Abstract
The islands of Jersey, Guernsey, Alderney and Sark lie close to the Normandy coast of France. They expose a largely Precambrian crystalline basement of metamorphic and igneous rocks – Jersey and Alderney also expose some early Palaeozoic clastic sediments – and all have a thin but widespread Quaternary sedimentary cover. The three largest islands were progressively fortified by the British between the early 13th and mid-19th centuries, and by German forces during occupation in World War II, a legacy illustrated by the castles, forts and numerous German coastal fortifications that still adorn them. A German military geologist based on Jersey from mid-1941 to mid-1944, and a military geological team on Guernsey and Alderney during 1942, generated hydrogeological maps and reports that were then in advance of understanding of crystalline basement aquifers elsewhere in the British Isles. All the major documents have now been found in Germany, the USA and UK, although none survived on the islands themselves. Geological mapping and hydrogeological studies postwar under the auspices of the British Geological Survey were completed without access to German data. However, German and British data together now facilitate an appraisal of the heavily stressed aquifers on these small, hard-rock islands over an unusually long (65 year) timespan.
The Channel Islands are British Crown Dependencies situated close to the Normandy coast of France (Fig. 1): Jersey (116 km2); Guernsey (64 km2); Alderney (8 km2); Sark (5 km2); the lesser islands of Herm, Jethou and Brecqhou; and adjacent (mostly uninhabited) islets, rocks and reefs. Despite its small size, the archipelago has been noted (Rose 2009), among the islands of the world, as of particular geological and historical significance.
Location of the largest Channel Islands.
Historically part of the Duchy of Normandy, the islands have owed allegiance to the English monarch since the Norman conquest of England in 1066. Following English defeat in later conflict, mainland Normandy was incorporated into the then Kingdom of France in 1204, and the islands were progressively fortified until the mid-19th century to withstand attack from the French. Alderney then became a ‘Gibraltar of the Channel’ (Anon. 1963): one of the fortified bases from which the Royal Navy sought to dominate the maritime world (Partridge & Davenport 1993). However, during World War II the fall of France to German attack in May/June 1940 left the islands too vulnerable to defend (Cruickshank 1975). The British garrison was evacuated, and German troops occupied the archipelago from mid-1940 to the end of the war in Europe in May 1945. During that time they fortified the largest islands (Partridge 1976). By 1944, when Hitler was expecting an Allied invasion of Europe, he designated the Channel Islands as one of 12 ‘fortresses’ that were to be defended to the ‘last man, round and ration tin’ (Davenport 2003, p. 6). Yet the massive fortifications were bypassed by the Allied invasion of Normandy in June 1944 and surrendered to the British Army essentially untested and intact on 9 May 1945.
Water supply has always been a critical factor in determining the ability of any fortress to withstand siege, and the Germans deployed significant military hydrogeological expertise on the islands to facilitate their fortification. It has long been assumed (e.g. by Mourant 1973; Bishop & Launert 1977, 1979) that the relevant maps and reports were all destroyed prior to surrender. However, recent studies (Rose 2005a–c, 2007, 2008; Robins & Rose 2005; Rose & Willig 2009) have shown this to be wrong. The main maps and reports still exist in Germany or among archives now preserved in the UK or USA. These reveal that their classification of basement aquifers, although elementary, was technically innovative for the British Isles. Together with studies in more recent years by the British Geological Survey (e.g. Robins & Smedley 1991, 1994, 1998; Robins et al. 2002), they facilitate an appraisal (on Jersey and Guernsey) of the hydrogeology of these small hard-rock islands over an unusually long (65 year) timespan.
Geology of the Channel Islands
The islands constitute one of the classical areas of British geology, for uniquely in the British Isles they reveal (Fig. 2) the eroded remnants of part of the Armorican Massif, a terrain more extensively exposed in Brittany and western Normandy, and named from Armorica, the region centred on these areas in Roman times. The islands were isolated from France primarily by Holocene sea-level rise. Their rocks have been studied intensively by geologists, amateur or professional, since the early 19th century (see Rose & Renouf 2005), and key elements of the geology (cf. Parkinson & Plymen 1929; Mourant 1933) were well known prior to the German occupation. However, the first medium-scale (1:25 000) geological maps were only published postwar (Institute of Geological Sciences 1982; British Geological Survey 1986), generated by detailed studies undertaken largely by geologists from Queen Mary College, University of London, under the auspices of the Institute of Geological Sciences (an organization in 1984 re-named the British Geological Survey). Postwar work (e.g. Bishop & Bisson 1989; Roach et al. 1991) has provided a benchmark for continuing academic studies and publications in recent years. Coastal exposure is excellent, and reveals evidence for a range of Precambrian events and spectacular outcrops of calc-alkaline igneous rocks that form a natural laboratory for the study of magmatic processes (Power 1997).
Geological sketch maps of the largest Channel Islands, based on Institute of Geological Sciences (1982) and British Geological Survey (1986), much simplified.
In essence, the northern part of the Armorican Massif comprises (i) gneisses that form a crystalline basement (the Pentevrian) yielding isotopic ages older than 900 million years (exposed on Guernsey, less certainly on Sark and Alderney) and (ii) a series of metamorphic and igneous rocks (Guernsey, Sark and Alderney), low-grade metamorphic sediments (Jersey and Guernsey), plutonic intrusions (all main islands), volcanic rocks (Jersey), a conglomerate (Jersey) and sandstone (Alderney), all associated with a late Precambrian to early Palaeozoic period of accretion at an active continental margin commonly known as the Cadomian Orogeny.
Jersey principally reveals metasediments (the Jersey Shale Formation) and overlying volcanic rocks of very late Precambrian age, intruded by end-Precambrian and Ordovician granites and associated igneous rocks. This complex is unconformably overlain to the NE by Cambro-Ordovician conglomerates with subordinate sandstones and mudstones. The metasediments are predominantly fine- to medium-grained turbiditic sandstones, siltstones and mudstones, and the volcanic rocks are lavas, tuffs, and agglomerates of andesitic and subordinate basaltic composition, succeeded by rhyolitic ignimbrites, lavas and air-fall or water-laid pyroclastics, all affected by low-grade regional metamorphism. Minor intrusions are common, mostly as dykes, and indicate several phases of emplacement.
