1. Characterization of Surface Waters, Wetlands and Groundwater

1. Characterization of surface waters, wetlands and groundwater

1.1 River basin

1.1.1 Landscape and soil types Figure 1.1.1 Topography of Odense Odense River Basin encompasses an area of ap- River Basin. prox. 1 046 km2 and includes approx. 1 100 km Source: National Sur- of open watercourse and 2 600 lakes and ponds vey and Cadastre and (>100 m 2). The present landscape of Fyn was Fyn County. primarily created during the last Glacial Period Height above Danish Zero 11.500 to 100 000 years ago (Figure 1.1.1). The 0-10 m most common landscape feature is moraine plains covered by moraine clay that was depos- 10-20 m ited by the base of the ice during its advance. The 20-30 m meltwater that ﬂ owed away from the ice formed 30-40 m meltwater valleys. An example is the Odense ﬂ oodplain, which was formed by a meltwater 40-50 m river that had largely the same overall course as 50-60 m today’s river. The clay soil types are slightly dominant and 60-70 m encompass approx. 51% of the basin, while the 70-80 m sandy soil types cover approx. 49% (Figure 80-90 m 1.1.2). The moraine soils of Fyn are particularly well suited to the cultivation of agricultural 90-100 m crops. Agriculture has therefore left clear traces 100-110 m in the landscape. Deep ploughing, liming and 0 5 10 km the suchlike have thus rendered the surface soils Over 110 m more homogeneous. Watercourse

Figure 1.1.2 1.1.2 Land use, population and Soil conditions in wastewater Odense River Basin. The dominant soil Just as elsewhere in Denmark, land use in types at a depth of 1 m Odense River Basin is dominated by agricultural are shown. Mapping exploitation (Figure 1.1.3). Farmland thus ac- carried out by Danish counts for 68% of the basin. Of the remainder, Institute of Agricultur- approx. 16% is accounted for by urban areas, 10% al Sciences 1975–79. by woodland, and 6% by natural/seminatural ar- Coarse sandy and eas (meadows, mires, dry grasslands, lakes and fine sandy soil wetlands, which are protected by Section 3 of the Clayey sand soil Protection of Nature Act). The corresponding Sandy clay soil ﬁ gures for farmland, woodland and natural/seminatural areas for Denmark as a whole are 62%, Clay soil

11% and just over 9%, respectively. Special soil type The population of Odense River Basin num- bers approx. 246 000, of which approx. 182 000 Humic soil inhabit Odense, which is Denmark’s third largest Unclassified: city. Approx. 90% of the population in the basin discharge their wastewater to a municipal waste- Urban areas water treatment plant. The remaining 10% of the Woodland population live outside the towns in areas with- Inland waters (lakes, out access to sewerage. There are a total of 6 900 0 5 10 km watercourses residential buildings in these sparsely built-up Residual areas areas outside the sewerage system catchments.

Odense PRB Odense Pilot River Basin 11 1.1 River basin

Figure 1.1.3 seepage. Many of the dairies and abattoirs were Land use in Odense subsequently closed down due to centralization, River Basin. and serious efforts were initiated to treat urban Source: Areal Infor- mation System (AIS). wastewater. Measures of change were not made Ministry of the Envi- until the 1980s and early 1990s when treatment ronment. of urban and industrial wastewater was considerably improved, and unlawful discharges of Urban areas silage juice etc. by the agricultural sector were Cultivated land Woodland stopped. Inland waters (lakes, watercourses) Natural contryside (mires, meadows, etc.) 1.1.3 Agricultural conditions

In 2000, there were approx. 1 870 registered farms in Odense River Basin, of which approx. 960 were livestock farms. Livestock production in the basin amounts to approx. 60 000 livestock units (LU) (1999–2002), consisting of 59% pigs, 37% cattle and 4% other livestock. Livestock density averages 0.9 LU/ha farmland, corresponding to the national average, but varies to some extent within the individual catchments that comprise 0 5 10 km Odense River Basin (Figure 1.1.4). Overall, livestock production in the basin has increased in recent years. This masks a production decrease in Figure 1.1.4 dairy farming and a marked increase in pig farm- Livestock density in ing. Based on the applications for permission to the river basins and increase herd size submitted to Fyn County (EIA local water body catch- screening) it is concluded that livestock produc- ments of Fyn County tion will continue to increase in the coming in 2002 expressed in years. Thus in the period from 2000 to mid 2003, relation to the area of permits have been granted to increase livestock farmland in the catchments (LU/ha). production in the basin by 17%. Source: Central Live- The dominant crop in Odense River Basin is stock Register under cereals (2/3 winter cereals), while only 10% is ac- the Ministry of Food, counted for by grass/green fodder (Figure 1.1.5). Agriculture and Fish- The concentration of market gardens in the basin eries. is relatively high, occupying approx. 3% of the 1.4 and above farmland. 1.2-1.4 1-1.2 0.8-1 Historical development – Artifi cial drainage 0.6-0.8 and land reclamation in Odense River Basin 0.4-0.6 0-0.4 Over the past 50–100 years, artifi cial drainage no information has been established on an estimated 55% of the cultivated land in Odense River Basin. The aim has been to ensure rapid drainage of the arable 0 5 10 km land and to optimize the possibilities for cultivation. In addition, mires, meadows, watercourses Due to increasing industrialization and the and shallow lakes and areas of the fjord have spread of fl ushing toilets, poorly treated waste- undergone major physical changes or have been water from towns, dairies, abattoirs etc. started erased from the map due to reclamation of the to be discharged into the water bodies of Fyn at land for cultivation. Thus 72% of the major ar- the beginning of the 20th Century. In the 1950s, eas of meadow and mire have disappeared from the agricultural sector also started to cause seri- Odense River Basin over the past 100 years ous pollution of the aquatic environment by dis- (Figure 1.1.6). Much of this has been turned into charging silage juice, liquid manure and manure farmland through regulation of watercourses

Odense 12 PRB Odense Pilot River Basin 1.1 River basin

Root crops Market gardens Figure 1.1.5 and regular watercourse maintenance. Moreover, 3% Permanent grass 5% Crop distribution on many watercourses and ditches have been piped, 4% and many watercourses have been straightened. Spring cereals farmland in Odense Grass/green fodder 23% River Basin (2000). 10% A corresponding reduction has taken place in Source: Danish Plant the number of minor lakes and ponds. All in all, Pulses Directorate. decades of land reclamation and drainage of wet- 2% lands and farmland have considerably reduced the natural self-cleansing ability of Odense River Seed crops Basin. Speciﬁ c knowledge of currently drained 8% farmland in Odense River Basin is lacking. This knowledge is important for determining the pressure placed on the environment by the individual agricultural areas, and is necessary for Winter cereals future prioritization of environmental initiatives 45% in the agricultural areas.

Historical development – farming in Fyn County Figure 1.1.6 Danish agriculture has undergone considerable Distribution of fresh- structural development since the 1950s. Among water meadows, salt other things, holding structure has become more marshes and mires in specialized, and livestock herds have become Odense River Basin concentrated. Since the 1950s, the number of in 1890 and 1992. Pre- pared by Fyn County holdings in Fyn County has decreased by 75%, on the basis of maps while their average size has increased by 250%. from 1890 (1:20 000) Diversifi ed farming with mixed livestock and and the National Sur- crop production has increasingly been replaced vey and Cadastre map by specialized cattle or pig holdings or exclu- from 1992 (1:25 000). sively crop holdings. Watercourses are not Crop distribution in the agriculture of the indicated on the map. 1950s refl ected the diversifi ed holding structure, with mixed livestock and crop production. At Distribution in that time, up to half of the farmland was cultivat- 1890 ed with grass, green fodder and root crops, while 1992 the other half was cultivated with cereal crops, primarily spring cereals. Among other things, agricultural development has entailed a more than 50% reduction in the area of grass, green fodder and root crops since the 1950s, while the area of cereals has increased. At present, only ap- 0 5 10 km proximately 18% of farmland is cultivated with grass, green fodder and root crops, while more than 60% is under cereals (Figure 1.1.7). The reduction in the area under grass, green fodder to an increase in the pig herd, while the cattle and root crops to the benefi t of cereal crops has herd continues to decrease in size. increased leaching of nitrogen due to the fact Due to the great specialization the livestock that the long growing season for grass and green herd – in contrast to previously – is much more fodder crops renders them better able to retain unevenly distributed between the individual the nitrogen than cereals. Since the 1980s, the holdings. The pig herd in particular has become proportion of the land under winter cereals has concentrated in holdings with very large herds. increased at the expense of spring cereals. The livestock herd and hence manure and the From the 1950s to the end of the 1980s the application of it have thus become concentrated livestock herd declined in size by 35% due to a on a far smaller area than previously. Due to new decrease in the cattle herd. Since the end of the legal requirements pertaining to utilization of 1980s, the size of the herd has begun to increase manure, however, it is once again being applied again (Figure 1.1.8). The increase is attributable to a greater area in recent years.

Odense PRB Odense Pilot River Basin 13 1.1 River basin

Figure 1.1.7 Crop distribution 1980-2002 Prior to the 1950s, manure was the chief % Trend in crop distribu- source of nutrient input to the crops. From the tion in Fyn County – 1950s onwards, the use of commercial fertilizer (1980 2002). 80 became increasingly widespread. Consumption Source: Statistics Den- mark. of commercial fertilizer increased steadily up 60 through the 1960s and 1970s, leading to a marked Set-aside increase in total nitrogen input to the crops of Permanent grass 40 86% over the period 1950 to 1980. Total nitrogen Pulses input remained high during the 1980s (Figure Root crops 20 1.1.9). Approx. 63% of the total nitrogen input Grass/green fodder was accounted for by commercial fertilizer, and Plantation 0 lllllllllllllllllllllllllllllllllll this alone was sufﬁ cient to meet crop nitrogen Seed crops 80 82 84 86 88 90 92 94 96 98 00 02 requirements. This reﬂ ects the fact that the ferti- Spring barley lizer value of manure was considerably underes- Winter cereals timated at that time. Livestock density 1980-2002 From 1992/93, nitrogen consumption began LU Figure 1.1.8 to fall. Among other reasons, consumption of Trend in the livestock 250000 commercial nitrogen fertilizer fell as a result of herd in Fyn County a number of legal measures aimed at reducing – (1980 2002) expressed 200000 pressure on the environment from agriculture, in terms of livestock e.g. requirements on improved utilization of units (LU). 150000 the nitrogen content of manure. Consumption Other of commercial fertilizer continued to decrease 100000 Cattle up through the 1990s until the present day. In Pigs 50000 all, total consumption of commercial nitrogen fertilizer has decreased by approx. 20% since its

0 lllllllllllllllllllllllllllllllllll peak in the 1980s. 80 82 84 86 88 90 92 94 96 98 00 02 The increasing consumption of commercial fertilizer up through the 1960s and 1970s also entailed an increase in phosphorus input to the Tonnes Nitrogen consumption 1979/80-2001/2002 crops, which increased by 20% from the mid Figure 1.1.9 N 1950s to the mid 1970s. Input of phosphorus far Trend in fertilizer 50000 exceeded the amount of phosphorus removed nitrogen consump- with the crops, resulting in phosphorus accu- tion in Fyn County 40000 mulation in the soil. From the beginning of the (1980 –2002). 1980s, the total input of phosphorus began to de- Source: Statistics Den- 30000 mark and Fyn County. crease due to a fall in consumption of commercial phosphorus fertilizer, but input continued to ex- 20000 Applied N ceed removal with the crops. From the mid 1980s N in commercial fertilizer to the present day, total phosphorus input to the N in manure 10000 crops in Fyn County has decreased by 28%. 0 lllllllllllllllllllllllllllll The major source of phosphorus input to 81/82 85/86 89/90 93/94 97/98 01/02 crops today is manure, accounting for approx. 72%. The amount of phosphorus in the applied manure roughly corresponds to the average Hkg Cereal yield 1980-2002 amount removed with the crops. As the livestock Figure 1.1.10 /ha herd is now concentrated on fewer holdings with Trend in average cereal very large herds, application of the manure is yield in Fyn County 60 concentrated on a far smaller area of farmland (1980 –2002). than previously. As a consequence, land fer- Source: Statistics Den- tilized with manure is particularly subject to mark. 40 overfertilization with phosphorus. Continued overfertilization with phosphorus will increase

20 phosphorus loading of the aquatic environment from agricultural land. The marked increase in fertilizer consumption

0 lllllllllllllllllllllllllllllllllll in the 1960s and 1970s is reﬂ ected in increasing 80 82 84 86 88 90 92 94 96 98 00 02

Odense 14 PRB Odense Pilot River Basin 1.1 River basin

River Odense at Bro- byværk near the village of Broby.

Photo: Jan Kofod Winther

cereal yields (Figure 1.1.10). The transition to year. After a fall in the mid 1970s, consumption cultivation of winter cereals, the use of high-yield increased again to peak at 7.5 million kg active varieties, improved cultivation techniques and in- substance in 1982/83, whereafter it fell steadily creased use of pesticides have helped increase the to 3.5 million kg in 2001. The reduction in con- cereal yields. sumption of active substance has been accompa- The agricultural sector is the greatest consumer nied by an increase in their potency such that the of pesticides in Denmark. Statistical information real pesticide application frequency (number of on pesticide consumption in agriculture is only applications per crop) has only decreased slightly available at the national level, and the following (Bichel Commission, 1999). The application fre- information thus pertains to Danish agriculture quency in 2000 and 2001 was only approx. 15% in general. The upturn in agricultural consump- less than at the end of the 1990s. tion of pesticides began in the 1950s, increasing Information on pesticide consumption at hold- ﬁ ve-fold from the mid 1950s to the beginning of ing level in Odense River Basin is not easily ac- the 1970s. In 1973, consumption amounted to cessible. Such information can be important in 6.7 million kg active substance corresponding to connection with planning measures in relation to 2.3 kg active substance per hectare farmland per water bodies and natural habitats in the basin.

Odense PRB Odense Pilot River Basin 15 Photo: Bjarne Andresen, Fyn County

Odense 16 PRB Odense Pilot River Basin 1.2 Climatic conditions

1.2 Climatic conditions

Precipitation and runoff Figure 1.2.1 Climatic conditions infl uence the state of na- N Average annual pre- ture and the environment. Thus variations in cipitation over Fyn precipitation and evaporation are important County (1961–1990). determinants of the amount of water that can Odense River Basin is fl ow towards the groundwater and watercourses, indicated. and variations in temperature and wind power can greatly infl uence the development of such phenomena as oxygen defi cit in the lakes and fjords. In warm, calm summers, nitrogen and phosphorus loading of these water bodies might cause particularly serious and protracted oxygen defi cit. When environmental state and trends are to be assessed it is therefore important to be able to differentiate between the variation in state attributable to climatic variations, and that attributable to the impact of anthropogenic pollution and other pressures. The annual precipitation in Fyn County varies considerably, typically being 200–300 mm 0 5 10 km greater over central Fyn than over the coastal areas (Figure 1.2.1). There is therefore a marked 550-600 mm 600-650 mm 650-700 mm precipitation gradient across Odense River Basin, 700-750 mm 750-800 mm 800-850 mm precipitation being highest in the upper (southern) parts of the basin and least near Odense Fjord. Over the period 1981/82 to 2001/02, aver- runoff is considerably greater than the variation age total annual precipitation for Odense River in precipitation. In summer the riverine runoff Basin varied from 400 mm to just under 1 100 is therefore typically only around 20% of that in mm (Figure 1.2.2, left), and riverine runoff to the winter months (Figure 1.2.3, right). the fjord varied correspondingly (Figure 1.2.2, In order to lead the water rapidly away from right). Particularly noteworthy is the very dry the typically clayey soils in wet periods, drain- year 1995/96, when riverine runoff was excep- age has been established in a considerable part of tionally low. In that “drought year”, the runoff the farmland in Odense River Basin. Although mainly consisted of water fl owing into the wa- subject to some uncertainty, it is estimated that tercourses from groundwater and of wastewater approx. 50% of the basin is drained. This has an discharged from the wastewater treatment plants (unknown) impact on the natural water cycle. in the basin. The average monthly precipitation in Odense Temperature and wind River Basin varied between approx. 40 mm The average air temperature in Fyn County (April) and 90 mm (December/January) (Fig- is 8.2ºC (1961–1990), and the typical variation ure 1.2.3, left). A large part of the precipitation during the year is illustrated in Figure 1.2.4. evaporates, especially in summer, and only a The wind usually blows from the west, and the minor share reaches the watercourses. As a power/energy with which the wind affects (stirs consequence, the variation in monthly riverine up) the water bodies varies (Figure 1.2.5).

Odense PRB Odense Pilot River Basin 17 1.2 Climatic conditions

Precipitation Riverine runoff mm mm Figure 1.2.2 Diffuse Precipitation over Wastewater 1000 500 Odense River Basin and riverine runoff to 800 400 Odense Fjord (annual values). 600 300

400 200

200 100

0 llllllllllllllllllllllllllll 0 llllllllllllllllllllllllllll 83/84 86/87 89/90 92/93 95/96 98/99 01/02 83/84 86/87 89/90 92/93 95/96 98/99 01/02

Precipitation Riverine runoff mm mm Figure 1.2.3 Diffuse Average monthly pre- 100 50 Wastewater cipitation over Odense River Basin and riv- 80 40 erine runoff to Odense Fjord (1981/82–2001/ 60 30 02). 40 20

20 10

0 0 JFMAMJJASOND JFMAMJJASOND

Air temperature 3 m / Summer wind energy 1976-2002 °C Max sec Figure 1.2.4 Kegnæs weather station Air temperature, Fyn Normal 1961-90 Normal June-August 20 500 County (monthly val- Min ues). 15 400

10 300

5 200 Figure 1.2.5 Wind energy in the 0 100 summer months

(June–August). -5 0 lllllllllllllllllllllllllllllllllllllllll JFMAMJJASOND 76 78 80 82 84 86 88 90 92 94 96 98 00 02

Odense 18 PRB Odense Pilot River Basin 1.3 The water cycle

1.3 The water cycle

Most of the precipitation that does not evaporate Figure 1.3.1 percolates to the groundwater, drains and water- Percolation from land courses. This amount of water is termed the per- N in Odense River Basin colation (Figure 1.3.1). Part of the precipitation (1999/2000). falls on paved areas, however, where it cannot Percolation percolate, but instead is conducted to watercours- mm es (stormwater discharges). Total percolation in 450-492 Odense River Basin amounts to an average of 298 400-450 mm (Table 1.3.1). The sum of percolation and es- 350-400 300-350 timated runoff from paved areas (306 mm in all) 250-300 corresponds very well to the total riverine runoff 200-250 (305 mm) calculated from measurements made in 150-200 100-150 the watercourses in the basin (Table 1.3.1). 50-100 The riverine runoff consists of wastewater (30 -1-50 mm) and of diffuse runoff to the watercourses (Table 1.3.1). The wastewater runoff consists of water that has been abstracted from the groundwater for various purposes (Table 1.3.2). The diffuse runoff can be roughly subdivided into two components:

•Groundwater runoff

0 5 10 km •Near-surface runoff.

