The INCA-Pathogens model: An application to the Loimijoki River basin in Finland

STOTEN-19947; No of Pages 11 Science of the Total Environment xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

The INCA-Pathogens model: An application to the Loimijoki River basin in Finland K. Rankinen a,⁎, D. Butterfield b, M. Faneca Sànchez c, B. Grizzetti d, P. Whitehead b, T. Pitkänen e,1, J. Uusi-Kämppä f, H. Leckie b a

Finnish Environment Institute (SYKE), Mechelininkatu 34a, FI 00250 Helsinki, Finland University of Oxford, School of Geography and the Environment, South Parks Road, Oxford, OX1 3QY, UK Deltares, Princetonlaan 6-8, 3584 CB Utrecht, P.O. Box 85467, 3508 AL Utrecht, The Netherlands d European Commission Joint Research Centre (JRC), via Enrico Fermi 2749, 21027 Ispra, VA, Italy e National Institute for Health and Welfare (THL), Water and Health Unit, Neulaniementie 4, FI 70700 Kuopio, Finland f Natural Resources Institute Finland (Luke), Tietotie 4, FI 31600, Jokioinen, Finland b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• We present the new INCA-Pathogens model application to Finnish river basin. • We simulate the effect of climate and land use changes on pathogens in the river water. • We also discuss about self-purification capacity of waters as an ecosystem service.

a r t i c l e

i n f o

Article history: Received 30 November 2015 Received in revised form 6 May 2016 Accepted 6 May 2016 Available online xxxx Keywords: Faecal bacteria Mathematical modelling Water quality monitoring Catchment scale Hygienic quality

a b s t r a c t Good hygienic quality of surface waters is essential for drinking water production, irrigation of crops and recreation. Predictions of how and when microbes are transported by rivers are needed to protect downstream water users. In this study we tested the new process-based INCA-Pathogens model in the agricultural Loimijoki River basin (3138 km2) in Finland, and we quantified ecosystem services of water purification and water provisioning for drinking and recreation purposes under different scenarios. INCA is a catchment scale process based model to calculate pollutant transfer from terrestrial environment and point sources to the catchment outlet. A clear gradient was observed in the numbers of faecal coliforms along the River Loimijoki. The highest bacterial counts were detected in the middle part of the main stream immediately after small industries and municipal sewage treatment plants. In terms of model performance, the INCA-Pathogen model was able to produce faecal coliform counts and seasonality both in the low pollution level sampling points and in the high pollution level sampling points. The model was sensitive to the parameters defining light decay in river water and in soil compartment,

⁎ Corresponding author. E-mail address: katri.rankinen@ymparisto.fi (K. Rankinen). 1 Present address: National Institute for Health and Welfare (THL), Water and Health Unit, P.O. Box 95, FI 70701 Kuopio, Finland

http://dx.doi.org/10.1016/j.scitotenv.2016.05.043 0048-9697/© 2016 Elsevier B.V. All rights reserved.