Guernsey has two distinct parts. Its southern metamorphic complex is principally of Pentevrian basement, dominantly ‘Icart’ Gneiss (of granitic composition) across the south and ‘Perelle’ Gneiss (a foliated quartz diorite) in the centre and west of the island, all cut by a large number of minor intrusions, mostly dykes (Roach et al. 1991). Its northern plutonic igneous complex is Cadomian in age, and largely diorite, flanked to the east by gabbro, to the north by granodiorite and to the west by granite, mostly emplaced (with relatively few minor intrusions) by c. 500 million years ago.
Alderney also has two distinct parts, but these are a Precambrian basement (of gneiss and granodiorite) to the west, and a feldspathic sandstone (which is now generally considered to be coeval with the post-orogenic conglomerates of Jersey) to the east.
Sark is formed by ancient metasedimentary gneisses and younger foliated, coarsely crystalline granodiorites of Precambrian age. The gneisses include quartz biotite, augen and quartz diorite gneiss with darker bands reflecting former mudstones and lighter bands former sandstones.
Deposition of the conglomerates and sandstone was followed by a timespan of c. 500 million years that left no local onshore sedimentary record. The largest islands developed a surface topography of low plateaux, now gently inclined and incised by small streams in steep valleys. The youngest rocks in the archipelago overlie these features: unconsolidated superficial sedimentary deposits of Quaternary age, which reflect changing climates and relative sea levels from the Middle Pleistocene to the present day (Keen 1978, 1981). Largely deposits of interglacial or periglacial origin, they comprise raised beaches (often backed by fossil cliffs), blown sands, ‘head’ (periglacial solifluxion breccias), peat or alluvium in coastal or valley areas, and widespread loess on the plateau surfaces. Steep cliffs fringe much of the present coast of all the large islands, indented by sandy bays. Tides have a high range (12 m in Jersey to 6 m in Alderney), locally exposing extensive rocky wave-cut platforms or sandy beaches at low water.
British fortification
The earliest stone-built fortifications on the islands were castles on Jersey (Fig. 3) and Guernsey (Fig. 4), to provide places of refuge or to dominate key harbours. On Jersey, Mont Orgueil Castle was founded in the early 13th century and progressively strengthened thereafter to counter developments in weapon technology (Rybot 1978); Grosnez Castle (a headland hill protected by walls and a ditch, presumably as a place of refuge) was built in the early 14th century; and in the late 16th century, as siege cannon became more effective, Mont Orgueil was superseded as the principal island fortress by Elizabeth Castle, on a granite/diorite islet at the entrance to the main port, St. Helier (Rybot 1986). Mont Orgueil, built on an elevated granite promontory, had a well 19 m deep (Ford 2007) and withstood French attacks in the 13th, 14th and 15th centuries. Grosnez lacked a secure water supply, so was easily if briefly captured during a raid by the French in 1373, and had been allowed to fall into ruin by the 16th century. A secure water supply for Elizabeth Castle was provided by cisterns to store rainwater runoff, and although a well was dug in the castle parade in 1800 this source proved brackish and additional rainwater tanks were added (Ford 2008).
Map of Jersey indicating sites of British-built 13th- to 19th-century fortifications and principal streams. After Rose et al. (2002).
Map of Guernsey indicating sites of existing British-built 13th- to 19th-century fortifications. Compiled from Guernsey Tourist Board literature.
On Guernsey, Castle Cornet, sited on a rocky granodiorite islet facing the island's capital, St. Peter Port, was also founded in the 13th century and developed through later centuries as the island's principal fortress (Guernsey Museum Team 2008). A well at Castle Cornet was originally located on the eastern side of the islet, outside the main castle walls. It was brought within the fortifications by construction of the Well Bulwark as earthworks in the latter half of the 16th century, later rebuilt in stone. Plans of the castle dating from 1680, 1690 and 1734 show two wells, by 1734 both brackish (H. Glencross, Monuments Curator, Guernsey Museum, pers. comm.). Ivy Castle (the Château des Marais) was founded by the 13th century, Vale Castle by c. 1400, both enhanced subsequently, but now largely in ruins.
Alderney was not fortified until much later (Fig. 5) (Partridge & Davenport 1993). Work was begun on its Essex Castle in 1546, but soon abandoned, as government policy changed. The present ‘castle’ is a barracks (later hospital) built in the 1840s.
Map of Alderney indicating sites of British-built 13th- to 19th-century fortifications. Compiled from Alderney Society literature.
In the 18th and 19th centuries, redoubts and batteries were constructed to defend those coastal embayments on the main islands deemed vulnerable to sea-borne invasion. Coastal defence became of increased importance from 1778, when the French became allies of the Americans in their War for Independence and it seemed likely that they would seize the opportunity to attack the islands. On Jersey, a series of towers was built to defend the beaches, at first with protected musket fire but later also by each mounting a single gun on the roof (Rose et al. 2002). Twenty-two round towers and one square tower were eventually built, by 1798. Eight more substantial (Martello) towers were constructed between 1807 and 1838, with large cisterns to store rainwater from the roof. A total of 31 towers thus protected the lowland shores of the island (Fig. 3). Forts protected the few navigable bays that penetrated the cliff-dominated northern coast.
On Guernsey (Fig. 4), batteries existed on several headlands (particularly of the east coast) during the 17th century, and 15 round towers of distinctive local style were built in 1778 or shortly thereafter. Three Martello towers (each mounting at least one 24-pounder carronade, to counter naval assault) were built in 1804. These three towers formed the nucleus of more extensive fortifications and artillery batteries, developing into forts. Three of the round towers on Guernsey were later destroyed, as were four of the towers on Jersey.
Alderney (Fig. 5) was protected by over a dozen batteries sited during the 18th century along its low-lying northern coast, although not by towers.
In the 1840s the French created a strongly fortified naval base at Cherbourg (Fig. 1) on their Normandy coast and, some 40 km to the west, the British countered this by construction of a ‘harbour of refuge and observation’ at Braye Bay on Alderney (Anon. 1963; Partridge & Davenport 1993). This was complemented by perimeter forts constructed between 1850 and 1858 on all but the southern, cliff-bounded coast of the island. Fort Albert, overlooking Braye Bay, was developed as the site of main defence. This series of fortifications was thus more substantial and more recent than military works on Guernsey and Jersey, where construction of new harbours was planned (on Guernsey) or begun but never completed (on Jersey at St. Catherine's Bay, north of Mont Orgueil Castle). All the forts except the first, Grosnez, were designed by William Jervois, then a Royal Engineers captain but later to become Lieutenant-General Sir William Jervois (Kinross 2004). In 1860 (as a young major) he was to be appointed ‘design leader’ for a programme of fortifications on the UK mainland to defend against a potential French invasion: the largest system of fortifications that the British Isles had ever seen or, arguably, ever would see (Crick 2012). They were a major step in a career of fortress construction and public service that would see the works of Jervois spread across the world, from the UK to North America, Bermuda, India, Australia and New Zealand.