The groundwater runoff consists of water that groundwater will correspond to the near-surface has been underway for a long time from falling as runoff, for example in the form of drain water precipitation to reaching the watercourses. The runoff (i.e. 305 mm – 30 mm – 91 mm = 184 mm) magnitude of this runoff can be assessed from (Table 1.3.1). the water ﬂ ow in the watercourses during the Most of the water that ﬂ ows in the watercourses “drought year” 1995/96 (91 mm), when the near- in Fyn County thus consists of water that reaches surface diffuse runoff – via drains etc. – was neg- the watercourses relatively quickly. If this water ligible. Assuming that the groundwater runoff is is polluted – for example via nitrogen loss from constant throughout the period, the difference cultivated land – the pollution will rapidly reach between measured total diffuse runoff and the the watercourses. Conversely, measures that re-

Freshwater input to Table 1.3.1 Category Transport Odense Fjord Water budget for Odense River Basin Precipitation 825 mm (1990/91–2000/01). Evaporation 519 mm

Percolation 298 mm

Paved areas (stormwater) 8 mm 306 mm

Riverine runoff (measured) 305 mm

Groundwater (=1995/96) 91 mm

Near-surface (diffuse) 184 mm

Wastewater (WWTPs, industry) 30 mm

Odense PRB Odense Pilot River Basin 19 1.3 The water cycle

Table 1.3.2 the Water Framework Directive in Odense River Category mm Water abstraction in Basin, fi rst and foremost to describe the temporal Odense River Basin Public waterworks 19 and spatial variation in water transport towards (1998). Odense Fjord, but also as a basis for assessing the Institutional wells <1 magnitude and variation in nutrient loading and transport in the basin. Farm and market garden wells, etc. 4 The water cycle in Odense River Basin is af- Industrial wells 3 fected by a number of human activities, for example abstraction of water for various purposes (Ta- Minor waterworks (1–10 households) 1 ble 1.3.2). Most of this abstraction is accounted Total abstraction 28 for by waterworks (20 mm). The share of water abstraction accounted for by farms and market gardens is relatively modest. Compared with water abstraction for these purposes in the sandy duce discharges from cultivated land will rapidly areas in western Denmark, the water require- affect water quality in the watercourses. ments of the agricultural sector in Fyn County A slightly different water balance for Fyn are considerably less. The abstraction carried out County as a whole has been published by Hen- affects water fl ow in watercourses. The effect is riksen & Sonnenborg (2003). In this case, the modest if assessed solely on the basis of the total groundwater runoff to watercourses is calculated annual runoff in watercourses, but water fl ow in to be relatively less and the near-surface runoff certain watercourses can be signifi cantly affected relatively greater than indicated in Table 1.3.1. during the summer when water fl ow in the wa- A dynamic water balance model for Odense tercourses is least. This aspect is examined more River Basin is currently being developed via closely in Section 4.2.1. cooperation between national sector research in- Finally, the water cycle has been considerably stitutions and Fyn County together with associ- affected by drainage and channelization of water- ated consultants. The results of this work will be courses in order to rapidly lead water away from utilized in the future work on implementation of the cultivated fi elds.

Odense 20 PRB Odense Pilot River Basin 1.4 Watercourses

1.4 Watercourses

1.4.1 Location, size and density is believed that it will hereafter remain necessary to operate with several levels of monitoring in or- Danish watercourses, including those in Odense der to achieve a suffi ciently adequate description River Basin, are typically lowland watercourses of watercourse status to enable prioritization of (terrain height under 200 m). In a European environmental initiatives and the selection of the perspective, they are all very small (Figure 1.4.1). most appropriate measures to improve water- The largest watercourse in Odense River Basin is course environmental quality. the River Odense (catchment 630 km2), the main reach of which is just under 60 km long and up to 30 m wide. The density of open watercourses 1.4.2 Typology in the basin as a whole is approx. 1.0 km/km2. The original (natural) density of the watercourse According to the WFD, the watercourses have network was probably somewhat greater (up to to be differentiated into types in order to es- 1.5-fold). However, the density is of the same tablish relatively uniform groupings for which magnitude as in the other Nordic countries and the natural reference conditions are relatively Great Britain (as well as in parts of the USA). homogenous. Annex II of the WFD provides In Odense River Basin, the very small water- two typology systems for differentiating the courses account for a considerable proportion watercourses, namely System A and System B. of the watercourse network (Table 1.4.1). Even As Danish watercourses are so small, the use of though the watercourse system is incomplete, System A in its present form is inappropriate. the relative distribution of the watercourses in Two other systems for differentiating Danish size categories does not differ very much from watercourses have therefore been tested. the expected natural distribution (Horton’s laws One of the systems in principle resembles Sys- of stream number and stream lengths; see Dodds tem A, but contains elements of System B (Table & Rothman, 1999). 1.4.2). This typology, which has been developed Due to resource constraints, the many small by the National Environmental Research Insti- watercourses are relatively under-represented tute, has so far only been notifi ed nationally. in the current monitoring (Table 1.4.1). They According to this system, Danish watercourses encompass both spring brooks, summer-dry are differentiated according to location east or (temporary) watercourses (which should be in- west of the line formed by the Weichsel ice front cluded as watercourses – and not as wetlands) and ditches established in connection with drainage of wetlands. Even though our knowledge of the Width 2 m 2–10 m >10 m Table 1.4.1 small watercourses is generally less than that of Size distribution of Watercourse the larger watercourses, the small watercourses 730 330 42 watercourses and length (km) monitoring stations in are considered to be a very important part of Percentage of Odense River Basin. 66 30 4 the watercourse network. Thus the small wa- total length tercourses together contain just as many species No. of monitoring 101 192 15 of macroinvertebrates as the larger watercourses stations (including also special spring species), just as Percentage of 33 62 5 many small watercourses are important spawn- stations ing and nursery grounds for salmonids. Given that the small watercourses are of Table 1.4.2 such great signifi cance, they, like the larger Typology of Danish watercourses, should also be included when watercourses proposed identifying water bodies (Section 1.4.3). As it is Type 1 2 3 by the National Envi- impracticable to investigate them all, it would be Stream order (Strahler ronmental Research In- 1–2 3–4 >5 appropriate to aggregate them into groups. Ag- system) stitute (M.L. Pedersen, personal communica- gregation could be carried out for each type (Sec- 2 Catchment area (km ) <10 10–100 >100 tion 1.4.2) according to the main threats/pres- tion). The watercourses in Odense River Basin Width (m) 0–2 >2–10 >10 sures. The monitoring and subsequent reporting lie east of the line could thereafter be restricted to a representative Distance to source (km) <2 2–40 >40 formed by the Weichsel fraction of the water bodies within each group. It ice front.

Odense PRB Odense Pilot River Basin 21 1.4 Watercourses

Figure 1.4.1 tested in part of Odense River Basin (Figure Map of Odense River 1.4.2) and is fully practicable when appropriate N Basin showing the calculation methods are used to defi ne catchment complete watercourse network. boundaries and to determine watercourse length. Watercourse width (if possible the winter width of the water surface, otherwise the summer width of the water surface) and the catchment area are the parameters that are easiest to work with. That stream order and distance to source are more diffi cult to work with is due to the fact that it can be diffi cult to locate the source (often piped) and to differentiate artifi cial reaches from natural reaches in the upper parts of the watercourse systems. For Danish watercourses, it has been shown that the composition of the plants, macroinvertebrates and fi sh is closely correlated with watercourse size (Fyn County, 2000a; Wa- terFrame, 2003). The macroinvertebrate fauna also correlates with the watercourse slope, which in turn is negatively correlated with watercourse size (Wiberg-Larsen et al., 2000). The second system (Table 1.4.3) is based in 0 5 10 km part on theoretical considerations, according to which the magnitude and variation of water fl ow and the watercourse slope considerably infl uence and thereafter according to size. The typology conditions for the fl ora and fauna. Thus many of each individual locality is assessed from four reaches can more or less regularly dry out in appropriate (and mutually positively correlated) summer, especially the very small watercourses size components. A locality is assigned a type and the upper reaches of the larger watercourses. in accordance with the dominance principle. In Furthermore, the slope is decisive for fl ow rate cases of equal distribution between two types, and the composition of the bottom substrate. the size of the catchment area is the decisive de- Runoff is described from the median minimum terminant of type. This principle is employed to value. For Fyn County as a whole this is deter- ensure that the system can be used throughout mined with an average density of 1 location per 5 Denmark irrespective of regional or local dif- km2. For Odense River Basin, the corresponding ferences in topography. The method has been density is 1 location per 4.4 km2. The project part- ner, DDH, has tested this method in the Lunde Stream system and concluded that the selected Table 1.4.3 Alternative type 1 2 3 parameters can be accurately determined, and An alternative typol- that the method is easy to use and not very time- ogy of Danish water- 2 Catchment area (km ) <10 10–100 >100 consuming (DDH, 2003). The relationship to the courses proposed by DDH (2003). In this Runoff, median min biological conditions still remains to be tested, 2 <0.2 0.2–2,0 >2.0 system the three vari- (l/s • km ) though. There is no doubt, however, that at least the naturally summer-dry watercourse reaches ables should be consid- Slope (m/km) <0.25 0.25–0.75 >0.75 ered independently of have reference conditions that differs markedly each other.

Table 1.4.4 Physical pressure Specification of extent Physical modiﬁ cations initially examined in Piped reaches km (and pipe diameter) the Ryds Stream catch- Other regulation (straightening, deepening) Grade 0–3 ment and the main Bank stabilization (stones/fascines) +/- course of the River Maintenance (weed cutting/dredging) Grade 0–3 Odense. Obstructions for fish Number (height and upstream damming effect) Stormwater discharges (hydraulic stress) Number (and frequency) Sand transport Grade 0–3 Drainage (flood plain) Grade 0–3

Odense 22 PRB Odense Pilot River Basin 1.4 Watercourses

Figure 1.4.2 N Map indicating the N watercourse typology for the Ryds Stream catchment and the upper parts of the River Odense system.

Type 1

0 1 2 3 km Type 2

Type 3 from that of permanently ﬂ owing watercourse Basin reaches. Moreover, the reference conditions of watercourse reaches with a naturally very gentle slope can be expected to differ markedly from that of those with a steeper slope.

1.4.3 Delineation of watercourse reaches

The complete method for delineating watercourse reaches was initially tested on the Ryds Stream 2 catchment (area 46 km ) and the main course of 0 5 10 km the River Odense (length 60 km) calculated from the source of Rislebæk Brook to the outlet in Odense Fjord (Figure 1.4.3). In addition, part of the physical conditions is available it is compared the identifi cation process (excluding assessment with the results of the biological monitoring. In of physical modifi cation) has been tested in the addition, the presence of different forms of na- Lunde Stream catchment (area 70 km2) (DDH, tional and regional protection is incorporated in 2003). Applying the criteria in the EU horizon- the assessments. Delineation of water bodies has tal guidance “Identifi cation of water bodies” to thus primarily been based on knowledge of the these three test areas together with the criteria watercourses’ current status assessed from the ex- for identifying heavily modifi ed water bodies isting knowledge about the physical and biologi- (Section 1.4.4) yields 13, 16 and 14 water bodies, cal quality. It should be noted that the biological respectively, i.e. an average of 1 water body per 4 status is primarily described from the existing km of watercourse. The individual water bodies Danish assessment system, whereby macroin- vary in length from 1 to 12 km. If the present vertebrates are utilized as “indicators” for the level of detail is maintained, an estimated 280 biological conditions in general (Danish Stream water bodies should be identifi able in Odense Fauna Index; Skriver et al., 2000). The status is River Basin solely from the watercourse systems. measured as fauna classes on a scale from 1 (poor- This fi gure is preliminary and could be revised as est) to 7 (best). According to the guidelines, the experience with the pilot project progresses. Fur- riparian zone is counted as a biological quality el- thermore, aggregation of the smaller watercourse ement. No decision has been made as to how this reaches in particular is being considered. zone should be delineated from wetlands along To facilitate the identifi cation of the water bod- the watercourses, though, and consequently the ies, watercourse physical conditions have been zone has not been included as an element in their charted on the basis of information from the identifi cation. watercourse regulations (only county and municipal watercourses), preliminary investigations in connection with wetland projects and moni- 1.4.4 Physical modifi cation toring results (see also Section 1.4.4). The whole process of assessing the physical conditions is An analysis of large watercourses in Fyn County very time-consuming, among other reasons (county watercourses) reveals that approx. 46% because much of the data is usually only avail- are regulated, and have thereby been shortened able in paper form or has to be collected from by 5–10%. The extent to which the minor wa- the municipal authorities. When information on tercourses have been regulated is unclear, among

Odense PRB Odense Pilot River Basin 23 1.4 Watercourses

Figure 1.4.3 Map indicating the wa- N RWB 3 N ter bodies in the Ryds RWB 4 RWB 2 Stream catchment and RWB 5 RWB 13 RWB 11 the main course of the River Odense. RWB 10 RWB 12

RWB 1 RWB 8 RWB 7 RWB 6 RWB 9 OWB 16

0 1 2 3 km OWB 14 OWB 15

OWB 13 OWB 12 other reasons because many changes of course OWB 11 and length were made long ago (before 1850). A general phenomenon, though, is that the upper sections of the watercourses are more heavily OWB 10 modifi ed than the lower sections. In many places, the upper sections are piped, often including OWB 8 OWB 9 OWB 7 the natural source. An analysis of known piped sections in the whole of Odense River Basin OWB 4 OWB 6 Lake Arreskov OWB 5 shows that approx. 25% of the total watercourse OWB 3 network is piped, although the real fi gure is un- OWB 2 doubtedly signifi cantly higher. Of the remaining OWB 1 open watercourses in Odense River Basin, 60% 0 5 10 km are estimated to be regulated. All in all, the majority of the watercourses are therefore modifi ed by man to some extent. At the same time, of physical modifi cation, an analysis has been though, it is clear that some of these modifi ca- made of the main physical pressures affecting tions are very old such that the environmental a number of sub-reaches (Table 1.4.4). It should quality of at least some reaches have considerably be noted that it has not been possible to identify improved in the course of time through “natural actual artifi cial reaches in the areas tested with re-establishment” (Brookes, 1984), and hence can certainty, even among the smaller, upper ends no longer be considered as heavily modifi ed. of watercourses. Different forms of artifi cially The procedure for the provisional identifi ca- established bypasses through which part of the tion of heavily modifi ed water bodies (Guid- watercourse water passes, for example at dams ance Document 2.2) has in the fi rst instance and weirs in the main course, are thus included been tested on the Ryds Stream catchment and as part of a modifi ed water body. the main course of the River Odense (see also Table 1.4.4 is based on work with two test lo- Section 1.4.3). For use in assessing the degree calities. It is not complete with regard to either

Table 1.4.5 WB No. HM FC Remarks Water bodies (WB) in the main course of the 1 + 4 Regulated, restoration planned, mean slopes 1.2–2.9 m/km River Odense indicat- 2 - 7 Unregulated, sinuous, mean slopes 2.8–6.9 m/km ing the occurrence of 3 + ? Regulated, mean slopes 0.1–4.6 m/km heavy modiﬁ cation 4 - - Lake Arreskov (HM), biological status (FC = fauna 5 + 3–4 Regulated, affected by fluctuation in water level (periodic outflow of water from the lake) class) assessed from the 6 - 5–6 Regulated, but with a rather varied bottom substrate macroinvertebrates 7 - 5–6 (pre) Under restoration (remeandering) (Danish Stream Fauna 8 + ? Regulated, affected by damming, technical structures Index), and remarks 9 - 5–7 Unregulated, sinuous 10 - 5–7 Regulated, but with a rather varied bottom, restoration planned concerning physical 11 - 5–7 Unregulated, sinuous potential. 12 + 3–4 Affected by damming, technical structures 13 - 5–6 Regulated, but with a rather varied bottom substrate 14 + 3–5 Regulated, several technical structures 15 - 4–6 Slightly regulated 16 + 4 Affected by salt water, cooling water and impoundment

Odense 24 PRB Odense Pilot River Basin 1.4 Watercourses

watercourses in Fyn County or in Denmark as dams, weirs, etc. that cause signifi cant impound- a whole. For example, it lacks groundwater ab- ment of the water or the formation of upstream straction, which in some cases can considerably lakes are also a major physical modifi cation. In affect watercourse fl ow and hence also hydro- the main course of the River Odense, 7 out of 15 morphology (especially in the watercourses of reaches corresponding to 33% of the main course eastern Denmark, to which those of Fyn County (20 km) have thus been provisionally designated belong, and which are affected by water abstrac- as heavily modifi ed (Table 1.4.5). The main tion to supply the major towns). In general, course also contains other reaches that have been though, the analysis is a good tool for use in the severely regulated over the years, but where most subsequent design of restoration scenarios and of them have a fauna class >5 (i.e. good status or for prioritization of efforts to improve the status better) and a relatively varied bottom substrate. It of the water bodies. During the work process it has been decided not to designate these reaches as transpired that information about drainage of the heavily modifi ed water bodies. fl oodplain is diffi cult to obtain, and only sporadic While subdivision of the main course is in prin- information is available about sand transport and ciple “quite simple”, this is not the case with the bank stabilization. In addition, it is characteristic minor watercourses due to the fact that they are that the data material is best for the major wa- often very fragmented by piped sections/dams tercourses, whereas knowledge about the small or other forms of regulation (straightening/ watercourses is more limited. With two of the deepening). Subdivision into sub-reaches accord- tributaries of Ryds Stream catchment it has been ing to physical status might therefore result in necessary to make a fi eld inspection. In these an enormous number of water bodies, which is cases it will be necessary to clarify the degree of naturally undesirable. Conversely, merging of detail to be recorded, and to draw up unambigu- too many sub-reaches into a single water body ous operational methods for assessing the extent entails that it will be possible to designate them of physical modifi cation at the national level. all as heavily modifi ed irrespective of the fact The most extensive physical modifi cations are that some sub-reaches could have a very good piping and straightening/deepening, where the status as assessed from fauna class and physical form of the watercourse and drainage status of conditions. There is therefore a general need to the surroundings are radically altered, and where draw up more exact criteria for what is meant by efforts are subsequently made to maintain this “heavily modifi ed”. status through intensive watercourse mainte- The water bodies in the Ryds Stream catch- nance (frequent weed cutting and dredging of ment have provisionally been designated as heav- bottom substrate). The establishment of major ily modifi ed according to the criteria in Table

Table 1.4.6 Physical pressure Extent (percentage of reach length) Criteria for prelimi- Piped reaches >50 nary identiﬁ cation of Regulation (grade 2–3) >75 heavily modiﬁ ed water Piped reaches and regulation (grade 2–3) >75 bodies. Maintenance (grade 3) >50 Hydraulic stress (considerable pressure) >50

Table 1.4.7 WB No. HM FC Remarks Water bodies (WB) 1 + 4 Upper parts piped, mean slopes 4.4–5.6 m/km in the Ryds Stream 2 + 5 Partially piped, with spring, mean slopes 6.4–20 m/km catchment indicat- 3 + ? Highly regulated, lower part piped, mean slope 0.9 m/km ing the occurrence of 4 + - Completely piped heavy modiﬁ cation 5 + - Completely piped (HM), biological status 6 + 4–5 Several piped/regulated sections, lower part sinuous, mean slopes 5.4–11 m/km (FC) assessed from the 7 - 4–6 Upper parts piped, remainder sinuous, many small obstructions, mean slope 10 m/km macroinvertebrates 8 + 4–5 Upper parts piped, with springs, mean slope 9.2–13 m/km (Danish Stream Fauna 9 + ? Upper parts piped, mean slopes 1.5–3.2 m/km Index) in the open part 10 - 5 Unregulated, sinuous, mean slope 9.0 m/km 11 - 5–7 Unregulated, slightly sinuous, mean slope 2.4 m/km of the watercourse, and 12 - 6–7 Regulated, but sinuous, mean slopes 1.2–1.8 m/km remarks concerning 13 + 4 Unregulated, sinuous, very affected by hydraulic stress, mean slope 3.8 m/km “physical potential”.