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

2

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

as well as to the amount of faecal coliforms in the manure spread on the fields. The modeling results showed that the number of faecal coliforms repeatedly exceeded 1000 bacteria 100 ml−1. Moreover, results lead to the following conclusions: 1) Climate change does not cause a major threat to hygienic water quality as higher precipitation increases runoff and causes diluting effect in the river, 2) Intensification of agriculture is not a threat as long as animal density remains relatively low and environmental legislation is followed, 3) More intensive agriculture without environmental legislation causes a threat especially in tributaries with high field percentage and animal density, and 4) Hygienic water quality in the River Loimijoki can best be improved by improving sewage treatment. We conclude that this catchment scale model is a useful tool for addressing catchment management and water treatment planning issues. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In classical water quality management the concentration of faecal indicator bacteria is a measure of water safety for consumption, irrigation, or recreational use. Faecal indicator organisms include bacteria which naturally occur in the gastrointestinal tract of mammals, and are correlated to the presence of waterborne disease-causing pathogens which are usually difficult and expensive to measure directly (Myers et al., 2014). In a Nordic setting, the most relevant waterborne human pathogens include noroviruses and Campylobacter jejuni, which both have feacal origin and a low infective dose (Zacheus and Miettinen, 2011; Hokajärvi et al., 2014). Noroviruses originate mainly from sewage, while C. jejuni is common in the faeces of all warm-blooded animals (Pitkänen, 2013). The three major indicator bacteria groups for hygienic water quality are total coliforms, faecal coliforms and faecal enterococci (Chapra, 1997). Total coliforms are not specific indicators of faecal pollution, as they occur in the faeces of mammals but also in both polluted and unpolluted soils and waters (Myers et al., 2014). Faecal coliforms are a subgroup of total coliforms that represent better the bacteria of the intestinal tract of mammals. In recent decades, Escherichia coli, a member of the faecal coliform group, has replaced the use of faecal coliforms in water quality monitoring. E. coli originates almost solely from faeces and thus it is often considered as the best indicator of recent faecal contamination of water (Tiwari et al., 2015). Faecal enterococci (also known as intestinal enterococci, previously also classified as streptococci) are found in the intestines of humans and other mammals, but at lower numbers than coliforms, and have faecal and non-faecal sources. Before the recent development and use of the host-specific genetic source identifiers (Pitkänen et al., 2013), the ratio of faecal coliforms to faecal enterococci (FC/FS) has been used to estimate the sources of contamination (Chapra, 1997). It has been claimed that if bacteria originates from human faeces, FC/FS ratio is over four, but in case of animal faeces, the FC/FS ratio is lower than one. In drinking water distributed from waterworks for the human consumption the quality standard for E. coli and coliform bacteria is b1 bacteria per 100 ml (STM, 2000). However, in private wells in Finland the presence of up to 100 colony forming units (CFU) of coliform bacteria other than E. coli per 100 ml is accepted (STM, 2001; Pitkänen et al., 2015), highlighting the fact that coliform bacteria may originate from environmental sources while E. coli is considered to indicate fresh faecal pollution (Tallon et al., 2005). Bathing water quality thresholds for a single sample from an inland bathing site are 1000 bacteria per 100 ml for E. coli and 400 bacteria per 100 ml for intestinal enterococci (STM, 2008). In Finland the monitoring of indicator bacteria counts in water started in the 1960's. The long-term bacterial monitoring time series are based on the use of the historical hygiene parameters: total coliforms, faecal coliforms and faecal enterococci. In agricultural areas the numbers of indicator bacteria often exceeded the limit of acceptable bathing water especially during high flow seasons (Niemi and Niemi, 1991). Even in pristine areas the indicator bacteria were detected in

about half of the samples, sometimes in counts up to 100 bacteria per 100 ml water. The origin was assumed to be wild animals (Niemi and Niemi, 1991). There are no national quality standards for hygienic quality of purified waste waters in Finland and therefore the hygienization of the wastewater effluents is rare. Outside of the centralized sewerage systems, the wastewater treatment systems of private houses may act as sources of human derived faecal microbes into the water bodies in rural areas (Kauppinen et al., 2014). The other significant source of faecal pollution, agriculture, is indirectly regulated by Nitrates Directive (EEC, 1991) and Finnish Agri-Environmental Programme (FAEP). Nitrates Directive regulates fertilizer use mainly by controlling manure spreading on fields. According to the national implementation of the Nitrates Directive, only 170 kg total N in manure is allowed in one year, and the last date to spread manure in autumn is 15th of October. In FAEP manure spreading during growing season is encouraged. Also relatively strict limits for P fertilization in FAEP may limit manure use as fertilizer. Manure application is not allowed on field areas with an average slope of over 15% (Finlex, 2014). Water bodies are now impacted by a complex mix of stressors resulting from urban and agricultural land use, and climate change, rather than a major single stressor like point sources. One impact is the microbial contamination of water and some of these microbes can be pathogenic to humans (Hokajärvi et al., 2013). In order to understand multiple stressor impacts on pathogen numbers in river waters, scientific studies on sources, transport and survival dynamics of microbes are required to develop informed health risk assessments, policy reforms and alternative land-use options (Kay et al., 2010; Meays et al., 2004; Oliver et al., 2016). A wide range of processes affects the transport and retention of microbes in rivers. In addition to natural mortality of bacteria, sunlight may inactivate them (Chapra, 1997; Kadir and Nelson, 2014) or they can be lost to the bottom sediments via settling (Chapra, 1997; Drummond et al., 2015). Thus inactivation depends on environmental condition, such as temperature, sunlight, physical and chemical water properties as well as runoff (Blaustein et al., 2013; Kadir and Nelson, 2014; Sokolova et al., 2012). This self-purification capacity is relevant in the context of ecosystem services (MA Millenium Ecosystem Assessment, 2005) where this contributes to the production of clean water (water purification and the prevention of waterborn disease (pest and disease control). Further, good water quality is essential for other ecosystem services, such as drinking water provisioning and recreational activities. One of the goals of the conceptualization of ecosystem services is to make their key role in supporting multiple human benefits more visible. Several classifications and conceptual frameworks have been proposed to analyse ecosystem services, such as the Millennium Ecosystem Assessment (MA, 2005), the Economics of Ecosystems and Biodiversity (TEEB, 2010), and the Common International Classification of Ecosystem Services (CICES, Haines-Young and Potschin, 2013). The application of ecosystem services concepts in water resource management is relevant to highlight the benefits of conservation and restoration of aquatic ecosystems (Grizzetti et al., 2016).