British Channel Island fortifications in general were not particularly innovative in their use of groundwater. Constructed to resist sea-borne assault rather than individually withstand siege, they typically relied for their water supplies on ready access to surface waters in the respective island hinterland, supplemented by cisterns to store rainwater. The exception is Fort Regent, the last British fortress to be constructed on Jersey (Davies 1971). This was founded in 1806 when defensive works throughout the island were in hand to counter the threat from Napoleonic France, and completed in 1814 as the threat came to an end. By this time Elizabeth Castle was considered too isolated to be of much value, except as a refuge of last resort, and the extensive new fort was constructed as a replacement defence for the island's main port. It was to be garrisoned by British regular troops until 1932.
Sited for defensive reasons on high ground overlooking St. Helier, the island's principal town, the fort was provided with a well (Fig. 6). The fort was designed and its construction supervised by John Humfrey, then a major in the Royal Engineers. Humphrey had begun his military career as a ‘Practitioner Engineer’ (=2nd Lieutenant) in January 1776, and in 1781 been ‘placed in charge of sinking wells for vital water supplies at Fort Townshend in Sheerness and also those known as the King's Lines at Harwich’ (Davies 1971, p. 64), so had prior experience of providing secure water supplies for fortified sites, although contemporary literature credits Captain Thomas Hyde Page as the senior officer in charge (Mather 2012). The Fort Regent well was sunk at a diameter of 3 m to a depth of 235 ft (72 m) through solid ‘syenite’ (a term used at that time to describe a granite in which hornblende is present), using explosive charges of gunpowder. Twelve miners worked day and night from December 1806 to October 1808, in total for 23 months, thus achieving an average rate of sinking of c. 3 m per month.
View of Fort Regent overlooking St. Helier Harbour (top left), with cross-section through the granite Town Hill (its summit levelled for construction of the Fort), showing the Fort's well, constructed in 1806–1808. The lower horizontal line indicates the low water level of Spring tides; the upper line normal low water level. Rest water level shown within the well is as at 25 June 1839, and is clearly higher than levels in wells marginal to Town Hill. The figure (top right) of a man with divining rod shows a then well-known local dowser, indicative of serious interest in this technique. From Jones (1840).
The well was mostly dry to its final depth, which was below sea level. One final blast at that depth (72 m) caused water to enter the well from a confined fracture and rise 70 ft (22 m) up the shaft to a near static level reflecting the local potentiometric surface. Another Royal Engineers officer (Jones 1840) subsequently described the critical moment of well construction: ‘After sinking through 235 ft of compact [igneous] rock, and upon firing a blast, the spring was laid open … when [water] poured in like a torrent, to the great astonishment of the miners, who were suspended in the bucket, waiting the effects of the explosion’. A pump was installed to raise water to two casemated cisterns, from which it could be pumped to parade ground level. Initially, the main pump was operated by a capstan with two horizontal spars. Jones (1840) estimated that by this means ‘twenty-four men working for two hours, without fatiguing themselves, can with ease pump into the cisterns 800 gallons of water’: a rate of abstraction equivalent to a short-term yield of 0.5 l s−1. Compared with the earlier (probably 16th century) well at Mont Orgueil Castle, also sunk through granite but to a depth of only 57 ft (19 m) (Rybot 1978, p. 24), the Fort Regent well marked a significant technical advance in terms of both depth and yield. It was still constructed with no real understanding of hydrogeological principles, but Jones's (1840) later description seemingly correlates with an increasing military awareness of the practical applications of geology in the 1840s (cf. Mather & Rose 2012).
Jones (1840) was clearly intrigued by the upwelling of water in the Fort Regent well and made some interesting observations, significant as his work precedes Henry Darcy's investigations of head and flow in a porous medium by 15 years. Jones noted that the water strike elevation was about the same as the water level in the town well located at the bottom of the hill occupied by Fort Regent (Fig. 6). He also noted that the water strike lowered the water level in that well permanently and assumed that one water body supplied both wells and others in the area, such as the barrack-office well, in which the water level was at the same general elevation. He later changed his mind when in the dry summer of 1835 the Fort Regent well had to be used to supply the town as the town wells had dried up. However, it is likely that the storage available to the Fort Regent well was greater than that of the town wells, ensuring a more sustainable supply. Jones made a further check on provenance by analysing the water from the respective sources, noting that ‘the Fort Regent Well contained about 4 grains of fixed salts in a pint of water [c. 450 mg l−1], and the barrack-office well not quite so much. The difference in the amount was due to the smaller quantity of lime in the barrack-office well’.
German fortification
The Germans adapted many of the British castles, forts and towers to suit their own needs, because these were strategically sited and sufficiently massive in construction to shelter troops and weapons deployed to counter amphibious attack by the Allies. These fortifications were augmented between 1941 and 1945 by a construction programme intended to make the islands as a whole an impregnable fortress. The programme is well documented and illustrated in annual volumes of the Channel Islands Occupation Review (currently to number 39, issued in May 2011), a series of Archive Books also published by the Channel Islands Occupation Society, and many other books (e.g. Cruickshank 1975; Ramsey 1981; Stephenson 2006).
Jersey (Fig. 7) was fortified ultimately with 14 batteries of coastal artillery, together with an impressive series of associated direction-finding towers and observation posts that ringed the coast, about 100 fortified infantry defence areas (many of the larger ones, the ‘resistance nests’ or ‘strongpoints’, equipped with 105 mm calibre guns, or 47 mm anti-tank guns, set within massive casemates of reinforced concrete), a series of anti-tank walls and ditches, an extensive minefield complex, a variety of tunnels for storage and shelter, bunker complexes that provided central and local command and control centres, and numerous anti-aircraft guns set on concrete standings (Rose et al. 2002). A similar range of defences was constructed on Guernsey (Fig. 8) (Gavey 2001) (cf. maps illustrated by Stephenson 2006, pp. 18–19). Alderney was smaller than the other islands, but for its size was fortified to a relatively greater degree (Fig. 9), with five coastal artillery batteries, up to 22 anti-aircraft batteries, 13 strongpoints, 12 resistance nests, three defence lines and 30 000 land mines (Davenport 2003). Very few fortifications were constructed in Sark, limited essentially to defences in the harbour area, and tunnels.