Odense PRB Odense Pilot River Basin 25 1.4 Watercourses

1.4.6. As a consequence, 9 out of 13 water bodies ing to the expected conditions if undisturbed by can be considered to be heavily modifi ed (Table human activity. For each watercourse reach of 1.4.7), corresponding to 65% of the total water- known type, an assessment has to be made of course network in the catchment. It is empha- the extent to which the current status deviates sized that the criteria established are provisional, from the “natural background conditions”. There and that changes can occur as the pilot project are several possibilities for describing the refer- progresses. ence conditions, namely using current data from No attempt has been made to identify heavily virtually undisturbed Danish watercourses or physically modifi ed watercourse reaches in the comparable foreign watercourses, historical and Lunde Stream system. palaeolimnological data, or, if necessary, data Based on experience to date, and applying the from expert judgements. criteria in Table 1.4.6, it must be presumed that Assessment of the extent to which suitable around 60% of the total watercourse network in reference stations exist is primarily based on Odense River Basin will be provisionally desig- the macroinvertebrate fauna, for which the data nated as heavily modifi ed. material is generally greatest. For natural reasons (cf. the island biogeographic principles on spatial 1.4.5 Reference conditions extinction and immigration), there are clear differences between the macroinvertebrate fauna Under the provisions of the WFD, reference con- in Jutland, Fyn + Zealand, and Bornholm. In ditions must be established for each type of wa- addition, man has enhanced this difference in tercourse (Guidance Document 2.3) correspond- various ways.

Table 1.4.8 Data series Watercourse Locality Type No. of data sets (yr) Stations assessed as being suitable as Historical Lindved Main road A1 2 4 (1947–1950) reference stations for biological conditions Stavis Upstream of Lærkehus 2 4 (1951–1952) in watercourses in Fyn Stavis Lake Langesø forest 2 1 (1952) County. The data in Stavis Morud forest 2 1 (1947) (3 outliers excluded) question primarily Stokkebæk Mullerup forest 2 1 (1945) concern macroinverte- Stokkebæk Skrams Vænge 2 1 (1945) brates and are included Odense Lykkensprøve 3 5 (1942–1955) in the analysis shown in Figure 1.4.4. Rislebæk Sollerup 1 Several (2001 selected) Present-day Stamperenden Idyllendal 1 Several (2001 selected) Ørredbæk Forest road 1 Several (2001 selected) Brende Tanderup 2 Several (2001 selected) Hattebæk Upstream of fish farm 2 3 (1996–1998) Hattebæk Faldsled–Jordløse road 2 Several (2001 selected) Hågerup Downstream of Hågerup 2 Several (2001 selected) Kongshøj Ågård 2 Several (2001 selected) Lindved Main road A1 2 Several (2001 selected) Stokkebæk Dyregård 2 Several (2001 selected) Tange Downstream of Skovmøllen 2 Several (2001 selected) Traunskov-afløb Vissenbjerg–Morud road 2 Several (2001 selected) Vindinge Rønninge 2 Several (2001 selected)

Odense Vibæk 3 Several (2001 selected)

Table 1.4.9 Objective Classification (Danish EPA, 1983a) Required FC Existing system of objectives used in Fyn Reference area for scientific studies A (Areas of special scientific interest) >5 County and the requirements stipulated Salmonid spawning and nursery waters B1 and B2 (Salmonid waters) >5 for fulﬁ lment of the Fish waters for angling/fishery B (Cyprinid waters) >5 objective. 3 Aesthetically satisfactory, etc. C–F (Eased objectives) >4

Odense 26 PRB Odense Pilot River Basin 1.4 Watercourses

The fi rst step taken was to examine which MDS for stream macroinvertebrates Type 3_ref_old regional data could be used to describe the refer- Type 3_ref_new Figure 1.4.4 Type 3_other_new Multidimensional ence conditions for watercourses in Fyn County Type 2_ref_old Type 2_ref_new scaling (MDS) for ma- (Table 1.4.8). As is apparent, the data material Type 2_other_new is generally rather sparse (7 old and 14 new sta- Type 1_ref_new croinvertebrates in the Type 1_other_new watercourses of Fyn. tions). Comparison of the macroinvertebrate The data derive partly fauna by multidimensional scaling (MDS) on from new (new) and the basis of the Bray Curtis similarity shows that historical (old) refer- there is a clear difference between old (histori- ence stations (ref, see cal) and new data sets (Figure 1.4.4). In making Table 1.4.8), and partly the analysis, attempts were made to compensate from remaining (other) for the fact that the methods employed earlier new watercourse sta- in time differ from those currently employed tions where the com- (quantitative data have been normalized, a few position and density of macroinvertebrates species have been aggregated; in addition, species have been comprehen- composition was tested for alone, with the same and 3.000 years ago, respectively (Klink, 1989; sively investigated. result.) Overall, the results indicate that it is not Wilkinson, 1987). Such studies are also feasible The distance between possible – even using the best stations currently in Denmark, for example in the fl oodplain of the points refl ects the available – to fi nd a macroinvertebrate fauna the River Odense. The method has already been difference in the fauna quite resembling that which previously existed tested on 10 300 year-old watercourse sediments between the individual (certain species have become extinct, and with from the Great Belt (Wiberg-Larsen et al., 2001). stations. The stress of others immigration can be expected to take a A good initial suggestion for the reference the ordination is 0.14 very long time). The same undoubtedly applies conditions for watercourses in Fyn County is (i.e. a useful 2-dimen- sional picture). to plants, for which the data material is even reaches with a natural course, good hydraulic smaller (e.g. Riis et al., 1999). contact with the surroundings, no or very ex- As impoverishment of the watercourse fl ora tensive utilization of the surrounding land, gen- and fauna due to human activities has been con- erally varied physical conditions in and around siderable, especially over the past 100 years (e.g. the watercourses, clean water with a low nutrient Riis & Sand-Jensen, 2001; Jensen & Jensen, 1980), content and a naturally varied fl ora and fauna the lack of suitable reference data is expected to both in and around the watercourses. These apply to large parts of the country. There is thus conditions will be close to that described as high a need to establish a suitable network of reference ecological status in Section 1.4.6. stations in Danish watercourses, and it will probably also be necessary to utilize data from reference stations abroad (southern Sweden, northern 1.4.6 Provisional establishment of Germany, southern Baltic countries, northern objectives Poland). There is a particular lack of data for all relevant biological quality elements (plants, Fyn County’s current Regional Plan stipulates a macroinvertebrates and fi sh), and knowledge of quality objective stating the use for which each the hydromorphological quality elements under watercourse should be suitable (Table 1.4.9). reference conditions also needs to be improved. Whether a watercourse fulfi ls its quality objec- For comparative purposes, consideration should tive is determined biologically from the fauna be given to the use of independent expert judge- class (Danish Stream Fauna Index). Since the ment for selected quality elements, for example fauna class at a given location can vary from year for part of the macroinvertebrate fauna (regional species lists) and fl ora (regional species lists). Finally, it is possible to gain an impression of Ecological Required FC Requried FC Table 1.4.10 the composition of the macroinvertebrate fauna status (slope >0.1 m/km) (slope 0–0.1 m/km) Summary of how the fauna class (cf. DSFI) and fl ora in watercourses of the past – before the High 6–7 5 (if possible) impact of human activities became extensive – by can initially be used to employing palaeolimnological methods. Investi- Good 5 4 assess watercourse ecological status. It should gations of subfossil insect remains in “old” water- Moderate 4 3 be noted that a greater course sediments have thus been successfully uti- number of quality ele- lized to describe the macroinvertebrate fauna in Poor 3 2 ments will be included the lower part of the Rhine (Germany/Holland) Bad 1–2 1 in the assessment in and the River Avon (England) for 100–250 future.

Odense PRB Odense Pilot River Basin 27 1.4 Watercourses

Table 1.4.11 Water Physical Surroundings Size Biology Variables included in chemistry conditions the data analysis for Percentage of farmland in Catchment NH -N, mean Degree of shading No. of species/taxa: 103 monitoring sta- x catchment area NO -N, mean Percentage Macrophytes tions in watercourses Woodland, nearest 25 m Winter width x stone+gravel Macroinvertebrates PO -P, mean in Fyn County. The Trees/bushes, nearest 2 m Max depth 4 Degree of sinuousity Fish physical index is cur- High (>0.5 m) herbaceous BOD5, mean Pools + riffles Coverage: rently calculated as a plants, nearest 2 m Roots All macrophytes modiﬁ ed Aarhus Index Herbaceous plant diversity, Degree of regulation Submerged (after Kaarup, 1999; nearest 2 m Physical index macrophytes but with plants, cf. the Fauna class (DSFI) unpublished proposal Fish index drawn up by the Physi- cal Index Working Group), and the ﬁ sh index is calculated according to WaterFrame (2003).

Table 1.4.12 BOD NH -N NO -N PO -P Status Physical index DSFI Fish index 5 x x 4 Summary of how dif- mg/l (mg/l) (mg/l) (mg/l) ferent variables can High >33 6–7 55–60 <0.5 <0.05 <0.8 <0.020 be used to assess the Good 25–33 5 47–54 0.5–2.0 0.05–1.0 0.8–2.0 0.020–0.040 ecological status of Moderate 17–24 4 38–46 2.1–3.5 1.1–2.5 2.1–5.0 0.041–0.090 watercourses. It should Poor 8–16 3 24–37 3.6–5.0 2.6–5.0 5.1–7.5 0.091–0.170 be noted that these Bad <8 1–2 12–23 >5.0 >5.0 >7.5 >0.170 considerations are only preliminary. The basis for the classifi cation is as follows: Physical to year, among other reasons due to differences over, the WFD prescribes the use of considerably index: Kaarup (1999); in precipitation and hence runoff, the objective more quality elements, including both physico- Fish Index: Water- is only considered to be fulfi lled in regional envi- chemical elements and other biological elements Frame (2003); Water chemistry: Prelimi- ronmental administration if the minimum fauna (Guidance Document 2.3). nary data and for high class requirement has been met for at least fi ve In order to determine what consequences status among others consecutive years. this can have, a more detailed analysis has been Kristensen & Hansen Under the provisions of the WFD, watercours- made of data from 103 monitoring stations in (1994). es have henceforth to be subdivided into fi ve Fyn County for which more detailed informa- classes of ecological status. The objective here is tion is available, for example on land use and that all water bodies should achieve good ecologi- physico-chemical and biological conditions. As cal status by 2015 at the latest, while at the same regards size and fauna class, the stations turned time preventing deterioration in the existing sta- out to be reasonably representative for the whole tus. A simple and manageable means of convert- regional station network, as well as for the sta- ing to the new system is to initially continue to tion network in Odense River Basin, although, employ the macroinvertebrate fauna to assess ful- as mentioned earlier, the smallest watercourses fi lment of the objective (Table 1.4.10). This is not are under-represented in the monitoring (Section without problems, though, as the DSFI employs 1.4.1). The analysis encompassed an initial selec- a 7-step scale and the WFD a 5-step scale. More- tion of the strongest descriptive variables in the Table 1.4.13 data set (Table 1.4.11) and subsequent statistical Watercourse total N analysis of these variables (see Annex 1.4). From and total P concen- Ecological Total N Total P this it is apparent that there is a positive correla- trations for different status (mg/l) (mg/l) tion between the selected biological variables and ecological status classes. High <1.0 <0.030 watercourse size, the presence of natural, varied This also gives the physical conditions in the watercourse and the requirements that wa- Good 1-0–2.5 0.030–0.060 presence of woodland around the watercourse, tercourses will have to respectively. Correspondingly, there is a nega- Moderate 2.6–6.0 0.061–0.125 meet in order to attain tive correlation to the degree of anthropogenic high and good ecologi- Poor 6.1–9.0 0.126–0.250 cal status in lakes and pressure described by land use for agriculture, watercourse regulation and input of easily me- coastal waters, respec- Bad >9.0 >0.250 tively. tabolizable organic matter (BOD5) and nutrients

Odense 28 PRB Odense Pilot River Basin 1.4 Watercourses

Ecological status from the surroundings, respectively. It should 100 Fauna class (DSFI) Figure 1.4.5 be noted that even though BOD5 is of relatively minor signifi cance in this data set (Annex 1.4), an 80 Distribution of 103 analysis of small watercourses alone would show watercourse stations 60 according to ecological that its signifi cance for biological conditions is 40 status classes for the considerably greater (Fyn County, 2001c). physical, chemical and When the indices employed for the 103 moni- 20 biological variables toring stations are distributed according to eco- listed in Table 1.4.12. 100 logical status classes it is seen that the status is Fish index These variables were generally assessed as being poorer for fi sh and 80 selected on the basis of physical conditions than for macroinvertebrates the statistical analysis 60 (Figure 1.4.5; Table 1.4.12). That the fi sh index described in Annex 1.4. indicates a relatively greater share of watercours- 40 es with bad status than the physical index is in 20 part attributable to the fact that certain forms of physical disturbances – for example dams, weirs 100 Physical index and drying-out – are not included in the physical 80 index (just over 40% of all the stations rated by the fi sh index as having bad status are affected 60 by one or more of these). At the same time it is 40 emphasized that speedy completion of the fi nal physical index is vital, and that the development 20 of a suitable index for plants is highly desirable. 100 Attempts have also been made on a trial basis BOD5 to subdivide into status classes for the water 80 chemistry variables that were included in the 60 statistical analysis (Figure 1.4.5; Table 1.4.12; Annex 1.4). It should be noted that even though 40 nitrogen and phosphorus are usually input to 20 the fl owing water in such large amounts that the 100 watercourse plants are not limited by them, and NH x-N the macroinvertebrates and fi sh are hardly likely 80 to be affected either (Section 4.2), the input of (%) stations watercourse of Proportion 60 these substances is of great signifi cance for the status of other categories of water body (lakes, 40 coastal waters, etc.) fed by the watercourses. In 20 many cases it will therefore be necessary – out of consideration for these water bodies – to include 100 NO x-N nitrogen and phosphorus as an important part of 80 the assessment of watercourse status. A suggestion for the distribution of the total content of 60 these substances across the status classes is shown 40 in Table 1.4.13. Apart from the above-mentioned methods, the status of Danish watercourses can 20 be assessed from other variables such as maxi- 100 mum temperature and pH, as well as the content PO 4-P of oxygen, ochre (especially in Jutland) and 80 various hazardous substances, and the necessary 60 amount of water (see also Figure 4.2.1). In any 40 event, it will be necessary to decide how these variables should be weighted against each other 20 in the fi nal system for determining the ecological status of the watercourses. High Good Moderate Poor Bad Part of the River Odense system, including large parts of the main course and the tributaries Hågerup, Sallinge and Lindved Streams, are

Odense PRB Odense Pilot River Basin 29 1.4 Watercourses

Ryds Stream – a reach surrounded by woodland.

