some instances, prolonged drought episodes since 1890. Many of the more recent droughts have been documented in detail, but for earlier droughts, especially before 1976, the records can be fragmentary. Much of the published literature has concentrated on the meteorological aspects of drought, or the consequent impacts of drought on economic activity, in particular agriculture and water supply. The effects of drought on groundwater have not been systematically documented in the same detail.

Bloomfield and Marchant (2013)[1] developed an index for standardising groundwater level time series and characterising groundwater droughts, the Standardised Groundwater level Index (SGI), and calculated SGI for 14 relatively long, up to 103 years, groundwater hydrographs from a variety of aquifers, including Chilgrove House and Dalton Holme. Here we have used the SGI time series for these two sites as the basis of the current investigation of the groundwater drought history of the UK.

SGI is a non-parametric normalization (the normal scores transform) of data that assigns a value to observations, in this case monthly groundwater levels, based on their rank within a dataset, in this case groundwater levels for a given month from a given hydrograph. When SGI is negative it indicates drought conditions and the more negative it is the more intense the drought. The normalised SGI values can be compared between sites over similar time periods to characterise the relative intensity of a drought between sites. More details about how SGI is calculated can be found in Bloomfield & Marchant (2013)[1].

The 9 groundwater drought episodes described in this report have been identified by analysing the SGI time series for Chilgrove House and Dalton Holme from Bloomfield and Marchant (2013)[1], i.e. for dates between 1910 and 2006. For each SGI time series a drought event is defined as any period where consecutive months are negative; a drought ends when SGI returns to a positive value. For each drought event, a total drought intensity can be estimated from the cumulative negative SGI over the drought event. In addition, the average intensity of the event can also be estimated by dividing the total intensity by the number of months of drought. Total and average drought intensities for all drought events in the Chilgrove House and Dalton Holme SGI time series have been estimated. Table 1 below shows the total and average SGI intensities for the seven most intense droughts (based on the average SGI where average drought SGI is <-1) in the record. The most intense drought event at each site is highlighted in yellow. In addition, it can be seen from Table 1 that the impact of drought episodes is not equally experienced at both sites and only sites where the average drought SGI is <-1 are highlighted on bold.

The drought events listed in Table 1, along with the drought known as the ‘long drought’ of 1890 to 1910 (Marsh et al., 2007)[2] and the drought of 2011 to 2012 (Marsh et al., 2013[3] — both identified as a major drought episodes but outside the scope of the SGI record of Bloomfield and Marchant (2013)[1], are the focus of this report.

In summary, this report details the major historical drought periods in the UK with respect to groundwater, from 1890 to 2012. The report consists of 9 standardised drought descriptions containing details of the droughts, as well as a summary, and in some cases an analysis, of the literature available (see Methodology for details). The aim of the report is to enable comparison between drought events, and assess the possibility of creating a drought classification based on event characteristics. There is also an attempt to review the literature and the public and academic discourse surrounding each drought.

Table 1 Total and average SGI for drought episodes at Chilgrove House and Dalton Holme where average drought SGI is <-1.
Drought episode Chilgrove House Dalton Holme
Start End Total SGI Av. SGI Start End Total SGI Av. SGI
1913 to 1914 Dec 13 Dec 14 2.32 0.17 Dec 13 Dec 14 -15.48 -1.11
1933 to 1935 Sep 33 Apr 35 -27.23 -1.36 Nov 34 Nov 35 -5.28 -0.41
1964 to 1965 Sep 64 Oct 65 -3.35 -0.24 Sep 64 Oct 65 -18.26 -1.30
1975 to 1976 Oct 75 Nov 76 -25.27 -1.81 Oct 75 Nov 76 -17.21 -1.23
1988 to 1993 Jul 88 Nov 93 -46.27 -0.71 Jul 88 Nov 93 -99.3 -1.53
1994 to 1998 Jul 95 Jan 98 -38.83 -1.25 Jun 94 Jan 98 -54.33 -1.23
2003 to 2006 May 03 Jan 06 -37.72 -1.14 May 03 Jan 06 -13.17 -0.40


This methodology describes and explains the definitions contained within the drought descriptions. In order for coherent comparison between the 9 droughts, the drought descriptions, along with the start and end dates are presented in a consistent manner, as described below.


There are many definitions and classifications of drought, and some recognised classes of drought include meteorological, agricultural and hydrological droughts (Tabony, 1977[4]; Marsh et al., 2007[5]; Rodda and Marsh, 2011[6]). The aim of this report is not to define drought or groundwater drought; rather it is to provide broad description of groundwater aspects of major drought episodes. The assumptions that have been made regarding identification of the drought episodes are explained below.