Map of principal German fortifications and streams on Jersey as at June 1944. For coastal batteries, data are provided for number, calibre, type and range of guns in place. After Rose (2008), and a larger-size figure within a poster illustrating Atlantic Wall defences 1940–1945: Jersey published in 1984 by Colin Partridge through Ampersand Press, Alderney.
Map of principal German fortifications on Guernsey as at June 1944. For coastal batteries, data as for Figure 7. After Rose (2005c) and a larger-size figure within a poster illustrating Atlantic Wall defences 1940–1945: Guernsey published in 1992 by Colin Partridge through Ampersand Press, Alderney.
Map of principal German fortifications on Alderney as at June 1944. For coastal batteries, data as for Figure 7. After Rose (2005c), and a larger-size figure within a poster illustrating Atlantic Wall defences 1940–1945: Alderney published in 1984 by Colin Partridge through Ampersand Press, Alderney.
Fortification began after the German Führer Adolf Hitler directed on 10 June 1941 that the Channel Islands were to become permanently and impregnably fortified outposts of the German state, the Third Reich. It continued as the islands were subsequently incorporated into the Atlantic Wall, a series of intermittent coastal fortifications that stretched from northern Norway to southwestern France and marked the western limit of German-dominated Europe. Work was directed by Fortress Engineer units that began to arrive in July 1941 in rapid response to the Führer's directive. These soon comprised Fortress Engineer Command XIV (based in Guernsey) with overall responsibility, plus Fortress Engineer Staff 14 on Jersey and Fortress Engineer Staff 19 on Guernsey (Ginns 1994).
Water was a particular concern. The period 1940 to 1943 was characterized by below average rainfall, very low winter rainfall and the lowest August rainfall on record (Maison St Louis rainfall data: Jersey Meteorological Office). Potable water was preferred to sea water for use in preparing the vast quantities of concrete used in construction of the fortifications. It was also needed to maximize development through irrigation of the agricultural resources of the islands and to support a substantial addition to the civilian population. On Jersey, there was a requirement for a construction workforce of up to 6000 men and a garrison of up to 11 500 troops in addition to the civilian population, which pre-war numbered about 50 000 but was reduced to nearer 40 000 by evacuation immediately before the occupation and deportations within it. The situation on the smaller but more exposed island of Guernsey was relatively similar; at the height of operations, in May 1943, there was a workforce of 5100 foreign labourers and a German garrison of 13 000 troops. Alderney's population of about 1200 had been almost completely evacuated before the occupation, but replaced by a garrison of up to 3800 troops, supplemented in 1942/1943 by a construction workforce of 2500 to 3000, and from 1943 to 1944 up to 1000 prisoners in an SS concentration camp. This greatly increased demand for water beyond pre-war levels. The Fortress Engineer Command, therefore, sought hydrogeological advice so as to maximize water supplies from all natural sources. The advice was duly contributed through two distinct organizational channels.
On Jersey, the largest of the Channel Islands, a resident military geologist was appointed at the outset as a full member of Fortress Engineer Staff 14 – Lieutenant (later Captain) Walther Klüpfel (1888–1964) (Bishop & Launert 1977, 1979; Rose et al. 2002). Klüpfel was already 53 years of age at the time of his call-up for military service on 31 March 1941. He had graduated with a doctorate in geology from the University of Strasbourg in 1914, prior to service as a military geologist in World War I. He received the Iron Cross, Second Class, for work on water supply to a section of the Western Front. He published (Klüpfel 1916) a significant account of groundwater occurrence in the Jurassic strata in the French/German borderland of Lothringen (Lorraine), and postwar contributed a substantial section (46 pages plus two maps) on Jurassic geology of this region to the volume (Kraus 1925) in Wilser's (1923–29) definitive German overview of the geology of World War I battlefields. Following demobilization he held several industrial posts in Germany and abroad before obtaining a non-established lectureship at the University of Giessen, at which time he supported himself largely by consulting work for the quarrying industry. He moved to Jersey in July 1941, thus soon after the Führer's directive of 10 June, beginning fieldwork the following month. Although both Fortress Engineer Command XIV and Fortress Engineer Staff 14 left the Channel Islands on 19 November 1942, leaving Fortress Engineer Staff 19 on Guernsey in overall technical direction (Ginns 1994), Klüpfel seemingly remained on Jersey until the Allied liberation of Normandy in June 1944 cut off supplies of cement from the mainland and so brought fortress construction to an end. Bishop & Launert (1979, p. 37) record that the last entry in his field notebooks is dated 23 April 1944. Klüpfel was thus a very experienced military geologist and hydrogeologist who worked on the islands for about three years, but geologically in near isolation as the only geologist in a Fortress Engineer Staff unit. Water supply assignments occupied most of his time; of 50 tasks undertaken between his arrival and October 1942, 19 were associated principally with problems of water supply, 10 with quarrying for construction materials, nine with tunnelling or excavation for underground facilities, six with site investigation for surface engineering works, and a further six were principally study tours or visits arranged to help orientate visiting geologists (Rose 2005a, table 1).
For Guernsey and Alderney, the outermost islands and so those most vulnerable to British assault, the situation was different. Geological and hydrogeological advice were contributed by Wehrgeologenstellen (military geology centres or teams) that formed part of the German Army's military geological service (Rose 2007). This organization, at its peak (in November 1943), contained 40 Wehrgeologenstellen. Each team was typically led by at least two officers or officials assigned major/captain status. Most such officials were designated as ‘TKVR’ (Technischer Kriegsverwaltungsrat: Military (War) Technical Administrator). These were geologists qualified to doctoral degree status, recruited or conscripted from the teaching staff of German universities or from Germany's national or regional geological survey organizations. In addition, each team included several non-commissioned officers, as assistants and/or draughtsmen. The teams were able to provide maps and reports generated remotely from a main base, as well as from brief consultancy visits by individual specialists, or deployments in theatre of the team as a whole.