Photo: Bjarne Andresen, Fyn County

encompassed by the Habitats Directive. It is would otherwise have to fulﬁ l in order to achieve therefore important that there is a good corre- good watercourse ecological status. lation between the two ways of assessing status Concrete suggestions for the provisional estab- that accompany the two Directives. Thus high lishment of objectives for the watercourse reach- and good status pursuant to the WFD should es in the two test areas – Ryds Stream catchment correspond to the designation “favourable con- and the main course of the River Odense – are servation status” pursuant to the Habitats Direc- given in Section 4.2.3, together with an assess- tive. However, it is presently uncertain whether ment of the likelihood that the environmental consideration for special habitat species can lead quality objectives will not be achieved by 2015. to requirements different from those that they

Odense 30 PRB Odense Pilot River Basin 1.5 Lakes

1.5 Lakes

1.5.1 Location of the lakes 1.5.2 Physical modiﬁ cation

There are 2 620 lakes larger than 100 m2 in Many lakes have been physically modifi ed over Odense River Basin. Their combined area is the years as a result of lowering of the water level, 1.106 ha, corresponding to 1% of the whole ba- fi lling in, damming, etc. A very large proportion sin. Their location is apparent from Figure 1.5.1, of the typically small, shallow lakes have com- and their size distribution from Table 1.5.1. pletely disappeared over the past 100 years or so. The WFD does not specify any lower size For example, the number of lakes in the Lake limit for the lakes it encompasses. The typology Arreskov catchment area has decreased by 76% “System A” in Annex II of the WFD suggests a from 276 around the year 1890 to 65 in 1992. lower size limit of 0.5 km2, but the horizontal Many of the existing lakes in Odense River Guidance Document “Indentifi cation of Water Basin have arisen as a result of human activity, Bodies” stresses that the WDF applies to all sur- e.g. peat mining, clay, marl or gravel quarrying, face waters. Member States can decide that also or as a result of dams etc. especially for operat- smaller water bodies are so important that they ing mills. Of 66 investigated lakes in the basin, should be individually identifi ed.” In Denmark, nearly half (45%) arose as a result of peat mining, all lakes larger than 100 m2 are protected by the and only 18 (27%) are natural (Table 1.5.2). Of Protection of Nature Act and are individually the larger lakes (>5 ha), however, nearly all are identifi ed. This protection is partly due to the natural. fact that many small lakes were disappearing, and partly because the many small lakes are Heavily modifi ed water bodies important natural elements in the very cultur- Millponds derived from damming watercourses ally infl uenced Danish landscape. For example, are classifi ed as heavily modifi ed water bodies. the small lakes and ponds together contain more Thereafter a decision has to be made whether species of macroinvertebrates such as worms, to preserve the dam, for example for cultural snails, mussels, crustaceans and insects than both historical reasons. If it is decided to eventually the large lakes and watercourses (Fog & Wiberg- remove the dam and re-create the watercourse, Larsen, 2002). the reach is thereafter classifi ed in the river basin As a consequence, all lakes larger than 100 m2 management plan as a watercourse and not as a have been identifi ed as discrete water bodies. heavily modifi ed water body. This emphasizes that the protection offered by the WFD applies to all surface waters. If the small lakes are not discretely identifi ed, there is the risk that this important habitat type could only be improved to the extent necessary to achieve good status in the water bodies with which they are in direct or indirect contact (cf. Table 1.5.1 Total Investigated Section 3.5 of the horizontal guidance “Identifi - Size Number area Number and total area cation of water bodies”). (ha) Number % of lakes of various size It is not practicable to investigate each of the categories in Odense more than 2 600 lakes, however. According to the >5 ha 14 606 11 79 River Basin. horizontal guidance “Identifi cation of water bod- >3 ha 21 639 11 52 ies”, it is permissible to aggregate lakes in relation to monitoring, reporting and management. The >1 ha 97 767 20 21 lakes are grouped on the basis of lake type and >0.5 ha 228 858 27 12 catchment (assessed from GIS), and monitoring is carried out for a randomly selected subgroup of >0.1 ha 1 058 1 032 50 5 the lakes. This aggregation and selection has not >100 m² 2 620 1 106 63 2 yet been carried out.

Odense PRB Odense Pilot River Basin 31 1.5 Lakes

Figure 1.5.1 tion. Moreover, good chemical status also has Location of the 2 620 to be achieved. Thus relative to good ecological lakes >100 m2 in N status, good ecological potential is only less Odense River Basin. stringent regarding the special physical conditions pertaining in connection with dams and the special physical nature of lakes originating from gravel quarries and peat mines. The status of the water body thus has to be measured against the maximum ecological potential. This describes the closest approximation to the reference conditions in the natural ecosystem that can be achieved with the hydromorphological conditions pertaining. With the majority of peat mine and gravel quarry lakes, which only differ slightly from natural lakes (e.g. very steep banks), maximum ecological potential will correspond to the reference conditions. Classiﬁ cation as artiﬁ cial water bodies thus does not entail the imposition of less stringent requirements concerning, for example, pollution of the water body than if it had been natural in origin.

0 5 10 km A special group consists of the lakes that have arisen in connection with Action Plan on the Aquatic Environment II. These have been Artifi cial water bodies formed with the main aim of enhancing retention Lakes created through human activity in a loca- and denitrifi cation of the nitrogen leaching from tion where there has not previously been a water agricultural land. Pursuant to the Danish EPA’s body are classifi ed as artifi cial water bodies. This current guidelines, quality objectives are not to particularly applies to peat mine, gravel quarry be set for these lakes, and consideration for their and marl/clay quarry lakes, duck ponds, and, to ecological status cannot in itself justify measures a certain extent, also to village ponds and fi re in the catchment to reduce the pollution. reservoirs. Unlike natural lakes, artifi cial and heavily Modifi cation of natural lakes modifi ed lakes do not have to achieve good eco- The natural lakes are often also modifi ed as a logical status, but rather “good ecological poten- result of changes in water level. The changes are tial”. Correspondingly, reference conditions do usually so slight, though, that the water body not have to be established for these lakes, but does not change type. This applies, for example, rather a “maximum ecological potential”. to Lake Arreskov, where the water level has been Through the use of good ecological potential lowered in several steps such that large parts of it is possible to impose just as stringent require- the former lake surface are now freshwater mead- ments regarding pressures, for example pollu- ows and mires. Despite the change, the lake can fulfi l the criteria for good ecological status and hence does not need to be classifi ed as a heavily Table 1.5.2 Origin No. of lakes % Origin of 66 investi- modifi ed water body. In Lake Langesø, among gated lakes in Odense Natural 18 27 others, the water level has been raised relative River Basin. to the natural level by a dam in the outlet. In Peat mine 30 45 this case too, the assessment is that this change Marl/clay quarry 7 11 in hydromorphology is so slight that it does not hinder achievement of good ecological status. Gravel quarry 4 6 Characterization of the lakes Village pond, etc. 3 5 The lakes have to be characterized as either natu- Dam/millpond 4 6 ral lakes, artifi cial water bodies or heavily modifi ed water bodies. Based on the above remarks, Total 66 100 Figure 1.5.2 shows the location and category of

Odense 32 PRB Odense Pilot River Basin 1.5 Lakes

the 66 investigated lakes. The lakes are shown Figure 1.5.2 subdivided by origin in Table 1.5.2. Of these, 18 Category and location are characterized as natural, 44 as artifi cial wa- N of 66 investigated lakes ter bodies (peat mine and marl, clay and gravel in Odense River Basin. quarry lakes, village ponds, fi re reservoirs, etc.) Natural water bodies and 4 as heavily modifi ed water bodies (dams and Heavily modified water bodies millponds). Artificial water bodies

1.5.3 Typology

According to the WFD, Odense River Basin lies in ecoregion 14: Central plains. Differentiation of lake types can be done using the typology “System A” (altitude, mean depth, surface area and geology) or “System B” (obligatory and optional physico-chemical factors), cf. Annex II of the WFD. For typology of lakes in Denmark it has been decided to use “System B”, with lake type being determined from alkalinity, colour, salinity and mean depth (Table 1.5.3), cf. National Environ- mental Research Institute (in prep.). 0 5 10 km This potentially gives 16 different lake types. In reality, though, only 11 of these exist in Den- mark. The lake types refer mainly to lakes larger than 5 ha. It appears to be inappropriate, though, example, mussels become more widespread, and that lake size cannot be taken more into account. the zooplankton shifts at a salinity exceeding In very small lakes (devoid of ﬁ sh), the biological approx. 10–12‰. Unfortunately, little is pres- conditions are very different from those in large ently known about the distribution of this type lakes. In addition, it should also be possible to of lake. differentiate out lakes with high salinity. This potentially gives 72 different types of This system should therefore be extended so as lake. In practice, many of the possible combina- to also take into account lake size, and a special tions will not exist, and as is apparent from Table class should be introduced for lakes with high 1.5.3c, no more than 17 types exist in Odense salinity (Table 1.5.4). It should thus be possible River Basin. to differentiate between very small lakes (<0.1 Due to the lack of knowledge it will be dif- ha), small lakes (0.1–1.0 ha) and lakes larger ﬁ cult to establish reference conditions for many than 1 ha. In addition, consideration should be of the lake types. For this reason, the National given to introducing a class for lakes with salin- Environmental Research Institute and the Dan- ity exceeding 12‰ since biological conditions ish Forest and Nature Agency are not presently change dramatically with increasing salinity. For prepared to propose more than the original 16

Parameter Low High Table 1.5.3 Typology of lakes >1 Alkalinity <0.2 meq/l 0.2 meq/l. ha. After National En- Colour <60 mg Pt/l 60 mg Pt/l vironmental Research Salinity <0.5‰ 0.5‰ Institute (in prep.). Mean depth 3.0 m >3.0 m

Parameter Low Medium High Table 1.5.4 Alternative typology of Alkalinity <0.2 meq/l 0.2 meq/l lakes in Odense River Colour <60 mg Pt/l 60 mg Pt/l Basin. Salinity <0.5‰ 0.5‰ <12‰G 12‰G Mean depth 3.0 mG >3.0 m Area <0.1 ha (0.001 ha)G 0.1 ha <1.0 haG >1 ha

Odense PRB Odense Pilot River Basin 33 1.5 Lakes

Type No. of Table 1.5.5 Area Alkalinity Colour Salinity Depth Examples Lake types in Odense No. lakes River Basin according 1 Small Low High Low Low Few/none Small brownwater lakes/ponds to the typology given Most of the small freshwater lakes/ in Table 1.5.4. 2 Small High Low Low Low Ca. 1 500 ponds Total of Freshwater-influenced coastal 1 577 lakes 3 Small High Low Medium Low Few/none lagoons

4 Small High Low High Low Many Coastal lagoons

5 Medium Low High Low Low Few/none Small brownwater lakes

6 Medium High Low Low Low Ca. 900 Most of the small freshwater lakes

7 Medium High Low Low Deep Some Gravel quarrry lakes Total of 954 Freshwater-influenced coastal 8 Medium High Low Medium Low Few/none lakes lagoons Quarry lakes influenced by salt 9 Medium High Low Medium Deep Few/none water

10 Medium High Low High Low Some Coastal lagoons

11 High Low High Low Low 1 Lake Sortesø

12 High High Low Low Low Ca. 90 Most of the freshwater lakes

13 High High Low Low Deep >1 Lake Søbo, gravel quarry lakes

Freshwater-influenced coastal 14 High High Low Medium Low Few/none Total of 98 lagoons lakes Quarry lakes influenced by salt 15 High High Low Medium Deep Few/none water 16 High High Low High Low Some Coastal lagoons

Quarry lakes influenced by salt 17 High High Low High Deep Few/none water

Table 1.5.6 Lake type Low alkalinity, shallow Alkaline, shallow Alkaline, deep Proposed reference values for three lake types <3 m, <0.2 meq/l <3 m, >0.2 meq/l >3 m, >0.2 meq/l (>1 ha). Summer mean values. Depth refers to Total P (mg/l) 0.010 0.015 0.08 mean depth. After Na- Total N (mg/l) 0.37 0.4 0.38 Chlorophyll a (µg/l) 2.5 3.7 3.9 tional Environmental Secchi depth (m) 4.1 3.8 5.4 Research Institute (in prep.).

types. The future monitoring work should aim data 3) Modelling including palaeolimnological to help improve knowledge of a number of the studies or 4) expert judgement. less well-known lake types, especially the small None of the lakes in Fyn County can be said to lakes, including small coastal lagoons. hold reference conditions as they are all more or less affected by enhanced nutrient loading. Palaeolimnological studies in Danish lakes 1.5.4 Reference conditions show that the best proposal for reference conditions from the temporal point of view is the pe- According to the WFD, the undisturbed ecologi- riod around 1850–1900 (National Environmental cal status – termed the reference conditions – is Research Institute, in prep.). to serve as the basis for the ecological classiﬁ ca- Palaeolimnological studies have been under- tion. Establishment of the reference conditions taken in a few lakes in Odense River Basin. The is therefore important, and there are several results are summarized below. The phosphorus methods for doing this: 1) Current status in un- content back in time has been calculated from disturbed Danish or foreign lakes, 2) Historical the composition of the diatoms in the sediment.

Odense 34 PRB Odense Pilot River Basin 1.5 Lakes

The concentrations are therefore subject to some layers. In 2002, the phosphorus concentration degree of uncertainty, although the method is was 0.172 mg/l. generally considered to be reliable. Lake Nørresø Lake Dallund Investigations of plant remains have shown that The lake is described on the basis of an approx. a well-developed submerged macrophyte veg- 11 m drill core covering approx. 7 000 years. etation formed at the end of the 18th Century, The results of the biological and physico-chemi- probably after lowering of the water level. The cal measurements indicate that the lake was in phosphorus concentration was approx. 0.030 a stable state up to the end of the Bronze Age mg/l around 1850 and increased to around 0.060 (around 750 BC). The phosphorus concentration mg/l up to 1900. Around 1930, and after 1950, was around 0.020 mg/l. At this point changes the concentration increased markedly, peaking at took place due to increased agriculture. Greater 0.120 mg/l around 1970. In 2002, the phosphorus changes took place in the Middle Ages after the concentration was 0.073 mg/l. introduction of the wheel plough and forest clearance, with a resultant marked increase in Lake Søby nutrient input. In that period, the phosphorus An investigation of algal pigment residues in the concentration reached as high as 0.175 mg/l. lake sediment indicates that the composition of After a minor fall in phosphorus concentration the algal community was relatively stable from in the 18th Century, the impact of human activ- the end of the 19th Century and up to around ity increased markedly again up to the beginning 1940, where after a marked increase in eutrophi- of the 1990s. After the cessation of wastewater cation took place over the period 1950–1977. The discharge into the lake and restoration in the phosphorus concentration has not been recon- form of biomanipulation, the phosphorus con- structed. centration has again fallen to 0.059 mg/l in 2000 (average for the summer period). Establishment of reference conditions The data material for establishing the reference Lake Langesø conditions for the various lake types is rather The phosphorus concentration in Lake Langesø weak. In addition, each lake has its own history, was already high in 1850 (0.135 mg/l) and as related above. Based on data for the cleanest increased considerably until around 1900 (to Danish lakes the National Environmental Re- approx. 0.180 mg/l) and again after 1950 (to ap- search Institute has proposed reference values for prox. 0.220 mg/l). The eutrophication that took water chemistry parameters for the three most place at the end of the 19th Century is probably common types of lake (Table 1.5.4). As is appar- attributable to a manor located near the lake. ent from Table 1.5.5, by far the majority of the New investigations of a deeper drill core show lakes in Odense River Basin belong to the alka- that the phosphorus content was also high before line, shallow type, with only a few lakes belong- 1850. The lake is apparently naturally eutrophic, ing to the other two types. None of the lakes are probably as a result of the inﬂ ow of phosphorus- expected to have reference conditions according rich water from naturally phosphorus-rich soil to these criteria.

No. of No. of lakes Preliminary future Table 1.5.7 Objective Lake name lakes meeting objective objective Current objectives and Arreskov, Brændegård, Sortesø, High ecological preliminary future A1 6 1 Store Øresø, Nørresø, Fjordmarken status objectives for lakes in Søbo, Langesø, Nr. Søby, Dallund, Good ecological Odense River Basin. B 6 0 Fjellerup, Brahetrolleborg Slotssø status No specific objective Good/high eco- set, but generally status 2 608 ? Table 1.5.8 logical status should meet objective B Proposed classiﬁ cation in ecological status classes based on Phosphorus content (mg/l) High Good Moderate Poor Bad phosphorus content (summer mean). After Shallow, alkaline 0–0.025 0.025–0.050 0.050–0.100 0.100–0.200 >0.200 National Environmen- Deep, alkaline 0–0.0125 0.0125–0.0250 0.025–0.050 0.050–0.100 >0.100 tal Research Institute (in prep.).

Odense PRB Odense Pilot River Basin 35 1.5 Lakes

In order to be able to establish reference condi- WFD objectives tions for the other lake types, further data and The WFD operates with fi ve classes of ecologi- possibly also expert judgements are needed. cal status: High, good, moderate, poor and bad. With high ecological status, no or little deviation is permitted from undisturbed status (reference 1.5.5 Provisional establishment of objec- conditions). With good ecological status, slight tives deviation is permitted. With moderate ecological status, moderate deviation from reference condi- Current objectives tions is permitted. The aim is that all surface Fyn County’s Regional Plan operates with two water bodies should achieve good status no quality objectives for the lakes in Odense River later than 2015. The National Environmental Basin: A1 – reference area for scientifi c studies, Research Institute (in prep.) has proposed criteria and B – fi sh waters for angling and fi shery. Both for the individual ecological classes for the most objectives entail that the lakes have to have a common lake types found in Denmark. The natural and divers fl ora and fauna that may only proposal encompasses both water chemistry and be slightly affected by pollution. The system of biological parameters. One of the fundamental quality objectives also encompasses an eased parameters is phosphorus content. Even though objective – C, but this is not employed in Fyn this does not in itself comprise a classifi cation County. In lakes with quality objective A1, the element under the WFD, phosphorus availability fl ora and fauna must be accorded special protec- determines the status of the majority of lakes. tion. Lakes for which a specifi c quality objective The proposed classes are shown in Table 1.5.8. has not been set are encompassed by the general The current basic objective, B, requires a natu- requirement to meet quality objective B. Six lakes ral and diverse fl ora and fauna. This is fundamen- in Odense River Basin have the specifi c quality tally similar to the WFD’s good ecological status, objective A1, and six have quality objective B which has to be achieved by all lakes by 2015. (Table 1.5.7) All the other lakes larger than 100 Pursuant to the WFD, the status of the lakes m2 are encompassed by the general requirement may not deteriorate, i.e. lakes that currently have to meet quality objective B. high ecological status must also preserve this No specifi c requirements have been established high ecological status in future. In a few gravel for the lakes for which a quality objective has quarry lakes the phosphorus concentration is less been set, but they are subject to a number of than 0.025 mg/l, and there is thus the potential general requirements concerning the input of to achieve high ecological status. In contrast, by polluting discharges. For example, wastewater far the majority of natural lakes are assessed as discharges should as far as possible be avoided. having moderate to poor status. At present, only 1 of the lakes for which a spe- When establishing the future objective it is cifi c quality objective has been set, Lake Øresø, reasonable that lakes currently designated as meets its objective. “reference areas for scientifi c studies”, quality objective A, should be assigned the environmental objective “high ecological status” pursuant to the WFD. However, this requires closer assessment of every individual lake, which will be carried out in connection with preparation of the fi rst river basin management plan. A number of the lakes are also encompassed by international protection, EC Habitat sites and EC Bird protection sites (see Section 2). Among others, this applies to Lake Arreskov, Lake Brændegaard, Lake Nørresø, Lake Sortesø, Lake Øresø and Lake Fjordmarken. In these areas, “favourable conservation status” has to be achieved for the species and habitat types for which the areas have been designated. In the case of the lakes, this typically entails certain species of aquatic plants and birds. The environmental objectives established pursuant to the WFD shall also help ensure that favourable conservation status can be achieved for these species.