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

Climate change may alter environmental conditions. In Finland annual precipitation is expected to increase by 13–26% and temperature by 2–6 °C by the end of the century, with the increases expected to be greater in winter than in summer (Jylhä et al., 2009; Ruosteenoja and Jylhä, 2007). A climate change modelling study (Veijalainen et al., 2010) found that this change in climate to cause a significant shift in the seasonality of runoff and floods, with increasing floods during autumn and winter, and diminishing floods in spring especially in southern and central Finland. These changes may increase pathogen leaching from fields as in field trials the highest number of bacteria in runoff water was measured after broadcasting slurry to wet soil followed by rainfall (Uusi-Kämppä and Heinonen-Tanski, 2008), conditions that may become more common in the future. Further, climate change is projected to favour Finnish agriculture within the coming decades (Peltonen-Sainio et al., 2010) due to the prolongation of the growing season and intensification may increase the pollution levels. When planning the measures to maintain high hygienic water quality or to restore larger water courses the origin and fate of bacteria have to be known. Statistical catchment scale models are based on monitoring data of water quality in the intensively studied catchments and then relating that to land cover data and other environmental variables (Kay et al., 2010; Kay et al., 2005; Kirschner et al., 2009; Schoonover and Lockaby, 2006). The use of these models is typically restricted to areas with corresponding land use and climate. The process-based models take into account in addition to the loading from sources the processes that determine microbe inactivation and transport in river and in some models also in terrestrial environment (Dorner et al., 2006; Ferguson et al., 2007; Servais et al., 2007). More detailed models describe the processes between river sediments (Wilkinson et al., 1995; Yakirevich et al., 2013). The use of genetic markers together with catchment scale models has enabled more accurate apportionment between different diffuse sources (Frey et al., 2013; Sokolova et al., 2012). Thus processbased models are more relevant option than statistical models for scenario analysis in cases where current land use or climate does not apply any more. In this study we tested the new process-based INCA (Integrated Nutrients in CAtchments) for Pathogens model (Whitehead et al., 2015) in a large river basin with both human and agricultural influence. INCA-Pathogens is a general model to describe microbes fate and transport in catchments. In this study our main interest was to test the model behavior in a Nordic setting in Finland. We used faecal coliforms as indicator bacteria for faecal pollution and calibrated the model against long time series of faecal coliforms observed in several tributaries in the river basin. We analyzed the sources and the fate of faecal coliforms in three different storylines representing possible development in future, and we considered how the ecosystem services of water purification and water for drinking and recreation purposes could be affected. The main aim was to test the potential of a dynamic catchment scale model in the river basin water treatment planning, and the effect of measures on ecosystem services in the future. 2. Material and methods 2.1. The Loimijoki River basin The Loimijoki River basin (3138 km2) is located in South Western Finland (Fig. 1), where agricultural land use has been traced back to the first centuries Anno Domini. The main soil type in the river basin is clay Vertic Luvic Stagnosols (FAO, 2014). Fields cover 38% of the catchment area, and main crops are cereals (34%), grass ley (3%) and special crops, like sugar beet (1%). There are also cattle breeding farms, mainly pig and poultry farming. Animal husbandry has been intensified during the last decade. At the lower reaches of the river the demand for clearing new fields to spread manure has increased in the last decade due to high animal density (Niskanen and Lehtonen, 2014).