In late 1941, TKVR Dr Walter Wetzel (1887–1978) of Wehrgeologenstelle 9, based on the French mainland, remotely generated appraisals for Guernsey and Alderney that included water supply (Rose 2005a). These pilot studies were followed by visits to Alderney or Guernsey by at least three German Army geologists, TKVRs Röhrer, Scherer and Hoenes, the last from Wehrgeologenstelle 7. Tasks were primarily associated with problems of water supply at sites of particular military interest, notably the ‘Mirus’ battery of coastal artillery on Guernsey, whose four 48-ton guns of 305 mm calibre were to be the most formidable on the Channel Islands (Fig. 10), and manned by a correspondingly large number of troops (Partridge & Wallbridge 1983).
One of the four gun emplacements for the Mirus (formerly Nina) Battery of NW Guernsey (cf. Fig. 8). Translation of the nine items listed top down in the German key shows that each site had its own shell store, cartridge store, ventilation system, engine room, fuel store, heating system, washroom/lavatory, accommodation (for 72 men), and site entrance. Outer walls were of reinforced concrete 2 m thick; the bunker complex measured some 31×31 m; and the gun barrel was 15.8 m in length. From Festung Guernsey, a record of fortifications prepared by the German armed forces during their period of occupation, courtesy of the Priaulx Library and the Royal Court, Guernsey.
In April 1942, following orders that initiated the main construction phase of the Atlantic Wall as such, Wehrgeologenstelle 4 was deployed as a unit for fieldwork on Guernsey and Alderney. It was led by a former Prussian Geological Survey officer with specialist expertise in ‘soft rock’ Quaternary geology: TKVR Dr Bernhard Beschoren (1898–1982). His deputy was a younger geologist with ‘hard rock’ expertise gained through post-doctoral research at the universities of Freiburg, Heidelberg and Berlin: TKVR Dr Dieter Hoenes (1912–1952). Between them they were to generate 10 substantial reports designated ‘expert opinions’ (Gutachten) and at least 11 reports additional to this numbered series (Rose 2007, tables 1 & 2), plus at least 10 specialist maps (Rose 2005a, table 3). Many of these related to groundwater conditions on the islands.
German hydrogeological investigations 1941–1944
Whereas a key groundwater investigation target today is the development of a conceptual groundwater flow model that encapsulates all the known reactions of a given aquifer, Walther Klüpfel set out initially to provide a groundwater typology map for Jersey, at a scale of 1:25 000 (Fig. 11). This accompanied a preliminary report that summarized groundwater knowledge for the island at that time. The report, dated 26 January 1942, was re-discovered only very recently (Rose & Willig 2009).
First draft of a ‘water map’ for the island of Jersey (Erster Entwurf einer Wasserkarte der Insel Jersey), original at scale of 1:25 000, compiled in January 1942 by the military geologist Walther Klüpfel during German occupation of the Channel Islands. For translation of key, see Table 2. From Rose (2005b): reproduced by permission of the Geologists’ Association, London, and the Bundesarchiv–Militärarchiv, Freiburg-im-Breisgau.
The report is a remarkable document that considers the diverse mechanisms and processes that control groundwater throughflow within a moderate-sized island predominantly comprising a complex shallow fractured hard-rock aquifer (cf. Banks & Robins 2002). Klüpfel's report does not dwell on the geological framework for the island but starts with a review of the prevailing climatic conditions to derive a simple water balance estimate for an ideal 1 km2 catchment. Using the approximation that annual precipitation is split equally between (actual) evaporation, runoff and groundwater recharge (so contrasting with a modern understanding of the split, as discussed in ‘British hydrogeological investigation postwar’, below), the water balance for a ‘normal’ year and a ‘dry’ year is given as in Table 1.
Water balance for normal and dry years
The January 1942 report continues with a detailed review of the occurrence of groundwater in which fracture porosity and permeability are separately identified from matrix storage and matrix transmissive properties. The report also discusses groundwater quality with regard to potential marine influence and the danger of sea water intrusion. It concludes with some observations on groundwater abstraction potential.
The accompanying map (Fig. 11) keys 12 discrete geological units (Table 2), which equate to hydrogeological types. These units correspond essentially with the modern Jersey Shale Formation (Fig. 2), granites, quartz-porphyry and porphyry (i.e. components of the Jersey Volcanic Group), conglomerate and the thickest Quaternary cover deposits. Additionally, it highlights ‘barren regions’ and potential water engineering sites as well as sites of drilling completed to date.
Translation of the key to Oberleutnant Walther Klüpfel's 1942 ‘First draft of a water map for the island of Jersey’, with colour or symbol if discernable on Figure 11
On 25 October 1942 Klüpfel issued a final report on water supply (Rose 2005a). This is one of the very first detailed reports to accompany a hydrogeological map of hard rock terrain for any part of the British Isles. Although hydrogeological descriptions had been prepared for major sandstone and Chalk aquifer systems in England and elsewhere, none had previously been attempted for a comparatively low-yielding fractured crystalline rock aquifer.
Two new maps were completed in November to accompany the final report. One (at a scale of 1:50 000) provided a synthesis of the earlier map in which areas were simplified into just ‘inland areas with shallow groundwater’, ‘areas where groundwater is deeper but may be accessed by drilling’ and ‘areas of coastal sands’ (Rose 2005c, fig. 3). This was presumably a planning guide to be used by German drilling crews and engineers when considering groundwater abstraction as a means of water supply. The other was a companion to this map at the original 1: 25 000 scale (Rose 2005c, fig. 4) which showed supporting features, remarkably including ‘water discharge areas to the sea’ and ‘channelling of groundwater flow to valleys’.
Klüpfel demonstrated a good understanding of the weathered basement aquifer on Jersey. His approach was constrained by working in two dimensions and nowhere does he appear to have considered the three-dimensional configuration of the aquifer, groundwater flow being a horizontal flowpath along the trace of a valley on a two-dimensional map. Nevertheless, his seminal work showed for the first time in the British Isles that:
Shallow weathered crystalline basement could indeed be an aquifer which contains groundwater, albeit an aquifer of limited storage potential.
The hydraulic character of the various basement lithologies could be categorized according to recognized individual properties.
Areas of different groundwater availability potential could readily be identified.