Lake Arreskov. Photo: Kjeld Sandby Hansen, Fyn County Odense 36 PRB Odense Pilot River Basin 1.7 Groundwater

1.7 Groundwater

The WFD requires that initial characterization Figure 1.7.1 of the location and boundaries of all groundwa- Map indicating loca- ter bodies be carried out in connection with the N tion and boundaries Article 5 report. In addition the initial charac- of aquifers in Odense terization must identify the pressures on the River Basin. groundwater bodies and the character of overlying strata, and must identify those groundwater bodies for which there are directly dependent surface water ecosystems or terrestrial ecosystems. If the initial characterization reveals that spe- ciﬁ c groundwater bodies are at risk of failing to meet the environmental objectives set, these groundwater bodies have to be characterized further. Among other things, their geological and hydrogeological characteristics have to be described. In addition, the chemical composition of the groundwater has to be characterized, as has groundwater recharge and exchange of water between the groundwater body and associated surface systems.

0 5 10 km 1.7.1 Initial characterization

This section describes the initial characterization of aquifers and groundwater abstraction in Table 1.7.1. From Table 1.7.2 it can be seen that Odense River Basin. Subsequently a single aqui- 20 aquifers are smaller than 15 km2, and that 12 fer has been selected for further characterization, of these are smaller than 5 km2. Table 1.7.1 would including subdivision into groundwater bodies. seem to indicate, moreover, that the majority of Odense River Basin is covered by aquifers. This Aquifers is not the case, however, in that the parts of the The location and boundaries of the aquifers aquifers that lie outside Odense River Basin are within Odense River Basin are illustrated in Fig- also included in the total aquifer area. Moreover, ure 1.7.1. The aquifers are included if just part of several aquifers may very well overlap in places, the aquifer lies within Odense River Basin, even separated by impervious clay layers. if only a limited part. As mentioned earlier, a number of the aquifers A summary of the aquifers is shown in Table are only partly located within Odense River Ba- 1.7.1 based on the data for the individual aquifers sin. For example, the major part of aquifers 3, 7N shown in Table 1.7.2 (Fyns Amt, 2000b). Due to and 39 are located outside the basin. These rela- the large range in aquifer size it makes a consider- tively large aquifers comprise approx. 50% of the able difference whether one considers the mean water resource. Abstraction from these aquifers or the median values, as is also apparent from is limited seen in relation to their size, however.

Max Min Median Mean Total Table 1.7.1 Statistics for all aqui- Aquifer area (km2) 187 0.4 9.8 29.7 1 009 fers encompassed by Odense River Basin. Aquifer resource (106 m3) 1 472 0.6 30 146 4 956

Waterworks abstraction in 1998 (103 m3) 4 173 0 152 565 18 627

Total abstraction in 1998 (103 m3) 5 628 0 161 772 25 498

Odense PRB Odense Pilot River Basin 43 1.7 Groundwater ) 3 m 0 1 0 0 1 6 33 82 30 83 90 49 38 161 307 644 188 108 603 281 256 109 157 253 516 160 1 220 1 576 5 628 4 173 1 072 3 378 3 041 1 260 (10 tion in 1998 Total abstrac- Total ) 3 m 0 0 0 0 0 0 6 56 97 13 72 30 83 90 49 34 294 273 188 591 274 256 855 152 253 515 130 1 217 1 246 1 585 4 173 1 800 3 041 1 260 (10 Waterworks abstraction in 1998 1998 in abstraction DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DS DG DS, IS BK, PK PK, BK DS, DG DS, DG DS, DG DS, DS DG, DG DS, DG DS, DG DS, Geology Geology U U C C C C C U C C C C C C C C C C C C C C C C, A A C, A C, U, C U, C U, C U, C U, C U, C U, C U, C U, C U, type Aquifer ) 3 m 6 7.5 9.6 0.6 8.1 9.0 9.6 5.1 3.6 28.2 68.4 16.8 21.6 59.7 18.0 51.3 45.9 64.2 7.05 8.25 63.3 30.9 129.6 223.5 112.5 582.6 148.5 13.95 13.05 821.7 626.4 163.8 111.15 1 472.4 472.4 1 (10 Resource ) 2 4.7 5.6 7.2 2.5 9.3 2.9 4.7 5.5 3.2 0.4 2.7 3.9 0.6 3.2 1.7 1.2 28.8 24.7 74.5 25.0 11.4 33.0 19.9 17,5 11.4 10.2 21.4 21.1 10.3 Area (km 122.7 186.9 182.6 139.2 A1 A1 A1 B1 A1 B2 B1 B1 A1 A3 A1 A1 A3 A2 B2 A1 B2 A3 A2 B2 A2 B2 B3 B1 B1 B1 B2 C2 C2 C1 C2 C1 C2 C2 C1 C1 class Cover layer 3 3 5 3 3 9 5 7 3 5 1 3 3 9 9 3 3 3 7 7 7 7 3 7 5 7 5 5 7 7 7 7 7 9 model Layer in geological 3 5 6 8 9 12 16 17 18 20 23 24 27 35 39 40 41 42 44 45 46 50 60 63 65 66 90 91 93 1V 7N 1Ø 102 105 Aquifer number Aquifer name/identity Tarup-Årslev Tarup-Årslev Langeskov Gislev Ringe-Ryslinge Nyborg Odense near complex Aquifer Holmehave Frøbjerg Vissenbjerg-Blommenslyst Korinth Arreskov Lake at Aquifer Aquifer north of Espe Limestone aquiferOdense – Deep aquifer at Holmehave Stenstrup-Lunde Kværndrup Lake Søndersø aquifer Odense Northern Seden-Bullerup Hasmark Mesinge Munkebo aquifer – upperpart Munkebo aquifer – lower part Morud Vester Hæsinge Tommerup at Aquifer Trunderup Broby Bøge Fruens at aquifer Deep Sanderum of north Aquifer Brylle of north Aquifer Aquifer northwest of Lake Søndersø Kappendrup

Table 1.7.2 Description of the individual aquifers encompassed by Odense River Basin. DS: Late-glacial meltwater sand; DG: Late-glacial meltwater gravel; BK: Danien bryozo limestone; PK: Selandien greensand limestone; IS: Interglacial freshwater sand.

Odense 44 PRB Odense Pilot River Basin 1.7 Groundwater

Class Description Cover layer class Description Table 1.7.3 Classiﬁ cation of cover A The cover layer thickness exceeds A1 A considerable part of the cover layer layer classes. 15 m in less than 25% of the total is less than 5 m thick. area of the aquifer. A2 There are no significant areas where the cover layer is less than 5 m thick, and heavy clay is not present to any great extent above the aquifer. A3 Heavy clay occurs above the aquifer. B The cover layer thickness exceeds B1 Areas are present where the cover 15 m in more than 25% of the total layer is less than 5 m thick. area of the aquifer. In addition, the B2 The cover layer is never less than 5 cover layer is less than 15 m thick in m thick, and only insignificant areas more than 5% of the total area of the of heavy clay occur above the aquifer. aquifer. B3 Significant deposits of heavy clay above the aquifer. C The cover layer thickness exceeds C1 There are no significant deposits of 15 m in more than 25% of the total heavy clay above the aquifer. area of the aquifer. In addition, the C2 Significant deposits of heavy clay cover layer is less than 15 m thick in above the aquifer. less than 5% of the total area of the aquifer.

A geological layer model has been established tion to the aquifers and the degree of exploitation in which layer 1 represents the unsaturated zone, vary considerably. layers 3, 5, 7 and 9 are the water-bearing layers (aquifers), with layer 3 uppermost and layer 9 Groundwater potential lowermost. Layers 2, 4, 6 and 8 are the inter- A potential map for the primary water table has mediate impervious layers, of which layer 2 is been prepared from sounding data (Figure 1.7.2). uppermost. The cover layer classes refer to the As this is primarily based on soundings from the distribution of the cover layer overlying the aqui- primary water table, it is possible that a second- fer (Table 1.7.3). The aquifer types are as follows: ary water table closer to the surface might exist U: Unconfi ned; C: Confi ned; and A: Artesian. locally. As areas without aquifers have also been Certain aquifers contain several types, including assigned a water table, the potential lines cannot both unconfi ned and confi ned. be utilized to accurately determine the magni- With a number of the aquifers it is apparent tude or direction of groundwater fl ow. that abstraction amounts to less than the minimum requirement of 10 m3/day for drinking Aquifer protection water, e.g. aquifers 12, 27, 91 and 102. They are Clay thickness above the aquifer nevertheless included because it is as yet unclear The thickness of the clay overlying the indi- whether groundwater fl ow in these aquifers vidual aquifers is indicated in Figure 1.7.3. The affects terrestrial ecosystems or surface water clay thickness map is not complete for the whole ecosystems. Moreover, the intention is that it of Odense River Basin, however. This is due to should be possible to abstract water from these the fact that well density is too low in some areas aquifers for the drinking water supply. For this to permit estimation of the thickness of the clay reason these four aquifers are also designated as above the aquifer, and that no aquifer is present protected areas (see Section 2). in some areas. In all, 34 aquifers have been characterized. These consist of both unconfi ned, confi ned and Nitrate-vulnerable abstraction areas artesian aquifers, primarily glacial meltwater It is estimated that approx. 303 km2 of Odense sand or gravel aquifers, but also some Tertiary River Basin is vulnerable to nitrate contamina- limestone aquifers. Some are located near the sur- tion (Figure 1.7.4). This fi gure encompasses the face and some are very deep with a very variable areas where groundwater recharge takes place cover layer thickness. As a consequence, infi ltra- and where the aquifer is already contaminated

Odense PRB Odense Pilot River Basin 45 1.7 Groundwater

Figure 1.7.2 (left) Map indicating groundwater potential for the primary water 0 table shown in metres 0 above sea level.

1 5 Figure 1.7.3 (right) 0 0

6 5 1 6 0 Map indicating clay 0 5 0 5 25 45 0 thickness above aqui- 40 3 3 0 fers. 50 5 Over 30 m 5 1 Under 15 m

0 15–30 m 2

Unknown 3 0 3 35 5 clay thickness

0 3

8 9 8 5 0 9 1 0 0 5 4 0 5 7 0 10 5 5 45 0 6 7 5 0 6 0 5 5 675 750 5 3

5 4

5 0 5 5 6 0

0 5 10 km 0 5 10KM

Figure 1.7.4 (left) Map indicating areas vulnerable to nitrate N N contamination.

Figure 1.7.5 (right) Map indicating soil type distribution at a depth of 1 m. Glacial meltwater gravel Glacial meltwater silt Glacial meltwater clay Glacial meltwater sand Post-glacial freshwater clay Post-glacial freshwater gyttja Post-glacial freshwater sand Post-glacial freshwater peat Post-glacial marine gravel Post-glacial marine silt Post-glacial marine sand Glacial moraine gravel Glacial moraine clay Glacial moraine sand 0 5 10 km 0 5 10 km Lake Late-glacial freshwater gravel Late-glacial freshwater clay Late-glacial freshwater sand

Odense 46 PRB Odense Pilot River Basin 1.7 Groundwater

Soil type Area % of Table 1.7.4 (left) 2 even if only a limited part. (km ) area A number of groundwater abstraction areas are Soil type distribution at a depth of 1 m Glacial moraine gravel 9.8 0.9 only partly located within Odense River Basin within Odense River (Figure 1.7.6). These only account for a limited Glacial moraine clay 679.5 65.5 Basin. part of the total amount abstracted, however. Glacial moraine sand 1.8 0.2 The total area of the groundwater abstraction areas is 441.2 km2. Several of the abstraction ar- Glacial meltwater gravel 17.5 1.7 eas are more or less coincident. Glacial meltwater clay 7.3 0.7 In addition to waterworks, groundwater is also abstracted by a large number of farms, Glacial meltwater sand 138.5 13.4 market gardens, industrial enterprises and small Post-glacial freshwater gyttja 40.3 3.9 waterworks (Table 1.7.5). In addition, there are

Post-glacial freshwater clay 1.9 0.2

Post-glacial freshwater sand 6.3 0.6 Permits in Abstraction in Table 1.7.5 (right) Category 3 3 3 3 10 m 1998 in 10 m Abstraction permits Post-glacial freshwater peat 57.0 5.5 Public waterworks 29 090 20 288 and groundwater Post-glacial marine gravel 2.8 0.3 abstraction in Odense Market gardens, 9 458 4 112 River Basin. Post-glacial marine clay 10.9 1.1 farms, etc. Industry 12 520 3 512 Post-glacial marine sand 11.4 1.1 Institutions 146 102 Late-glacial freshwater gravel 0.6 0.1 Minor waterworks 291 59 Late-glacial freshwater clay 1.6 0.2 (2–9 households) Private wells, 1 228 1 228 Late-glacial freshwater sand 49.6 4.8 assessed

with nitrate, or areas where there is poor geological protection against nitrate contamination.

Soil type maps Figure 1.7.6 The distribution of soil type at a depth of 1 m is Map indicating loca- summarized in Table 1.7.4 based on the soil type N tion of groundwater distribution map shown in Figure 1.7.5. abstraction areas The greater part of Odense River Basin is located within or en- covered by moraine clay, while a smaller part compassed by Odense River Basin. is covered by glacial meltwater sand and post- glacial organic-rich soil types. The distribution is uneven in that glacial meltwater sand is more frequent in the western and southern parts of the basin while the late-glacial freshwater sand is more frequent in the northeastern part. The post-glacial organic-rich soils are primarily seen around the River Odense ﬂ oodplain and in ad- joining ﬂ oodplains.

Abstraction and groundwater abstraction areas The ﬁ gures for groundwater abstraction are from 1998. Since then the amount of water abstracted by waterworks has decreased slightly, as has the total abstraction permit allocation. The groundwater abstraction areas have been included if just 0 5 10 km part of the area lies within Odense River Basin,

Odense PRB Odense Pilot River Basin 47 1.7 Groundwater

numerous private wells that supply individual In these areas there is a good chance that surface households. Within Odense River Basin, approx. water ecosystems or terrestrial ecosystems are di- 550 groundwater abstraction permits have been rectly dependent on upwelling from groundwa- granted to farms, market gardens, institutions, ter bodies. These areas are described as upwelling industries and waterworks supplying 2–9 house- areas, and among others include Kulemose Bog, holds. Together with the water abstracted from Stavis Stream, parts of the River Odense and private wells, this accounts for approx. 1/3 of the parts of Ulvebæk Brook. In connection with the total groundwater abstraction in Odense River further characterization of aquifer 8 (see below), Basin. a more detailed map of upwelling areas has been drawn up for that aquifer. Monitoring Under the provisions of the WFD, groundwater Conclusion bodies from which more than 36 500 m3/year is The overall assessment is that in view of the pres- abstracted for drinking water must be monitored. sure of land use and nutrient loading described in As this is the case for most aquifers it is conclud- Sections 3 and 4 and of the geological conditions ed that monitoring will have to be carried out at described in this section, many of the aquifers most of the aquifers listed in Table 1.7.2. are at risk of failing to meet the objective set and must therefore be considered to be threatened. Upwelling The aquifers therefore need to be characterized The areas where the groundwater potential lies further in order to more precisely determine the above ground surface are shown in Figure 1.7.7. magnitude of the risk.

Table 1.7.6 Contaminant Objective Limit value cf. Remarks Groundwater quality Groundwater Directive objectives and limit Nitrate 25 mg/l 50 mg/l The limit value applies to all ground- values for nitrate and water bodies apart from those in areas pesticides. vulnerable to nitrate contamination 1 according to Directive 91/676/EEC . In these areas Article 4 (1)(c) of the WFD applies. Pesticides and pesticide Non-detectable 0.1 µg/l metabolites

Figure 1.7.7 (left) Map indicating location of upwelling areas and watercourses. Watercourse

Upwelling area

Figure 1.7.8 (right) Map indicating location and boundaries of aquifer 8 and urban areas. Urban areas

Aquifer 8

0 5 10 km 0 5 10 km

Odense 48 PRB Odense Pilot River Basin 1.7 Groundwater

Layer Area Resource Geology Waterworks Total Infiltration Groundwater Utilization Table 1.7.7 in Fyn (km2) (106 m3) abstraction in abstraction (mm) recharge per (%) Description of aquifer 8. model 1998 (103 m3) in 1998 year (103 m3) (103 m3)

5 186.9 582.6 DS 1 585 5 628 100 18 691 30

1.7.2 Further characterization

Due to resource constraints, it was decided to re- strict further characterization to a single aquifer Figure 1.7.9 – No. 8, the aquifer complex near Odense. Map indicating cover The characteristics of the aquifer are summa- layer thickness above aquifer 8. rized in Table 1.7.7, and its location and boundaries are illustrated in Figure 1.7.8. As a ﬁ rst step the aquifer was subdivided into groundwater bodies. Thereafter the individual groundwater bodies were characterized. The analysis method followed was that outlined in the horizontal guidance document “Identiﬁ cation of water bodies”.