All dates referred to in the report under the headings of drought episode, start and end dates are applicable to groundwater, unless specified otherwise. These start and end dates are informed by the Standardised Groundwater Index (SGI; see Bloomfield & Marchant, 2013[1]) except where not available (1890–1910, the year of 2006, 2010–2012). Dalton Holme and Chilgrove boreholes have been taken as proxies for the North and South of England respectively.

Start Date
Usually defined by a significant period of negative SGI but also informed by groundwater level reports and data in some instances. Meteorological dates have been included for context and comparison.

End Date
Groundwater drought terminations can be highly spatially and temporally variable and depend on local aquifer storage and flow properties (Bloomfield & Marchant, 2013[1]) as well as hydrological processes (Tallaksen et al. 2009[7]) and meteorological inputs. As a consequence, the end of a groundwater drought is particularly difficult to constrain. For the ease of comparison between episodes, end dates have been derived from the SGI (SGI values return to positive for a prolonged period of consecutive months) and groundwater reports where possible. Where these sources of information were not available, groundwater data has been used in a qualitative way to estimate end dates. Some interpretation was required in the observation of general trends. Meteorological dates have been included for context and comparison.

Drought Classification
The drought classification provides a description of each drought period based on:

A single or multi-winter event, groundwater start date, meteorological start date, regional or national focus, drought intensity using SGI where possible, other significant comments and termination characteristics.

This classification will be compared later in the report to look for common trends and patterns between events (Drought classification and conclusions).

Regional Characteristics
This section describes the spatial and temporal variability of the drought episode with respect to groundwater. Some context of pre- and post-event conditions is included, as well as any particularly relevant meteorological or hydrological background. These are both explained in greater detail in individual sections in each drought description.

Where possible, existing reports have been used to populate the standardised drought descriptions. These are referenced and can be referred to for greater detail on a particular episode. For droughts without an accompanying report, particularly older droughts, rainfall reports and data as well as hydrometric data were used alongside groundwater data when available. Anecdotal evidence was used only in the absence of this information. In addition, other relevant data and literature sources were used where appropriate.

Drivers and impacts
Following the description of the groundwater drought, each description also includes notes on the relationship with meteorological drivers and impacts of the drought, including surface water impacts, environmental, agricultural, economic and societal impacts and water resource management and regulatory responses to the drought episode.

Observation wells
The following observation wells are referred to in the inventory:

Table 2 Observation wells referenced in report.
Well BGS Reference NGR Notes
Chilgrove House SU81/1 SU 8350 1430 Unconfined Chalk, longest continuous groundwater monitoring record. Hampshire.
Compton House SU71/23 SU 7550 4900 Unconfined Chalk. Hampshire.
Therfield Rectory TL33/4 TL 3330 3720 Unconfined Chalk. Hertfordshire
Dalton Holme SE94/5 SE 9650 4530 Unconfined Chalk. Yorkshire

Further data on these wells can be found at :http://www.bgs.ac.uk/research/groundwater/datainfo/levels/home.html


  1. 1.0 1.1 1.2 1.3 1.4 1.5 BLOOMFIELD, J P and MARCHANT, B P. (2013). Analysis of groundwater drought using a variant of the standardised precipitation index. Hydrology and Earth System Sciences discussions, 10(6), 7537–7574.
  2. MARSH, T, COLE, G and WILBY, R. (2007). Major droughts in England and Wales, 1800–2006. Weather, 62(4), 87–93.
  3. MARSH, T, PARRY, S, KENDON, M and HANNAFORD, J. (2013). The 2010–12 drought and subsequent extensive flooding: a remarkable hydrological transformation. NERC/Centre for Ecology & Hydrology.
  4. TABONY, R C. (1977). Drought classifications and a study of droughts at Kew. Meteorological Magazine, 106(1254), 1–10.
  5. MARSH, T, COLE, G and WILBY, R. (2007). Major droughts in England and Wales, 1800–2006. Weather, 62(4), 87–93.
  6. RODDA, J C and MARSH, T J. (2011). The 1975–76 drought-a contemporary and retrospective review. National Hydrological Monitoring Programme Series.
  7. TALLAKSEN, L M, HISDAL, H and VAN LANEN, H A. (2009). Space–time modelling of catchment scale drought characteristics. Journal of Hydrology, 375(3), 363–372.