The maps relied on geolines (map lines dividing different geological units) as the source of classification of groundwater occurrence on Jersey, much like the modern hydrogeolgical map of the island (British Geological Survey 1992). What Klüpfel could not do was apply the now widely accepted approach of quantitative analytical hydrogeology supported by models that synthesize aquifer processes. Klüpfel, nevertheless, was among those who pioneered the way towards geoline-based classification of groundwater availability that would later underpin the analytical understanding of aquifers. Although this is a significant outcome in the development of the science of hydrogeology, the more pressing practical outcome of support to the German fortification of the island was also achieved. His understanding of the aquifer was able to underpin groundwater abstraction for supply as well as dewatering for tunnel and trench work.
Walter Wetzel, meanwhile, working independently, remotely developed a water supply map for Guernsey (Rose 2005c, fig. 8), also at 1:25 000, to accompany a brief report of November 1941. This showed all the known water supply sources and distribution mains while alluding to groundwater flow being focused along valley bottoms. An important difference for the two workers was that Klüpfel had local access to more published geological maps and geological descriptions for Jersey than were available to Wetzel for Guernsey. Another important difference was that Guernsey was then marginally better endowed with water than Jersey, Guernsey relying in the 1930s almost entirely on a small group of large-diameter wells for public supply (Hawksley 1977).
Walter Wetzel's work for Guernsey was also supportive of the German fortification programme. However, in November 1941 one site was given absolute priority in the proposed 14-month construction programme: the Battery ‘Nina’, soon to be re-named Battery ‘Mirus’ (Partridge & Wallbridge 1983) (Figs 8 & 10). First surveys to site a heavy naval battery were conducted in the late summer of 1941, and from November work began in earnest on an extensive arc of land. Ultimately, some 47 000 m3 of concrete were to be consumed in the construction of this battery position by mid-1942, an amount that alarmed the local German command and proved a significant drain on material and manpower. To facilitate development of the site, reports on potential water supply were completed by TKVR Röhrer on 19 December 1941 and TKVR Scherer on 17 January 1942. Röhrer suggested three possibilities, including shallow wells, Scherer pumping from the nearest spring, but water supply was still a problem as the battery became fully operational, requiring a further report (by Wehrgeologenstelle 4) in June 1942 that unsuccessfully widened the search for a suitable source of water.
Wehrgeologenstelle 4 arrived on Guernsey in April 1942, but did not complete reports on potential water supply on the island as a whole until late in the year (Rose 2005a, table 2). A brief report was completed on 23 September, the main report not until 21 December, seemingly just prior to its departure from the islands (Rose 2005a, table 2). The reports contain a comprehensive spring and well inventory listing 279 sites, including well depth, normal yield and location, plus a number of wells with yields in excess of 10 l s−1. Interestingly, many of the wells, featuring wind-pumps, some even motor or electric pumps, subsequently fell into disuse as they were not located in a subsequent survey carried out by Robins et al. (2002).
While the reports were in preparation, effort focused on mapping the existing water sources along the coastal belt, including all boreholes and wells. Two groundwater maps at 1:25 000 completed in September 1942 (e.g. Fig. 12) showed well depth and salinity in the region of projected coastal fortification. Well locations were mapped and colour-coded according to three or four classes of depth to groundwater and qualitative water salinity (saline or fresh). Another contemporary map (Rose 2005c, figs 10 & 11) was generated from these to illustrate groundwater conditions by colour-coding coastal regions according to the three different categories of groundwater depth, or regions of saline groundwater. A final map in this 1:25 000 set (Rose 2005c, fig. 12) was circulated later, in December 1942. This indicated potential water supply for the island together with the location of existing features including springs and wells. This work greatly assisted the development of fresh groundwater sources to support engineering work and potentially allowed anti-tank trenches to be sited in areas that were least likely to flood with rising groundwater in winter, minimizing risk of damage to the trenches.
Groundwater map (Grundwasserkarte) of Guernsey, original at a scale of 1:25 000, compiled in September 1942 by Wehrgeologenstelle 4 during German occupation of the Channel Islands. Key to the six symbols at the right of the map indicates (from the top down) wells with water at depths of <2; 2–7; >7 m; springs; wells with saltwater; dry wells. From Rose (2005c); reproduced by permission of the British Cartographic Society, and the Bundesarchiv–Militärarchiv, Freiburg-im-Breisgau.
Wetzel remotely generated a water supply map for Alderney (Rose 2005c, fig. 20) as for Guernsey, but at a scale of 1:10 000, consistent with the best available topographic map. Compiled by October 1941 on the basis of published data, this showed the location of wells and boreholes, surface waters and small dams. No meteorological records were available: for Alderney these were begun only in 1955, unlike Guernsey, where records had been maintained for a much longer time (Anon. 1963).
Water supply on Alderney was seemingly a major problem. It was investigated by the military geologist Dieter Hoenes in 1942 on brief assignment from Wehrgeologenstelle 7 (Rose 2005a), prior to his appointment to Wehrgeologenstelle 4. Subsequently, Wehrgeologenstelle 4 issued three out of ten ‘expert opinions’ (on 27 May, 15 August and 15 December) and six out of 11 additional reports (on 26 April, 10 May, 18 May, 23 May, 7 July and 25 November) dealing specifically with Alderney's water supply. Hoenes looked at likely transfer potential for water to the major prison camp and at potential sites to develop infiltration galleries in the Quaternary sand aquifers to supplement supply. However, no galleries were constructed and groundwater investigation and development largely ceased on Alderney when Wehrgeologenstelle 4 was withdrawn from the Channel Islands towards the end of 1942, following the departure of Fortress Engineer Command XIV and Fortress Engineer Staff 14. The focus for hydrogeological investigation turned towards other fronts.
Although there is evidence that a geological appraisal was made of Sark (Fig. 13), the island was not significantly fortified, and seems not to have needed hydrogeological investigation at this time.
German geological map of Sark and Brecqhou, original at a scale of 1:25 000, preserved in file RW35/72 at the Bundesarchiv–Militärarchiv, Freiburg-im-Breisgau, and published with permission.
From the number of ‘hydrogeological’ maps and reports, and the expertise deployed to compile them, it can reasonably be inferred that they were intended for significant use by the German forces occupying the Channel Islands. However, these troops systematically destroyed most of their records before surrendering in May 1945 (Cruickshank 1975), so details of how the maps and reports were actually applied is not known. The maps and reports themselves were among the documents destroyed locally, but fortunately duplicate copies were sent to superior military engineering or geological staffs on the mainland of Europe, and it is these that have been discovered and described recently (Rose 2005a⇓–c, 2007, 2008; Robins & Rose 2005; Rose & Willig 2009).