3

There are also some areas with small-scale industry and several small but relatively densely populated settlements in the river basin. All together there are nine municipal sewage treatment plants discharging their effluents into the river basin. Local authorities have made plans to build a new central sewage treatment plant to collect and purify sewage from a larger area. The recreational value of the Loimijoki River valley has increased since the 1980′s when water quality started to get better due to improved waste water treatment in municipalities and industry. Nowadays there are both recreation fishing places and a canoeing route along the river. In the upper reaches of the river there is the Liesjärvi national park (21 km2). In the river basin there are also several small NATURA2000 areas. Over the period 1981–2010, the mean annual precipitation in the area was 600–650 mm and the mean annual temperature was 4.0 to 5.0 °C. In the Loimijoki River the mean daily flow was 24 m3 s−1 the maximum discharge was 328 m3 s− 1 and the minimum discharge 0.74 m3 s−1 during the same period. There are daily discharge measurements in three stations along the main river, and altogether 17 water quality sampling points. In 1995–2009 sampling density was 2–10 samples per year, the highest sampling density being in the main river. Faecal coliforms and faecal enterococci were enumerated using standardized culture-based membrane filtration methods on mFC agar medium at 44 °C for 22 h and on m-Enterococcus agar medium at36 °C for 44 h, respectively. Table 1 shows the characteristics of the 17 sampling points which location can be seen in Fig. 1. 2.2. The INCA-Pathogens model The INCA model is a process based model to calculate pollutant transfer from terrestrial environment and point sources to the catchment outlet (Futter et al., 2007; Wade et al., 2002; Whitehead et al., 1998; Whitehead et al., 2015:Lazar et al., 2010). With the dynamic and process based nature of the model, variations in rainfall and temperature, and changes of inputs, such as purified waste water discharges and manure from livestock, can be investigated in space and time across the catchment. This means the effects of climate change, population growth, land use change, and mitigation measures on the numbers of pathogens in the environment can be evaluated. The model constants can vary on a sub-catchment basis and according to soil or land-use type. These two factors allow the mass stored, process rates, hydrological pathways to vary spatially based on preconceived notions of variations in soil moisture, temperature, adsorption potential and land management. The terrestrial compartment of the model has three water storages: quick flow, soil water and ground water. Within the soil zone it is assumed the water can be partitioned into two volumes: drainage and retention. The drainage volume represents the water stored in the soil that responds rapidly to water inflow and drains under gravity; it may be thought of as macropore or drain flow (i.e. the flow that most strongly influences the rising hydrograph limb). The soil zone retention volume represents the water stored or retained in the soil after gravity drainage; it responds more slowly than the drainage water and represents the majority of water in the soil. The amount of ground water and soil water discharging to the river is defined by base flow index. The INCA-Pathogens Model has been designed to simulate the transport pathways and fluxes of generic pathogens in the land, water column, riverbed sediment, and groundwater phases. The following equation relates to the pathogens included in the model. The full set of equations is described in detail by Whitehead et al. (2015). The generic structure of microbe numbers (Min) is: Min ¼ manures ðeither aerial spreading or injection applicationÞ þ atmospheric deposition þ livestock=animal inputs þ wild animal inputs:

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

4

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

Fig. 1. Location of the Loimijoki river basin and its sub catchments.

The reach mass balance includes the upstream water quality together with diffuse inputs from the soil and groundwater zones, as well as direct effluent discharges and abstractions. A key concept in the water

column is that faecal bacteria are not all bound to sediments and it can float freely in the stream. However, the bacterial cells can be attached to the particles and agglomerate to form cell clusters which

Table 1 Animal density and point sources in the sub catchments. Sub-catchment

35.98 35.93 35.97 35.92 35.96 35.99 35.94 35.95 35.91 a

Main reach upper Main reach Main reach Main reach middle Tributary 1 Tributary 2 Tributary 4 Tributary 3 Main reach outlet

Fields

Animal densitya

[%]

[AU ha

22 40 43 59 48 60 42 42 42

0.5 0.5 0.6 1.4 0.6 2.2 1.2 1.4 3.4

−1

]

Manure application −1

[t ha 5 5 10 15 10 25 10 15 32

Point source input

Water quality sampling station

No No No Yes No Yes Yes No Yes

– Loimijoki 106 – Lojo 54 Koejoen suu Niini 16 Punkalaidun 20 Palojoen suu Lojo 68

FC/FS

] 3.88 2.53 1.81 3.92 2.02 1.56 3.66

Animal unit (AU). Cow corresponds to 1 AU, heifer to 0.4 AU, sow with piglets to 0.5 AU, other pigs to 0.3 AU, poultry to 0.003 AU.