Subdivision into groundwater bodies Subdivision into groundwater bodies was based on a survey of the aquifers. In order to determine 02.5 5 to what extent the aquifer should be subdivided km into a number of groundwater bodies, the following aspects were assessed: Accumulated clay thickness 15-30m Accumulated clay thickness <15m • Geology Accumulated clay thickness >30m • Groundwater chemistry Unknown clay thickness above aquifer • Flow direction/potential. Figure 1.7.10 Geology Map indicating soil Aquifer 8 is an aquifer complex comprised of type distribution above several smaller contiguous aquifers in meltwater aquifer 8 at a depth of sand. The complex was deposited at approxi- 1 m. mately Danish Zero Level with a general thick- Glacial meltwater gravel ness of 10 m. A belt of the aquifer around Odense Glacial meltwater silt is as much as 35 m thick, though, and in this area Glacial meltwater clay it reaches down to -30 m below Danish Zero Glacial meltwater sand Level. Moreover, geophysical mapping (TEM) Post-glacial freshwater clay indicates that the aquifer is up to approx. 20 m Post-glacial freshwater gyttja thick in the vicinity of Nørre Søby. The aquifer Post-glacial freshwater sand complex is covered by an approx. 10 m thick Post-glacial freshwater peat layer of moraine clay. The individual aquifers are Post-glacial marine gravel both unconfi ned and confi ned. Their boundaries Post-glacial marine silt to the north and south are uncertain, while those Post-glacial marine sand to the west and east are reasonably well defi ned. Glacial moraine gravel The aquifer complex is the most extensive in Glacial moraine clay Fyn County, and is probably in hydraulic contact with the regional Northern Odense Aquifer Glacial moraine sand (No. 40) to the north and possibly with the Vis- Lake Late-glacial freshwater gravel senbjerg-Blommenslyst Aquifer (No. 16) to the 0 2.5 5 km northwest and the Holmehave Aquifer to the Late-glacial freshwater clay west. It is provisionally assigned to layer 5 in the Late-glacial freshwater sand

Odense PRB Odense Pilot River Basin 49 1.7 Groundwater

Table 1.7.8 Area Type 2 % to analyse the potential data. The groundwater Distribution of cover (km ) potential map composed using the model is layer thickness above Clay thickness unknown 46 25 shown in Figure 1.7.14. Comparison of this with aquifer 8. the previously established potential map shows Clay thickness <15 m 97 52 that the two maps are in good agreement. The Clay thickness 15–30 m 41 22 ﬁ gure shows that aquifer 8 can be subdivided into a number of smaller areas on the basis of poten- Clay thickness >30 m 2 1 tial data for the primary water table.

Groundwater bodies geological layer model. Based on the above it is concluded that the aqui- The distribution of cover layer thickness above fer needs to be subdivided into six groundwater aquifer 8 is shown in Figure 1.7.9 and Table bodies in order to be able to characterize them in 1.7.8. a uniform manner, namely groundwater bodies The distribution of soil types within aquifer 8 8-1 to 8-6 (Figure 1.7.14). The subdivision was is indicated in Table 1.7.9, while the location of initially based on groundwater chemistry data. soil types within aquifer 8 is illustrated in Figure Final subdivision was made with the help of the 1.7.10. modelled potential map. Table 1.7.10 indicates In the southeastern part of the aquifer clay which of the groundwater bodies differ from the cover above the aquifer is thicker than in the re- expected natural state. mainder of the aquifer. Moreover, meltwater and From Table 1.7.10 it can be concluded that freshwater deposits predominate in the south- the problem facing the six groundwater bodies western part in contrast to glacial deposits in the differs. Thus it is not possible to group two or remainder of the aquifer. more of the groundwater bodies on the basis of their chemical properties. They are therefore Groundwater chemistry described individually. In order to assess the need to subdivide aquifer 8 into groundwater bodies an analysis of the Description of groundwater bodies groundwater chemistry data was made encom- The characteristics and status of each of the six passing: groundwater bodies into which aquifer 8 has been subdivided are described below. • Nitrate • 2,6-dichlorbenzamid (BAM) Geology • Conductivity. The area and geology of each groundwater body is shown in Table 1.7.11. In this connection, a simple model was established (Triangular irregular network model). Hydrogeological conditions The model is based on interpolation between The hydrogeological properties of each of the original data points in a network of triangles. groundwater bodies are summarized in Table These models are shown in Figure 1.7.11 (ni- 1.7.12. The hydraulic conductivity is based on the trate), Figure 1.7.12 (2,6-dichlorbenzamid) and geological description of the groundwater body. Figure 1.7.13 (conductivity). The reason that the Negative values for the depth of the groundwa- individual maps do not encompass the same area ter table indicate that the groundwater potential is that not all three of the selected parameters are lies above ground surface. analysed for in each of the selected wells. The resultant maps show that clear aggrega- Clay thickness above the groundwater bodies tions are apparent for certain parameters. For The distribution of clay thickness above the in- example, raised nitrate concentrations are seen dividual groundwater bodies is shown in Table in groundwater bodies in the southwestern and 1.7.13 based on the clay thickness map for aquifer northeastern areas, while lower concentrations 8. are seen in the northwestern and southeastern areas. Raised BAM concentrations are seen in the Effects of groundwater on surface water eco- northern part of the area. systems Groundwater affects both the quantitative and Flow direction/potential qualitative status of surface water ecosystems. A triangular network model was also established In this section an attempt is made to identify

Odense 50 PRB Odense Pilot River Basin 1.7 Groundwater

Figure 1.7.11 (left) Nitrate concentration 8-1 8-1 map based on well measurements. The map was composed 8-3 using a triangular 8-3 8-2 8-2 irregular network model. The locations 8-4 of the wells used for 8-4 the triangulation are indicated. 8-6 8-6 8-5 8-5 Figure 1.7.12 (right) 2,6-dichlorbenzamid (BAM) concentration 02.5 5 map based on well km measurements. The map was composed using a triangular Site of analysis Site of analysis irregular network over 50 mg/l over 10.0 µg/l 44-50 mg/l model. The locations 1.0-10.0 µg/l of the wells used for 39-44 mg/l 0.1-1.0 µg/l 33-39 mg/l 0.05-0.1 µg/l the triangulation are 28-33 mg/l 0.01-0.05 µg/l indicated. 22-28 mg/l 17-22 mg/l 11-17 mg/l 6-11 mg/l 0-6 mg/l

0 Figure 1.7.13 (left) 40 4 5 Conductivity map 1 0

3 5 8-1 5 8-1 based on well meas- 0 5 urements. The map

0 3 was composed using a triangular irregular 8-3 8-3 8-2 8-2 network model. The

5 locations of the wells 1 used for the triangula- 8-4 8-4 tion are indicated.

5 3 8-6 8-6 8-5 8-5 Figure 1.7.14 (right)

3 2 5 0

0 4 Groundwater poten- 4

30 6 tial map composed 5

0 0

5 5 using a triangular

0 2.5 5 0 2.5 5 2 irregular network 40 3 5 45 5 km km 5 50 7 model. The locations 55

0 5 6 of the data points used for the triangulation Site of analysis Measurement points Potential line are indicated. Subdivi- over 1930 µS/cm sion of the aquifer into 1750-1930 µS/cm Interpollated potential map groundwater bodies is 1580-1750 µS/cm 15 m 30 m 1400-1580 µS/cm also indicated. 1230-1400 µS/cm 5m -6 m 1050-1230 µS/cm 10 m 50 m 870-1050 µS/cm 20 m 45 m 700-870 µS/cm -5 m 520-700 µS/cm 35 m 350-520 µS/cm 0m 40 m 25 m 55 m

Odense PRB Odense Pilot River Basin 51 1.7 Groundwater

Table 1.7.9 Area surface water ecosystems that are dependent on Soil type 2 % Soil type distribution (km ) groundwater. within aquifer 8. Glacial moraine gravel 0.1 0.0 The areas in which the groundwater potential lies above the ground surface were determined Glacial moraine clay 114.8 61.7 from terrain altitude relative to Danish Zero Glacial moraine sand 0.1 0.0 Level in combination with potential data for the primary water table. These areas were compared Glacial meltwater gravel 2.7 1.5 with the areas known from the soil type distri- Glacial meltwater clay 0.2 0.1 bution map to consist of post-glacial freshwater sediments. From Figure 1.7.15 it can be seen that Glacial meltwater sand 29.5 15.9 by far the majority of the upwelling areas are Glacial variable moraine layer 0.0 0.0 located in areas with post-glacial freshwater sediments. From Table 1.7.14 it can also be seen that Post-glacial freshwater gyttja 10.5 5.7 the percentage of each groundwater body that consist of upwelling areas varies considerably. Post-glacial freshwater clay 1.0 0.5

Post-glacial freshwater sand 1.5 0.8 Conclusion Based on the groundwater chemistry data it Post-glacial freshwater peat 14.0 7.5 seems appropriate to subdivide aquifer 8 into Late-glacial freshwater gravel 0.5 0.3 six groundwater bodies in order to be able to describe status. Late-glacial freshwater clay 0.5 0.3 The characterization revealed that fi ve of the Late-glacial freshwater sand 10.5 5.6 groundwater bodies are at risk of failing to meet the objective. This aspect is more closely examined in Section 4.1, where the groundwater chemical status (environmental effect) is described for the individual groundwater bodies. The environ- Table 1.7.10 Groundwater Nitrate BAM Conduc- mental pressures on the individual groundwater Subdivision of aquifer body tivity bodies are described in Section 3.4.1. 8 into groundwater 8-1 bodies on the basis of groundwater chemis- 8-2 try. The grey, shaded 1.7.3 Provisional establishment of objec- areas indicate where 8-3 tives the water chemistry differs from the expect- 8-4 The WFD does not directly specify criteria for ed natural state. 8-5 assessing the quality of the groundwater. The criteria have been subsequently specifi ed in the 8-6 Groundwater Directive. In Fyn County an objective has been set regarding the pesticide content of the groundwater, namely that pesticides and pesticide metabolites may not be present in the groundwater. Among Table 1.7.11 Groundwater Area 2 Geology other things, the background for this objective body (km ) Area and geology of the is that it must be possible to base the drinking individual groundwa- Quaternary meltwater 8-1 15.09 water supply on uncontaminated groundwater. ter bodies of aquifer 8. sand/gravel Fyn County has also set an objective for the Quaternary meltwater 8-2 34.51 sand/gravel nitrate content of the groundwater, namely that Quaternary meltwater the nitrate content may not exceed 25 mg/l in 8-3 24.53 sand nitrate-contaminated groundwater zones in the Quaternary meltwater identifi ed groundwater bodies. 8-4 29.06 sand These objectives differ from the criteria es- Quaternary meltwater tablished for assessing good chemical status of 8-5 55.74 sand groundwater in the Groundwater Directive, Quaternary meltwater 8-6 27.98 namely 0,1 µg/l for pesticides and their relevant sand metabolites, and 50 mg/l for nitrate. The objectives and the limit values for good status are shown in Table 1.7.6.

Odense 52 PRB Odense Pilot River Basin 1.7 Groundwater

Groundwater Distance from the terrain to Hydraulic conductivity (m/s) Annual varia- Source of Table 1.7.12 body the primary groundwater tion in ground- groundwater Hydrogeological prop- table (m) water level (m) recharge erties of the individual Mean Max Min Mean Min-Max groundwater bodies of aquifer 8. 8-1 6.0 29.3 -9.0 5.0E-3 5.0E-5–1.0E-2 No data Precipitation

8-2 5.9 27.1 -14.6 5.0E-3 5.0E-5–1.0E-2 0.5–2 Precipitation

8-3 5.7 24.4 14.7 5.0E-4 5.0E-5–1.0E-3 0.5–2 Precipitation

8-4 2.4 22.0 -12.8 5.0E-4 5.0E-5–1.0E-3 0.5–2 Precipitation

8-5 2.8 23.3 -9.5 5.0E-4 5.0E-5–1.0E-3 0.5–2 Precipitation

8-6 3.4 32.9 -8.7 5.0E-4 5.0E-5–1.0E-3 0.5–2 Precipitation

Groundwater body 8-1 8-2 8-3 8-4 8-5 8-6 Table 1.7.13 Distribution of clay Area % Area % Area % Area % Area % Area % thickness above the individual groundwater Clay thickness unknown 6.1 40.2 12.9 46.2 7.1 29.1 0.7 2.3 14.5 26.0 12.9 46.2 bodies of aquifer 8. Clay thickness <15 m 8.0 53.2 10.0 35.7 12.3 50.2 21.7 74.5 21.2 38.0 10.0 35.7

Clay thickness 15–30 m 1.0 6.6 5.0 18.0 5.1 20.8 6.3 21.6 19.3 34.6 5.0 18.0

Clay thickness >30 m 0.0 0.0 0.0 0.1 0.0 0.0 0.5 1.6 0.8 1.4 0.0 0.1

Ground- Area Area with up- % area with Table 1.7.14 (left) water body (km2) welling (km2) upwelling Distribution of up- 8-1 15.1 0.9 5.7 welling areas in the individual groundwater 8-2 34.5 2.7 7.7 bodies of aquifer 8.

8-3 24.5 2.3 9.4 Figure 1.7.15 (right) 8-4 29.1 5.5 18.9 Map indicating areas of aquifer 8 where the pri- 8-5 55.7 8.5 15.3 mary water table lies above ground level, as 8-6 28.0 7.1 25.4 well as areas consisting 8 186.9 27.0 14.4 of post-glacial freshwater sediments.

Post-glacial freshwater sediments 0 2.5 5 km Upwelling area

Odense PRB Odense Pilot River Basin 53 Photo: Bjarne Andresen, Fyns Amt Seden Strand and Odense Fjord.

Odense 54 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

1.8 Transitional waters and coastal waters

1.8.1 Odense Fjord Northern Belt Sea Figure 1.8.1 ODF22B Map of Odense Fjord Odense Fjord is a shallow (average depth 2.25 indicating the bound- m) mesohaline estuary located in the northern aries of the inner fjord part of Fyn (Figure 1.8.1). The submerged area (Seden Strand) and outer fjord. The three encompasses 63 km2. The fjord can be subdivided monitoring stations, into a smaller inner fjord, Seden Strand, which SS8 in Seden Strand, has an average depth of 0.8 m and accounts for a ODF17 in the outer little less than 1/4 of the submerged area, and an part of the fjord and E g outer part which has an average depth of 2.7 m e n ODF22B in the boun- s G OTTERUP e a D b and accounts for the remaining approx. 3/4. e Odense e dary zone outside the e t p Fjord fjord are indicated. By far the major source of riverine runoff to the Outer fjord is the River Odense, which ﬂ ows into the fjord inner part of Seden Strand. The water exchange ODF17 between the fjord and the open sea (Northern MUNKEBO Belt Sea) takes place via Gabet, the narrow mouth KERTEMINDE Seden Ø

Strand e g i of the fjord. Using a hydrodynamic model it has t S SS8 l been calculated that the residence time of River Depth na Ca 30-40 m se n l R Odense water is short, around a yearly average 20-30 m de a n i O a v 10-20 m C e ld r of 17 days for the fjord as a whole and 9 days for 6-10 m O O d 4-6 m e n Seden Strand. 0 5 km 2-4 m s ODENSE Odense Harbour e Nutrient loading of the fjord from anthropo- 0-2 m genic sources is high due to the large size of the catchment (1 060 km2), which corresponds to one third of the area of Fyn and is characterized by intensive agricultural production and a high ﬂ uorescence are measured electronically every population density (see Section 1.1). Thus in 2001, 10–20 cm along the water column; the oxygen for example, total loading from the catchment and concentration is alternatively measured using the the atmosphere amounted to approx. 32 g N (or Winkler technique. Transparency is measured 2.3 mol N) and approx. 0.88 g P (or 0.03 mol P) using a Secchi disc. per m2 of fjord, with atmospheric deposition only Nutrient concentrations – total nitrogen, am- accounting for 5% or less. Given this high level of monium, nitrate (+nitrite), total phosphorus, loading, Odense Fjord can be characterized as a orthophosphate and silicate – are measured in eutrophic water body. surface water samples and in the bottom water at Both national and local initiatives initiated in the deeper stations. the late 1980s have helped reduce this high level Phytoplankton conditions are evaluated by of loading, and the point-source dominated P load measurement of chlorophyll a at several depths has since been reduced by a factor of 6–7. The N and of primary production (C-14 method). At load, which is predominantly attributable to dif- ODF17 in addition, carbon biomass is measured fuse loading from farmland, has only been reduced at species and taxa level. by around 1/3 though, (see also Section 3). A number of other variables are monitored concomitantly with these pelagic activities, but Monitoring programme with varying frequency. These are: Benthic in- Pelagic variables are monitored weekly at three vertebrate number, biomass and ﬁ ltration capac- stations representative of the fjord and its border ity; macroalgal and rooted macrophyte species zone: SS8 (depth 0.8 m) in the shallow inner composition, coverage and (to a lesser extent) fjord, Seden Strand, ODF17 (depth 8.5 m) in the biomass; nutrient and oxygen exchange across outer deep part of the fjord, and ODF22B (depth the sediment-water interface; continual hydro- 17 m) in the border zone outside the fjord (Figure graphic measurements for use in a deterministic, 1.8.1). hydrodynamic 3D-model, which is coupled to a The physical variables monitored are salinity, dynamic eutrophication model; and the content temperature and light attenuation, which togeth- and effects of hazardous substances in water, er with oxygen concentration and chlorophyll sediment and biota.