British hydrogeological investigation postwar
Postwar, an off-island consultancy firm was commissioned to investigate groundwater potential for public supply in Jersey and Guernsey (Hawksley 1976, 1977). Otherwise little further study of groundwater was made until drought in the years 1988 to 1991 provided the stimulus for a major investigation under the auspices of the British Geological Survey (British Geological Survey 1992; Robins & Smedley 1998; Robins et al. 2002).
These studies were concluded in the belief that significant records of the German investigations had not survived the war (Robins & Rose 2005). They generated a new ‘Hydrogeological Map of Jersey’ (British Geological Survey 1992). This, unlike Walther Klüpfel's January 1942 ‘Water Map for the island of Jersey’ (Fig. 11), which defined water-bearing rock on a typology basis and which was also at a scale of 1:25 000, used the UNESCO International Map Legend (UNESCO 1983). The new map thus characterizes groundwater bodies according to whether they are intergranular flow dominant, that is, the Quaternary cover; fracture flow dominant, that is, much of the bedrock; or weakly to negligibly permeable, that is, Klüpfel's ‘mostly barren’ type. Despite the different approaches, both Klüpfel's map of 1942 and the new map of 1992 are strikingly similar (Robins & Rose 2005). Both, for example, identify the rhyolites as the least productive unit of the basement rocks.
Blackie et al. (1996) derived a long-term average water balance for Jersey using modelled data for the years 1968 to 1998. Robins et al. (2002) later transposed this model to Guernsey (Table 3).
Long-term average water balance
Similar models to that of Blackie et al. (1996) can be applied on Alderney and Sark. However, Davies (1998) had to make the assumption on the smaller island of Sark that a perceived loss of 74% of the effective rainfall (P–AE) was discharged over the beach as shallow groundwater throughflow. This loss reflects the increasing role of discharge to the sea with reducing island area (see Falkland 1991).
The British work established that the largest island, Jersey, offers broadly the same hydrogeological constraints as found also on Guernsey (Fig. 14), Alderney and Sark. The main differences between the islands depend on the respective physiography of each. Groundwater is unconfined although it may occur under a confining head within individual fractures. On Jersey, groundwater flow is principally from north to south with the main discharge area along St. Aubin's Bay on the mid south coast. On Guernsey, groundwater flow is dominated by the higher ground in the south causing short flow paths to the south coast and longer slower pathways particularly across the central low plateau towards the NW coast. Alderney and Sark both offer a radial pattern of groundwater flow towards the coast. The hydraulic conductivity of the weathered aquifer in the Jersey Shale Formation lies in the range 10−3 to 1 m/d (Robins & Smedley 1998), while that of the crystalline basement on Sark could be as low as 10−4 m/d (Davies 1998; Cheney 2004).
Schematic cross-section showing the groundwater flow system in Guernsey as currently understood. From Robins & Rose (2005), courtesy of the Geological Society of London.
The shallow nature of the weathered zone creating the bedrock aquifer on the respective islands provides a valuable moderating feature that protects the systems from excessive and harmful over-abstraction. Both the hydraulic conductivity and storativity decrease with depth beneath the water table as the fractures tighten with the weight of overburden. The ability for the aquifer to drain as baseflow and to discharge to pumped boreholes diminishes as the water table falls during dry periods, so providing a critical mechanism for self-protection and healing. This effect is minimal in the rhyolite formations, which do not tend to weather and fracture to the same extent as other rock types.
The main island aquifer on Jersey is a shallow zone of weathering, typically only 25 m in thickness below the water table, within ancient igneous and metasedimentary rocks. There are also thin coastal sand aquifers along the west and parts of the east coasts of the island. A typical sustainable borehole yield is about 0.5 l s−1. The highest known yield from bedrock is 4 l s−1, but this is exceptional.
About half the long-term available recharge, or renewable resource, is currently used either as groundwater abstraction or as baseflow in surface water. In dry years, water levels fall, baseflow declines and many boreholes go dry, particularly on higher ground. The overall water balance is little affected by irrigation returns and interception of rainwater by hardstanding, and these in any case tend to counteract each other.
Groundwater quality on Jersey is characterized by oxidizing waters of Na–Ca–HCO3 or Na–Ca–Cl type, although some samples have a high SO4 concentration. Around 80% of the groundwaters sampled on the island by Robins & Smedley (1998) were acidic (pH<7), and most were undersaturated with respect to calcite. Most groundwaters are oxidizing, but reducing waters occur in the south and SE of the island towards the main discharge zone into St. Aubin's Bay, reflecting upwelling of deeper and longer groundwater flowpaths from the north of the island (Green et al. 1998). All the groundwater is susceptible to surface pollutants derived essentially from intense agricultural activity. Many Jersey groundwater samples have nitrate concentrations in excess of the European Community maximum admissible concentration for drinking water of 11.3 mg N l−1 and some sources also periodically exceed the limit for K and NH4.
The groundwater system on Guernsey comprises the same shallow weathered zone in ancient crystalline rocks as on Jersey. Discharge from storage in this zone is critical to the maintenance of baseflow during prolonged dry periods. Groundwater is moderately mineralized on Guernsey, with similar high nitrate concentrations to Jersey, with a mean reported by Robins et al. (2002) of 14.1 mg N l−1.
Alderney is slightly different from the other islands in that it contains a Cambro-Ordovician sandstone as well as ancient crystalline rocks. However, the sandstone is cemented and offers little primary permeability so that a shallow weathered and fractured zone provides the only real potential for groundwater storage.
The granodiorites and gneisses of Sark are similar in hydrogeological character to the ancient rocks found elsewhere in the Channel Islands. Again, groundwater storage and flow occurs entirely within discrete fractures separated by a rock matrix of negligible permeability. The average depth of boreholes on Sark, c. 36 m, reflects the weathered shallow fracture zone thickness. Davies (1998) calculated that a hydraulic conductivity of 3× 10−4 m/d would be compatible with the observed potentiometric surface within the island. Water level monitoring shows that selected boreholes respond rapidly to rainfall with water level fluctuations of several metres. Reaction in other boreholes is more subdued with a range of less than one metre (Cheney 2004). Variation in response reflects the fractures penetrated, that is, the degree of fracturing and connectivity with the surface. The potentiometric surface forms a subdued reflection of surface topography and is shallowest beneath the central plateau.