Odense PRB Odense Pilot River Basin 55 1.8 Transitional waters and coastal waters

Hydrographic conditions and nutrient condi- due to the high runoff, and are highest (median tions in Odense Fjord 3 500–4 000 µg N/l) in the inner fjord near the The seasonal variation in the salinity of the fjord main freshwater outﬂ ow, decreasing towards the and border zone is illustrated in Figure 1.8.2 open sea, i.e. approx. 1 400 µg N/l in the outer based on data for the past seven years. The sa- fjord and approx. 350 µg N/l outside the fjord. In linity in Seden Strand varies during the year be- the summer the concentrations of total nitrogen tween approx. 10 and 15 PSU (monthly median), are lower, the median values being around 650, being lowest in winter/spring when riverine run- 450 and 250 µg N/l, respectively. off is highest. The salinity is higher in the outer Corresponding estuarine winter gradients also fjord (16–21 PSU), and highest outside the fjord occur for total phosphorus (Figure 1.8.4). In this (16–24 PSU). The water temperature at the three case the winter median is approx. 110, 50 and 30 stations varies between approx. 0 and 25ºC. µg P/l in the inner fjord, outer fjord and border The estuarine gradient in nutrient concentra- zone, respectively. A special feature of the nutri- tions is illustrated in Figure 1.8.3 (nitrogen) and ent cycle in the fjord is the almost bell-shaped 1.8.4 (phosphorus). The concentrations of total course of phosphorus concentration over the nitrogen are generally higher in the winter period summer, which also comprises the yearly maxi-

Salinity, surface Total N, surface PSU µg N/l Figure 1.8.2 (left) 90% quantile SS8 7 000 90% quantile SS8 Surface water salinity 24 Median 1996-02 Median 1996-02 at three stations inside 10% quantile 6 000 10% quantile and outside Odense 20 Fjord (see Figure 1.8.1) 5 000 16 for the period 1996– 4 000 2002. The median and 12 10% and 90% quan- 3 000 8 tiles are shown together 2 000 with all the data. 4 1 000

0 0 JFMAMJJASONDJ JFMAMJJASONDJ

PSU µg N/l Figure 1.8.3 (right) Surface water con- 90% quantile ODF17 90% quantile ODF17 28 Median 1996-02 Median 1996-02 4 000 centration of total 10% quantile 10% quantile nitrogen at three 24 stations inside and 3 000 outside Odense Fjord 20 (see Figure 1.8.1) for 2 000 the period 1996–2002. 16 The median and 10% 1 000 and 90% quantiles are 12 shown together with all the data. 8 0 JFMAMJJASONDJ JFMAMJJASONDJ

PSU µg N/l

90% quantile ODF22B 700 90% quantile ODF22B 32 Median 1996-02 Median 1996-02 10% quantile 600 10% quantile 28 500 24 400 20 300 16 200

12 100

8 0 JFMAMJJASONDJ JFMAMJJASONDJ

Odense 56 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

Total P, surface mum. The highest median concentrations are µg P/l found in the inner fjord, reaching a maximum Figure 1.8.4 90% quantile SS8 Surface water con- of approx. 200 µg P/l as compared with approx. 500 Median 1996-02 centration of total 100 µg P/l in the outer fjord. The concentration 10% quantile phosphorus at three 400 peak is primarily attributable to release from stations inside and the sediment iron-bound phosphorus pool and outside Odense Fjord 300 (to a lesser extent) point-source discharges of (see Figure 1.8.1) for phosphorus and poorer water exchange during 200 the period 1996–2002. the summer. The phosphorus peak is not seen in The median and 10% the surface water outside the fjord; on the other 100 and 90% quantiles are hand, oxygen defi cit-dependent accumulation of shown together with all phosphorus (and nitrogen) often occurs through- 0 the data. out the summer and autumn in the bottom layer JFMAMJJASONDJ of these stratifi ed open marine waters. µg P/l During the growth season, nutrient concentra- 90% quantile ODF17 tions in the fjord are potentially limiting for the 160 Median 1996-02 phytoplankton, although exhibiting temporal 10% quantile and spatial variation (data not shown). Potential P limitation generally exists in the spring, and 120 potential N limitation in the summer. Moreover, nutrient limitation of the phytoplankton is gen- 80 erally more pronounced in the outer fjord (where silicate can also be potentially limiting for dia- 40 toms) than in Seden Strand. Not surprisingly, the greatest potential nutrient limitation exists at the 0 considerably lower nutrient levels found in the JFMAMJJASONDJ open marine waters outside the fjord. µg P/l 90% quantile ODF22B Biological conditions in Odense Fjord 50 Median 1996-02 Despite the high nutrient loading of the fjord, 10% quantile pelagic phytoplankton biomass, which is domi- 40 nated by diatoms, is relatively low (data not shown). At the same time, the seasonal devel- 30 opment of the phytoplankton biomass is very 20 dynamic, with rapidly shifting concentrations of chlorophyll a. 10 Apart from the fact that water exchange is high, one of the main reasons for the relatively 0 low phytoplankton abundance and the dynamic JFMAMJJASONDJ seasonal development is that the diverse benthic invertebrate fauna of Odense Fjord is dominated by fi lter feeders, i.e. polychaetes and mussels (Figure 1.8.5), which fi lter the phytoplankton from the water column. It is calculated that alone the daily fi ltration capacity of the dominant polychaete Neries diversicolor is several times Benthic fauna Figure 1.8.5 that of the water volume in Seden Strand, and Benthic fauna in approximately half the water volume in the outer Apportioned by species Apportioned by individuals Odense Fjord apportioned by number of fjord. To what extent the fi ltration potential is 29% 41% species and individu- expressed is highly dependent on the physical als. Mean of 50 sta- mixing of the water bodies. 24% tions, 1998–2001. That nutrient loading of the fjord is high gen- 10% erally favours growth of the rapidly growing Molluscs ephemeral macroalgae: For example, sea lettuce 2 % Crustaceans 11% Echinoderms (Ulva lactuca) dominates in Seden Strand. In 13% Polychaetes step with a demonstrated fall in nutrient loading 36% Other since the 1980s (see Figure 4.5.1), the abundance 34%

Odense PRB Odense Pilot River Basin 57 1.8 Transitional waters and coastal waters

of other ephemeral macroalgae with a lower Optional factors: nutrient requirement has been increasing in 4) Wave exposure Seden Strand, and the abundance of perennial 5) Mean depth macroalgae such as bladderwrack (Fucus vesicu- 6) Further geomorphological subdivision of losus) and of rooted macrophytes such as eelgrass fjords: Inner fjords, sill fjords and sluice (Zostera marina) and widgeon grass (Ruppia mar- fjords. itima) has been increasing in the outer fjord. The macrophyte community in the fjord is still very This typology yields 18 types of coastal water, of unstable, though, and still exhibits large interan- which 16 are found in Denmark. nual variation in coverage. Among other things, According to the proposed national typology eelgrass has virtually disappeared in Egense Deep Odense Fjord has to be differentiated into two in the northwestern part of the outer fjord since types of water body based on data from the the mid 1990s, and the overall eelgrass coverage monitoring stations SS8 in the inner fjord and was found to be very low during the last survey ODF17 in the outer fjord: in 2002 (see also Section 4.5.2). • Odense Fjord, inner part, denoted Seden Strand: Inner fjord, mesohaline (7–18 PSU). 1.8.2 Typology • Odense Fjord, outer part: Shallow (0–3 m mean depth), mesohaline (7–18 PSU) fjord. Ecoregion The Danish coastal waters are located within The border zone of Odense Fjord located outside two of the ecoregions defi ned in the WFD, the the mouth, Gabet, will be incorporated in the North Sea ecoregion and the Baltic Sea ecoregion description and establishment of quantitative cri- (cf. WFD Annex XI, map B). The boundary be- teria for Odense Fjord. According to the national tween the two regions runs from Sjællands Odde geomorphological subdivision of fjords, this zone through the southern Kattegat to Djursland. In is comprised of a third type of water body when the Habitats Directive, the boundary between using monitoring data from station ODF22B: the Atlantic bioregion and the Baltic bioregion runs along the Weichsel ice front line, i.e. from • Odense Fjord, border zone: “Inner marine Hirtshals to the southern coast of Norway. In water between Djursland/Sjællands Odde and the context of both the WFD and the Habitats Fyns Hoved/Røsnæs, depth <15m, salinity Directive, Odense Fjord will thus belong to 15–20 PSU”. Ecoregion 5, the Baltic Sea ecoregion, but with close association with the North Sea ecoregion in Three types of coastal water have thus been iden- the context of the WFD. The international ma- tifi ed in the Odense River Basin. rine convention OSPAR encompasses both open Delineation of water bodies outside the fjord marine waters and fjords, while the Helsinki cannot be fi nalized until the geographic bounda- Convention only encompasses open marine wa- ries of the new administrative units “River ba- ters. In Danish marine waters the OSPAR Con- sins” with associated coastal waters have been vention encompasses the area north of the WFD established for the forthcoming River Basin outer boundary (North Sea ecoregion), while the District Fyn County (Annex 0.2). The area Helsinki Convention encompasses the Baltic Sea will be encompassed by the WFD’s administra- region as well as the whole of the Kattegat up to tive delineation of fjords and/or coastal waters a line between Skagen and the western coast of within one nautical mile of the so-called baseline Sweden at Gothenburg. The two conventions from which the breadth of territorial waters are thus overlap in the Kattegat area. measured (cf. Nielsen et al., 2001; Danish En- vironmental Protection Agency, 1983b; Annex Typology 0.2). The status of these water bodies has to be The provisional national proposal for a typology classifi ed relative to reference conditions estab- of coastal waters (Nielsen et al., 2001) follows lished on the basis of biological quality elements. System B, with a number of optional factors, and The remainder of the border zone between 1 and operates with the following descriptors: 12 nautical miles from the baseline might have to be assigned to a coastal water body that, pursuant Obligatory factors: to the WFD, only has to be characterized relative 1) Geomorphological differentiation into main to physical/chemical quality elements. coastal waters 2) Tidal range 3) Salinity.

Odense 58 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

1.8.3 Delineation of water bodies Figure 1.8.6 Delineation and area In the context of the WFD a surface water body of the fi ve marine has to be a “discrete and signifi cant” element of Boundary zone water bodies encom- surface water that can be characterized and as- 74 km² passed by Odense sessed in relation to the environmental objectives River Basin, the inner fjord, NW, mid and set. A water body may only consist of a single NE outer fjord, and N type of water body, and can only hold a single the boundary zone. See ecological status class. According to the EU also Table 1.8.2. Guidance Document “Identifi cation of water ODF NW bodies”, protected areas have also to be taken OTTERUP 14 km² into account, and subdivision into several water ODF NE 19 km² bodies can be considered in cases where only part ODF mid of a water body is designated as a protected area. 13 km² The application of the above-mentioned typol- ODF inner/ MUNKEBO ogy based on morphological/hydrographical KERTEMINDE Seden Strand criteria yields three types of coastal water in 17 km² Odense River Basin and hence a minimum of three water bodies. Odense Fjord is also encompassed by both international and regional protection provisions. The borders of these areas do not ODENSE follow either the geographical subdivisions of the 0 1 2 3 km fjord or the above-mentioned morphological differentiation into types of fjord (Figure 1.8.8). Further subdivision of the outer part of Odense Fjord into three water bodies will thus Northern Belt Sea Figure 1.8.7 be necessary in order to be able to assess status Map of Odense Fjord and to assess whether the fjord meets the various indicating formerly submerged but now re- degrees of objective. claimed areas, location In all, fi ve coastal water bodies have thus been of the dykes and major identifi ed in Odense River Basin, namely Seden physical installations Strand, the outer fjord subdivided into Odense such as harbours, pow- Fjord northwest, Odense Fjord mid and Odense er stations, shipyards, Fjord northeast, and fi nally the border zone ad- Egense Harbour fairways, etc. E g e joining Odense Fjord. The delineation and areas n G OTTERUP s e a Reclaimed areas D Odense b of the fi ve individual water bodies are indicated e e e t p Fjord in Figure 1.8.6. Outer Dykes fjord Bregnør Harbour Klintebjerg Harbour Physical installations Boelsbro 1.8.4 Physical modifi cation Lindø Terminal Lindø ShipyardMUNKEBO Odense KERTEMINDE North landfill Seden Strand Harbours and navigation Stige Ø landfill Odense Fjord is connected to Denmark’s third largest city, Odense, via Odense Canal (Figure Fynsværket, CHP Plant 1.8.6). Since ancient times it has been a hub for Odense Harbour maritime trade and transport. Thus there are ODENSE 0 5 km a number of minor harbours and quays on the fjord. Odense Harbour, Lindø Shipyard and Lindø Terminal represent installations of considerable size that have entailed major physical in the form of large amounts of resuspended changes to the fjord’s natural coastline (Figure sediment that reduces the transparency of the 1.8.7). Moreover, shipping fairways have been water column and causes sedimentation on the excavated in connection with the harbours, and seabed. Periodic intensive navigation to and from these have to be dredged every few years (Figure Odense Harbour and Lindø Shipyard with large 1.8.7). This dredging work periodically places vessels also causes resuspension of sediment and marked physical pressure on the fjord, especially reduces the transparency of the water.

Odense PRB Odense Pilot River Basin 59 1.8 Transitional waters and coastal waters

Other physical installations Outer parts of Odense Fjord The physical environment of the fjord is also The main physical changes in the form of dyking affected by a number of other large installa- and harbour construction have long been com- tions located on the fjord. Stige Ø Landfi ll pleted, and the water balance in the fjord can has considerably increased the area of Stige Ø thus be expected to have reached equilibrium. peninsula which, after cessation of landfi lling In themselves, these changes do not hinder the and subsequent re-establishment, now forms a fjord in meeting the WFD objective of good eco- hilly peninsula projecting into the inner part of logical status. Moreover, as the recurrent work of Seden Strand. Since its construction, the intake dredging the fairways is carried out at intervals of cooling water by Fynsværket Combined Heat of many years, its impact is not expected to be and Power (CHP) plant has caused changes to the suffi cient to hinder attainment of good ecological physical environment in the form of temperature status either. and salinity changes, altered water exchange and enhanced mortality of planktonic organisms in Seden Strand. 1.8.5 Reference conditions

Dyking and land reclamation As described in Section 1.8.3, differentiation and Due to the shallow nature of the fjord and the fl at delineation resulted in Odense Fjord being subdi- topography of the catchment, major dyking and vided into an inner fjord (Seden Strand), an outer land reclamation works have been carried out in fjord comprised of three discrete water bodies, the fjord right up to the 1960s. This considerably and a border zone. reduced the area of the fjord, and long portions According to the WFD and the Guidance of the physical coastline were replaced by dykes Document on classifi cation of coastal waters (Figure 1.8.7). In addition to the physical chang- (GD 2.4), establishment of the reference condi- es, drainage and dyking considerably reduced tions can be based on data from existing water the capacity of the fjord and catchment to retain bodies assessed as being in an undisturbed state, nitrogen input from the catchment. on historical data, on modelling of historical and other data and on expert judgement. There are Shell mining no coastal water bodies in Denmark that can be The great abundance of calcareous oyster shells presently characterized as “undisturbed”. The in the seabed of the shallow parts of the fjord establishment of reference conditions for Odense were exploited right up to the 1970s. The min- Fjord has therefore been based partly on 3D mod- ing was conducted in large parts of Seden Strand elling of Odense Fjord carried out in connection right down to a depth of 4 m, and thus undoubt- with the cooling water intake and discharge edly comprised a major physical pressure. permits for Fynsværket (CHP) plant, partly on historical data on the macrophyte vegetation in Heavily modifi ed water bodies Odense Fjord around 1900, at which time an- Seden Strand thropogenic nutrient loading of the outer part Based on the guidelines given in the EU Guid- of Odense Fjord and the border zone, but not ance Document “Heavily modifi ed water bodies” Seden Strand, is considered to have been low, and it is concluded that the above-mentioned physi- partly on empirical modelling of these historical cal pressures are not of suffi cient magnitude to data. Selection of the variables described below justify classifi cation of all or part of the fjord as was thus based on what was rendered possible by heavily modifi ed. historical data and modelling within the provi- The intake and discharge of cooling water sions of the WFD concerning biological vari- by Fynsværket (CHP) plant cause changes in ables and supporting physico-chemical variables. the hydromorphological conditions in Seden It should be noted that the variables described Strand. In particular, water exchange, salinity below should not be considered as the fi nal list of and temperature are raised relative to the natural criteria for establishing objectives. In the further state (see Section 1.8.5). It is estimated that these work, the description of reference conditions changes are not of suffi cient magnitude to hinder will be supplemented with additional criteria for the fjord in meeting the objective of good ecolog- biological variables, especially pertaining to phy- ical status, however. This assessment will have to toplankton and benthic invertebrates, as well as be revised, though, if the above-mentioned con- additional criteria for supporting physico-chemi- ditions change, i.e. if the power station’s current cal variables to describe hazardous substances, discharge permit is altered. heavy metals, etc.

Odense 60 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

Under the WFD, a status class is only con- ous criteria for reference conditions are examined sidered to be attained if the criteria for all the below and in Table 1.8.1. biological variables are met (one out, all out principle). As a result of this strict requirement, the Reference conditions for the inner fjord, Seden EU guidance documents propose that variables Strand subject to large natural variation – which is often Macrophytes the case with the marine waters – can be excluded In 1957 there was a well-developed community from the array of variables included in the defi ni- of eelgrass (Zostera marina), widgeon grass (Rup- tion of reference conditions. As the data material pia maritima) and sago pondweed (Potamogeton for coastal waters is inadequate to establish many pectinatus) in the northern part of Seden Strand, detailed criteria for their biological state, it is while in the southern part there was an abun- essential to maintain that the criteria in combi- dance of sea lettuce (Muus, 1967). The present- nation must still provide the same broad picture day distribution of vegetation in Seden Strand is of the biological status, and that all the criteria examined in Sections 1.8.1 and 4.5.2, and illus- established must be fulfi lled. trated in Figure 4.5.3. The WFD does not specify over which time Based on historical data from other similar period these established criteria must be fulfi lled, fjords (Petersen, 1892; 1901), 3D modelling (DHI, however. As demonstrated in Roskilde Fjord, for 2002; 2003) and the studies of Muus (1967) and example (Andersen et al., in press), assessment Larsen (unpublished manuscript), it is concluded of status based on a single monitoring year will that reference conditions entail a considerably often lead to random variation in status clas- lower abundance of eutrophication-dependent sifi cation. Assessment of status should therefore ephemeral macroalgae (in Seden Strand, primari- be carried out in accordance with statistically ly sea lettuce and fi lamentous algae) concentrated well-founded guidelines that provide the best in the immediate vicinity of the main freshwater possible assessment over time in relation to the inputs than is presently the case. Using 3D mod- coming six-year reporting period, cf. also the rec- elling the reference conditions for macroalgae ommendations based on the existing experience has been quantifi ed for two different years (1998 with quantitative quality criteria and assessment and 2002) in so-called “Natural state scenarios”. of objective fulfi lment in the current recipient Based on these values the reference conditions quality planning (Nielsen et al., 2001). The vari- for ephemeral macroalgae has been established

Table 1.8.1 Odense Fjord Reference conditions Water body Type Reference status criteria for the ﬁ ve marine water bodies en- Secchi Macrophytes Nutrients compassed by Odense depth River Basin. Coverage (%) Biomass Depth distri- Total N Total P 2 (m) (depth interval) (g C/m ) bution (m) (µg/l) (µg/l) Filamentous Eelgrass Widgeon grass algae and Eelgrass sea lettuce Inner fjord (Se- 1 >80 (1.5–4 m) >80 (0.5–1.5 m) <10**** >4 <660* <30* - den Strand)

Outer fjord NW 2 >80 (1–6 m) >80 (0.5–1 m) <3*** >6 <370** <10** >7.2

Outer fjord NE 2 >80 (1–6 m) >80 (0.5–1 m) <3*** >6 <370** <10** >7.2

Outer fjord mid 2 >80 (1–6 m) >80 (0.5–1 m) <3*** >6 <370** <10** >7.2

Boundary zone 3 >80 (1–10 m) >80 (0.5–1 m) <3*** >10 <146 <5 >10.6

Remarks: The values apply for the individual water bodies as a whole. * The nutrient criteria are set at a salinity of 12.2 PSU. ** The nutrient criteria are set at a salinity of 18.1 PSU. Total N is up- and down-regulated by 49.3 µg/l total N per PSU with a decrease or increase in salinity, respectively. Total P is up- and down-regulated by 3.4 µg/l total P per PSU with a decrease or increase in salinity, respectively. *** Only anchored filamentous algae; freely floating filamentous algae must not be present. **** New calculations could lead to the value being lower.

Odense PRB Odense Pilot River Basin 61 1.8 Transitional waters and coastal waters

as an average for the whole of Seden Strand as solely established for the period 01/03–31/10, the the sum of the seasonal maximum biomass of the total N values derived have to be corrected for various species. The reason for operating with an the higher winter concentrations. This has been average value for the whole of Seden Strand is the done on the basis of ratios calculated from meas- physically determined passive mobility of these urements made by Fyn County over the period macroalgae. The resultant reference value is 7 g 1981–2002. C/m2 for 2002 and 10 g C/m2 for 1998, which In addition, total N concentrations have been is in good agreement with calculations based on calculated by DHI (2002; 2003) in “Natural state data in Larsen (unpublished manuscript). scenarios” based on two different meteorological The distribution of the rooted macrophyte situations in 1998 and 2002. The annual mean vegetation in reference conditions is considered concentration of total N at station SS8 was cal- not to differ markedly from the current distribu- culated to be 670 µg N/l in 1998 and 730 µg N/l tion, but the biomass and stability of vegetation in 2002. These values are comparable with the coverage would be considerably greater than is empirically determined value used as the crite- presently the case, and the rooted macrophytes rion for reference conditions, i.e. 660 µg N/l (see should thus occur in dense stands. The variables below). operated with for the rooted macrophytes are No corresponding empirical relationships ex- depth distribution and coverage. The reference ist for total phosphorus (total P). However, as values have been established on the basis of with total N, 3D modelling of “Natural state vegetation surveys made by Fyn County and his- scenarios” yields an annual mean concentration torical data from Odense Fjord (Ostenfeld, 1908; for total P at station SS8 of 30 µg P/l in 1998 and Fyn County, 1990; 2003b), as well as historical 50 µg P/l in 2002. However, new calculations are data from corresponding fjord waters (Nielsen et currently being made with the model in which al., 2003a). riverine nutrient loading has been corrected. It is expected that these calculations will yield a Macrophyte criteria for reference conditions: somewhat lower “natural state” nutrient level in • Depth distribution (main community) of eel- the fjord. For total P, the lower of the two above- grass: ≥4 m where depth in the area permits. mentioned values has therefore been selected as • Coverage of eelgrass: >80% in the depth inter- the criterion for reference conditions. val 1.5–4 m. As the nutrients fl ow into the fjord with the • Coverage of widgeon grass: >80% in the fresh water, and fl ow out of the fjord with water depth interval 0.5–1.5 m. consisting of a mixture of fresh water and the sea • Biomass (seasonal maximum) of sea lettuce water fl owing into the fjord, the nutrient content and freely fl oating fi lamentous algae (average in the fjord follows an estuarine gradient and for the whole of Seden Strand): <10 g C/m2 varies markedly depending on the actual mixing conditions at the time of measurement (see also Nutrients Section 1.8.1). It will therefore be necessary to Seden Strand is presently overfertilized by nutri- relate the nutrient criteria for reference condi- ents from the catchment (Sections 3.4 and 4.5.2, tions to salinity, i.e. to normalize the nutrient Figure 4.5.1). concentrations for differences in salinity. This With respect to nutrient concentrations, the also eliminates “noise” from variation in salinity reference conditions for Seden Strand should be when interpreting the monitoring results. The established such that the requirement for undis- relationships established for variations in total P turbed or virtually undisturbed biological state and salinity are not as good as those established can be met. for total N and salinity, primarily due to sedi- The criteria established for eelgrass depth dis- ment release of phosphorus during summer. It tribution in Seden Strand (and in the outer fjord has been decided to also employ the relationship and border zone; see below) of an undisturbed/ established for total P, though, as it is considered virtually undisturbed state is based on data for that the establishment of the operational criteria the depth distribution of eelgrass early last cen- would be problematic without the correction. tury (Ostenfeld, 1908). For total nitrogen (total N), the reference Nutrient criteria for reference conditions: conditions are hereafter calculated by empirical • At an annual mean salinity of 12.2 PSU at modelling of the relationship between total N station SS8 in Seden Strand, the annual mean and the depth distribution of eelgrass (Nielsen et concentration of total N must not exceed 660 al., 2002). As the empirical relationship has been µg N/l (empirical modelling for the period

Odense 62 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

01/03–31/10 regulated 40.5% up to an annual fjord), but that this is not used in establishing the mean value based on measurements over the reference conditions as the presence of such algae period 1981–2002). is incompatible with the historical data. • Total N is regulated up or down by 49 µg N/l for each PSU change in salinity relative to the Macrophyte criteria for reference conditions: baseline of 12.2 PSU (the 3D model yields • Depth distribution (main community) of eel- comparable values for salinity correction of grass: ≥6 m. 57–58 µg N/l per PSU). • Coverage of eelgrass: >80% in the depth inter- • At an annual mean salinity of 12.2 PSU at val 1–6 m. station SS8, the annual mean concentration of • Coverage of widgeon grass: >80% in the total P must not exceed 30 µg P/l (dynamic 3D depth interval 0.5–1.0 m. modelling, 1998 scenario). • Eutrophication-dependent algae must not oc- • Total P is regulated up or down by 3.4 µg P/l cur in a freely fl oating state. for each PSU change in salinity relative to the • Coverage of anchored fi lamentous algae: baseline of 12.2 PSU. <10% at all depth intervals. • Biomass of anchored fi lamentous algae (aver- Reference conditions for Odense Fjord, outer age for the whole outer fjord): <3 g C/m2. part The reference conditions for the three water bod- Nutrients ies into which the outer fjord has been subdivid- Nutrient loading of the outer fjord from the ed (see Section 1.8.4) is considered to be common catchment is presently considerable (Sections 3.4 as their hydrography are common and loading and 4.5.2, Figure 4.5.1). mainly derives from diffuse sources rather than With respect to nutrient concentrations, the point sources. reference conditions for the outer fjord should be established such that the biological state of Macrophytes the water body will be undisturbed or virtually Petersen (1892; 1901) and Ostenfeld (1908) de- undisturbed. scribed the depth distribution of eelgrass in the For total N, the reference conditions are outer fjord early in the last century and found hereafter calculated by empirical modelling of this to be 6–7 m. In reference conditions, eelgrass the relationship between total N and the depth thus covered the majority of the outer fjord, in- distribution of eelgrass (Nielsen et al., 2002). The cluding the deep northeastern basin and the deep resultant values for total N have been corrected channel Egense Deep in the northwestern part. as described above for Seden Strand. In addition, the rooted macrophytes should gen- The total N concentration has also been erally occur in dense communities. calculated by DHI (2002; 2003) for the above- The current distribution of vegetation in the mentioned “Natural state scenarios”. The annual outer fjord is described in Sections 1.8.1 and mean concentration of total N at station ODF17 4.5.2. (1 m) was calculated to be 260 µg N/l in 1998 It is assessed that the reference conditions and 320 µg N/l in 2002. These values are lower will correspond to the distribution of rooted than the empirically determined value used as the macrophytes early in the last century and that criterion for reference conditions, i.e. 370 µg N/l the occurrence of fi lamentous algae will also be (see below). limited to anchored specimens, such as described No corresponding empirical relationships exist for other corresponding fjords (Petersen, 1898). for total P. However, as with total N, 3D model- The biomass of anchored fi lamentous algae in ling of “Natural state scenarios” yields an annual reference conditions has been established on the mean concentration for total P at station ODF17 basis of measurements made by Fyn County in (1 m) of 10 µg P/l in 1998 and 20 µg P/l in 2002. such areas, as well as quantitative quality criteria New calculations currently being made with cor- established for other Danish fjords. The biomass rected riverine nutrient loading are expected to was converted to C from dry matter using the yield a somewhat lower calculated “natural state” relationships in Larsen (unpublished manuscript) nutrient level in the fjord. For the time being, and Fyn County (unpublished data). It should be the lower of the two above-mentioned values for noted that 3D modelling of “Natural state sce- total P is therefore used. This is supplemented narios” (DHI, 2002; 2003) yields a relatively low with correction of the nutrient concentrations seasonal peak biomass of freely fl oating macro- for salinity as described for Seden Strand. algae of 1.2–1.5 g C/m2 (average for the outer

Odense PRB Odense Pilot River Basin 63 1.8 Transitional waters and coastal waters

Nutrient criteria for reference conditions: relationships in Larsen (unpublished manuscript) • At an annual mean salinity of 18.1 PSU at and Fyn County (unpublished data). station ODF17 (1 m) in the outer fjord, the annual mean concentration of total N must not Macrophyte criteria for reference conditions: exceed 370 µg N/l (empirical modelling for • Rooted macrophytes must be present in dense the period 01/03–31/10 regulated 34% up to communities (>80% coverage): an annual mean value based on measurements • Depth distribution (main community) of eel- over the period 1981–2002). grass: ≥10 m. • Total N is regulated up or down by 49µg N/l • Coverage of eelgrass: >80% in the depth inter- for each PSU change in salinity relative to the val 1–10 m. baseline of 18.1 PSU. • Coverage of widgeon grass: >80% in the • At an annual mean salinity of 18.1 PSU at depth interval 0.5–1.0 m. station ODF17 (1 m) in the outer fjord, the an- • Eutrophication-dependent algae may not oc- nual mean concentration of total P must not cur in a freely fl oating state. exceed 10 µg P/l (dynamic 3D modelling, 1998 • Coverage of anchored fi lamentous algae: scenario). <10% at all depth intervals. • Total P is regulated up or down by 3.4 µg P/l • Biomass of anchored fi lamentous algae (aver- for each PSU change in salinity relative to the age for the whole of the outer fjord): <3 g baseline of 18.1 PSU. C/m2.

Secchi depth Nutrients The reference conditions for Secchi depth are With respect to nutrient concentrations in the also established from the relationship between water, the reference conditions for the border the depth distribution of the vegetation early in zone should be established such that the biologi- the last century (Ostenfeld, 1908) and the empirical state of the water bodies will be undisturbed cal relationship between this depth distribution or virtually undisturbed. and the necessary transparency of the water dur- The reference conditions for total N were ing the growth season (Nielsen et al., 2002). The established on the basis of empirical modelling Secchi depth is presently approx. 3 m (summer as described above (although without any correc- mean). tion for salinity) In the 3D modelling of the “Natural state Secchi depth criterion for reference conditions: scenarios” the nutrient concentrations in the • 7.2 m (mean, March–October). marine border (i.e. the present border zone) were reduced relative to the current concentrations by Reference conditions for the border zone backwards extrapolation of N and P values from Macrophytes Kieler Bight for the period 1960–1989 (Helsinki Petersen (1892; 1901) and Ostenfeld (1908) de- Commission, 1990). For nitrogen the reduction scribed the depth distribution of eelgrass in the was 67%, thus yielding total N values of 94–97 border zone north of Odense Fjord early in the µg N/l for the modelling years 1998 and 2002 last century and found this to be 9–10 m. In (33% of the actual annual mean values measured reference conditions, eelgrass thus occurred in a those two years in the surface water at station dense, broad belt around the coast and reached ODF22B). This is somewhat lower than the out to a depth of approx. 10 m. empirically determined value used as the crite- It is assessed that the reference conditions will rion for reference conditions, i.e. 146 µg N/l (see correspond to the distribution of rooted macro- below). phytes early in the last century, and that the oc- Useful empirical relationships do not exist for currence of ﬁ lamentous algae will be limited to establishing the reference conditions for total P. anchored specimens, such as described for other Using a similar backwards extrapolation for total corresponding fjords (Petersen, 1892; 1901). The P at station ODF22B (1 m) as that used for total biomass of anchored ﬁ lamentous algae in refer- N above in the “Natural state scenarios” yielded ence conditions has been established on the basis an annual mean concentration of 25% of the of measurements made by Fyn County in such measured value (75% reduction), corresponding areas, as well as quantitative quality criteria es- to a very low level of 5–6 µg P/l for the two tablished for other Danish fjords. The biomass years. was converted to C from dry matter using the New calculations are currently being made,

Odense 64 PRB Odense Pilot River Basin 1.8 Transitional waters and coastal waters

however, and it is expected that these calcula- Northern Belt Sea Figure 1.8.8 tions will yield a somewhat lower “natural state” Map of Odense Fjord level for the nitrogen concentration in the border indicating areas E g e n encompassed by the zone, but not for the phosphorus concentration. s G OTTERUP e a D b quality objective e Odense e e t p Nutrient criteria for reference conditions: Fjord "Reference area for Outer scientifi c studies" in • The annual mean concentration of total N at fjord Fyn County's Regional station ODF22B (1 m) must not exceed 146 Plan for 2001–2013. µg N/l (empirical modelling for the period Areas designated as MUNKEBO 01/03–31/10 regulated 3.7% up to an annual KERTEMINDE international protec- Seden mean value based on measurements made by Strand tion sites (Natura 2000 Fyn County over the period 1981–2002). sites, which in Odense Fjord encompass sites • The annual mean concentration of total P at Depth station ODF22B (1 m) must not exceed 5 µg 30-40 m protected under the 20-30 m Habitats Directive and P/l (dynamic 3D modelling, 1998 scenario). 10-20 m 6-10 m Birds Directive) are 4-6 m ODENSE 0 5 km 2-4 m also indicated. Secchi depth 0-2 m The reference conditions for Secchi depth are Reference site also established from the relationship between Natura 2000 site the depth distribution of the vegetation early in the last century (Ostenfeld, 1908) and the empirical relationship between this depth distribution and the necessary transparency of the water during the growth season (Nielsen et al., 2002). The tion Agency’s remarks on the draft bill on imple- Secchi depth is presently approx. 7.5 m (summer mentation of parts of the WFD and the Habitats mean). Directive, the overall aim for the marine waters is to transfer the current objectives directly such Secchi depth criterion for reference conditions: that areas with a general (basic) quality objective •9.0 m (mean, March–October). shall in future have to fulfi l “good ecological status” pursuant to the WFD. With regard to areas assigned a stringent quality objective in the 1.8.6 Provisional establishment of objec- current recipient quality plans, or areas where tives the state is actually better than the criteria for good status, the county authorities shall ensure Due to the natural qualities of Odense Fjord, the that these higher objectives are preserved. For northwestern part at Egense Deep has been des- Odense Fjord, this will entail that the area at ignated as a “reference area for scientifi c studies” Egense Deep that is currently assigned a high (stringent quality objective) in Fyn County’s Re- (stringent) quality objective will have to fulfi l the gional Plan for 2001–2013. The quality objective WFD requirements for “high ecological status”, for the remainder of the fjord is “fi sh waters for while the remainder of the fjord will have to ful- angling and/or fi shery” and, where the natural fi l the requirements for “good ecological status”. conditions permit, “spawning and/or nursery A large part of the fjord is presently designated grounds for fi sh” (general/basic quality objec- as a Natura 2000 site pursuant to the Habitats Di- tive). rective and Birds Directive, yet is only assigned a Moreover, the whole of the inner fjord, Seden general quality objective in the Regional Plan. Strand, and the western part of the outer fjord, In order to ensure adequate protection of the has been designated as an international nature nature qualities that justifi ed designation as a protection area (Natura 2000), see Figure 1.8.8. Natura 2000 site, i.e. achievement of good pres- The fjord is also encompassed by a number of ervation status for species and habitat types se- other protection regulations (see Section 2 on lected pursuant to the Habitats Directive or Birds protected areas, Figure 2.6 and 2.7). Directive, these areas will in future have to meet The quality objectives stipulated in the Re- at least “good ecological status” pursuant to the gional Plan serve as the basis for establishing WFD. It is as yet unclear whether consideration environmental objectives pursuant to the WFD. for the species that comprise the basis for designa- According to the Danish Environmental Protec- tion of sites in the fjord pursuant to the Habitats

Odense PRB Odense Pilot River Basin 65 1.8 Transitional waters and coastal waters

Table 1.8.2 Area Present Future Water body Type 2 HM The ﬁ ve marine water (km ) objective objective bodies encompassed by Inner fjord (Seden Strand) 1 14 No General, P G Odense River Basin indicating type and area ODF NW 2 14 No Stringent, P H and present and future objective. See Figure Outer fjord ODF NE 2 19 No General G 1.8.6 for delineation of ODF Mid 2 13 No General, P G the water bodies. Boundary zone 3 74 No General G

P: International Natura 2000 protected site (Habitat Site, EC Bird Protection Site). H: High ecological status. G: Good ecological status. General: Regional Plan objective that all coastal waters should be suitable as fish waters for angling/fishery and, where the natural conditions permit, suitable as spawing and/or nursery areas for fish. Stringent: Regional Plan objective that the recipient should be a reference area for scientific studies. HM: Heavily modified water body. Type 1: Inner fjord, salinity <18 PSU. Type 2: Fjord, salinity 7–18 PSU, mean depth 0–3 m Type 3: Inner marine waters between Djursland/Sjællands Odde and Fyns Hoved/Røsnæs, salinity 15–20 PSU, depth <15 m.

Directive and the Birds Directive, as well as for preservation status (Dahl et al., in press; see also their habitat conditions such as food resource in Section 2). the form of macrophyte vegetation, water level Areas of Odense Fjord that are designated as in wetlands etc. can entail requirements different Natura 2000 sites, but which have not been as- from those that have otherwise to be fulﬁ lled to signed the highest quality objective in the cur- meet good ecological status (Søgaard et al., 2003). rent recipient quality plans, are thus assigned In the case of certain marine habitat types, it is the provisional environmental objective “good considered necessary to achieve “high ecologi- ecological quality” (i.e. Seden Strand and ODF cal status” in order to be able to achieve good mid; Table 1.8.2 and Figure 1.8.6).

“LIV II” Fyn County’s surveillance vessel.

Photo: Fyn County

Odense 66 PRB Odense Pilot River Basin