Discussion
The three largest of the Channel Islands have been heavily fortified, by the British from the early 13th to mid-19th centuries and by the Germans in the 1940s. A secure water supply, ideally from groundwater, was a necessary requirement for those installations built to withstand siege. Thus, Mont Orgueil Castle, the earliest fortress on Jersey, had wells dug within its Middle and Outer Wards: that in the Middle Ward can still be seen (Rybot 1978). The castle withstood a French siege in 1373, whereas Grosnez Castle, which lacked a secure water supply, was quickly captured. Mont Orgueil was superseded by Elizabeth Castle for tactical reasons from the late 16th century, to defend the island's principal port, at a relatively safe distance from potential siege cannon. However, although Elizabeth Castle had a well and underground cisterns for storage of rainwater, its position on a rocky islet in St. Aubin's Bay gave little scope for groundwater abstraction. Lack of water was always a problem: additional supplies had to be shipped or (from 1874) piped in (Rybot 1986). Elizabeth Castle was in turn superseded early in the 19th century by Fort Regent, on a hill above the port, which had a far more effective well (Davies 1971).
The German Army built its fortifications of reinforced concrete rather than stone, their outer walls typically 2 m thick, requiring a great deal of water (ideally but not necessarily fresh water) as well as cement for their construction. Irrigation water was a requirement to maximize local agriculture, and potable water to supply camps providing the construction workforce and troops manning the completed fortifications as well as the remaining civilian population. German military geologists deployed on the islands were therefore involved in intensive hydrogeological investigations as well as assessments of sand and aggregate as construction materials, and site investigations, especially for the excavation of underground facilities. Data compiled by German geologists in the early 1940s, together with those from more recent investigations by off-island consultants (Hawksley 1976, 1977) and subsequent more comprehensive investigation guided by the British Geological Survey (Robins & Smedley 1998; Robins et al. 2002), provide detailed information that extends over a time range (65 years) that is unusually long for basement aquifers in the British Isles.
In the UK, almost no investigation of hard-rock aquifers had been undertaken at catchment scale prior to World War II (Mather 2012b): hard rock areas in the north and west of the country were adequately supplied by surface waters and there was little need to develop groundwater. Understanding of the hydrogeology of the British Isles was limited to studies of major aquifers such as the Cretaceous Chalk of southern and eastern England and the Permo-Triassic sandstones of the English Midlands. The investigations of German geologists during the occupation of the Channel Islands were thus both innovative and, at that time, unique for the British Isles. Pre-war British investigation into hard-rock aquifers in the UK had focused primarily on engineering issues such as the need to dewater metal mines in Cornwall, Wales and elsewhere (Younger 2004). British geologists pioneered exploratory work on basement rocks in Africa prior to World War II (Dixey 1931), but apart from work undertaken by geologists developing groundwater sources in Scandinavia (well advanced by the early years of the 20th century) and German or Austro-Hungarian military geologists studies in World War I (Willig & Häusler 2012), medium-scale regional mapping of European hard-rock hydrogeology in Lower Palaeozoic and Precambrian basement aquifers did not become of major significance until the 21st century, with the advent of the Water Framework Directive (Council of European Communities 2000).
Much of the German military effort remained largely unknown until applied hydrogeology had developed through the postwar years to a point that had overtaken the understanding of the hard-rock aquifer processes and mechanisms demonstrated on the Channel Islands in the early 1940s. Similarly, German work was not available to guide later hydrogeological studies of Jersey and Guernsey. It remained largely undetected until these studies were essentially complete (Robins & Rose 2005). Of course, none of the German workers had at his disposal the technical advancements that we enjoy today. As a consequence, the water balance derived for Jersey grossly overestimated both runoff and recharge. Nevertheless, the German technical accomplishments coupled with vision of how a shallow hard-rock fractured aquifer works would have provided a significant enhancement to applied hydrogeological understanding had they been published.
Conclusions
The Channel Islands are one of the classical areas of British geology, studied intermittently for some 200 years, their bedrock revealing evidence for a range of Precambrian events and spectacular outcrops of calc-alkaline igneous rocks, the thin Quaternary cover changing climates and relative sea levels from the Middle Pleistocene to the present day.
A high density of coastal fortification constructed over a 700-year timespan is a major feature of the terrain on the three largest islands. British stone-built castles, forts, gun emplacements and towers protect most beaches deemed vulnerable to sea-borne assault, particularly in the 18th and 19th centuries, supplemented by German anti-tank walls, casemates, gun positions, observation towers, infantry strong points and a variety of bunkers, mostly constructed of reinforced concrete with outer walls up to 2 m thick during occupation of the islands by German troops between mid-1940 and May 1945.
German military geologists initiated detailed hydrogeological studies and hydrogeological mapping of the islands to assist fortification, primarily by the provision of potable water for the construction workforce and their garrison troops, whose numbers significantly increased the island population during years of low rainfall, but also water for agriculture and the preparation of concrete. Their maps and reports, now known to be preserved as archive documents in Germany, the USA and the UK, although seemingly all destroyed on the islands themselves prior to German surrender at the end of the war, reveal an understanding of low-storage fractured hard-rock aquifers and the processes and mechanisms that controlled recharge and throughflow within them that was in advance of such understanding elsewhere within the British Isles at this time.
Military use of hydrogeology by German geologists has provided a detailed database for comparison with observations made during subsequent groundwater studies in the Channel Islands. All of their maps and all but one of their reports remained unknown to British hydrogeologists operating in the 1970s and then more intensively in the 1990s, but their relatively recent near-complete rediscovery now facilitates an appraisal of the heavily stressed aquifers on these small, hard-rock islands over an unusually long (65 year) timespan. Fortunately, the inferences drawn by British hydrogeologists in the 1970s and 1990s, the last 30 years, are compatible with those of German military hydrogeologists of the 1940s, representative of the first 30 years. Annual rainfall has been declining on the islands, at least on Guernsey, since the 1940s, and although the available surface and groundwater has sustained the population even in times of drought in the past, it may not do so in the future.
Acknowledgments
We gratefully acknowledge owners of copyright for permission to reproduce illustrations, as duly acknowledged in the figure captions, and constructive comments by John Mather, John Renouf and an anonymous reviewer. Historical details not given a literary source in the text have been gleaned at sites on the islands where local information has been put on display, typically by the respective island Tourist Boards.
- © The Geological Society of London 2012
References
Basement hydrogeology and fortification of the Channel Islands: legacies of British and German military engineering