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

can be deposited on the stream bed (Chapra, 1997). It is difficult to determine how the bacterial cells move from one state to another and then disperse again. Thus the sediment equations (e.g. Lazar et al., 2010) are not directly included in the model structure, though pathogens may behave like fine sediments. Bacteria are also subject to die off in the water column and this decay can be temperature and solar radiation dependent. The automatic multisite calibration and sensitivity analysis was performed by the model-independent Parameter ESTimation tool (PEST). The optimization problem is iteratively solved by linearizing the relationship between a model output and its parameters by using a Taylor series expansion. Parameter ranges, initial values and parameter increments must be given by the user. After each iteration the solution of the linearized problem is taken as the optimal set of parameters. The parameter vector is updated at each step using the Gauss-MarquardtLevenberg algorithm (Levenberg, 1944; Marquardt, 1963). The derivatives of model outputs (daily simulation of discharge and microbe numbers) with respect to its parameters provide also a measure of the parameter sensitivities.

2.3. Model set up to the Loimijoki River basin Land use of the Loimijoki River basin was based on the CORINE 2000 land cover data. The amount of cattle and other domestic animals was based on the annual statistics and the proportion of different crops (Table 2) was defined by detailed field parcel data provided by the Information Centre of the Ministry of Agriculture and Forestry (http:// www.mmmtike.fi/en/front-page.html). Meteorological data from Jokioinen Observatory was used as input data (Supplementary material 1). Global radiation was available from year 1997, and in two previous years default data based on geographical location was used. The setup procedure is explained in detail by Rankinen et al. (2009). All the manure produced in the area was assumed to be applied primarily to spring cereals with secondarily to grasses. The amount of manure applied in autumn was assumed to reach the maximum amount (e.g. 20 tn ha−1 cattle slurry or 15 tn ha−1 pig slurry) allowed, according to the Nitrate directive. This is a common practice in the region because farmers empty manure storages before winter. In areas of low animal density (≤1.4 animal units ha−1) only autumn spreading was assumed, but in areas of high animal density (N 1.4 animal units ha−1) also spring spreading was allowed as the amount of manure exceeded the limits of Nitrate Directive. The Loimijoki River basin was divided into main branch and five tributaries based on location of water quality sampling points and animal density, so that there were monitored tributaries with low animal density and with high animal density, as well as tributaries with high point source loading (Table 1). Effluents from point sources were calculated from observations of the river water sampling point before and after the effluent discharge point. The model was calibrated against Table 2 Land cover and main crops in sub basins. Sub basin

35.98 35.93 35.92 35.97 35.91 35.96 35.99 35.95 35.94

Land use Area

Forest

Set aside

Grass

Winter wheat

Spring barley

Sugar beet

Pasture

[km2]

[%]

[%]

[%]

[%]

[%]

[%]

[%]

211 405 470 193 422 465 224 228 425

78.0 60.0 40.0 57.0 57.8 52.0 39.9 58.0 58.0

2.0 2.6 3.8 3.5 1.4 3.1 3.3 2.2 3.0

6.5 6.4 10.3 7.5 2.5 7.0 4.5 4.9 6.3

0.2 3.6 5.4 2.6 3.3 3.2 8.1 2.2 2.0

11.3 26.6 39.3 28.0 28.4 33.8 43.8 32.0 30.2

0.9 0.1 0.1 0.2 6.3 0.4 0.2 0.2 0.2

1.1 0.7 1.1 1.2 0.3 0.5 0.2 0.5 0.3

5

observations in the lowest river water sampling point of the tributary (Table 1) as they collect all inputs from that area. In the automatic calibration procedure, parameter values were allowed to vary based on the range of values found in the literature. The amount of faecal coliforms in municipal and industrial effluents and in manure was allowed to vary ±20% as they do not have a consistent quality and also to indirectly include sewage of households outside municipal sewerage systems. The numbers of faecal coliforms can be as high as 6.8 × 105 CFU 100 ml−1 in ditch water nearby outlets of wastewater drains (Uusi-Kämppä, 2005). In a recent study, we analyzed the presence of general and host-specific contamination source identifiers in the neighboring small catchment (Pitkänen et al., 2014). The results indicated that the faecal bacteria in the surface water of small-scale area originated from human, pig and cattle faeces while the poultry markers remained absent. Wild animals were not included in the modelling though rough estimates of the typical games are available via the Natural Research Centre. Simulated counts were adjusted against observed counts of faecal coliforms along the river (water quality stations in the Table 1) as faecal coliforms are considered better estimates of faecal contaminants as total coliforms or faecal enterococci. The calibration period was 1995–2004 and validation period 2005–2009, due to high sampling frequency in these years. 2.4. Vulnerability to changes in climate and human behavior In this study our main interest was focused on the coming decades which are influenced by our current decisions. In Southern Finland annual temperature is assumed to increase by 1–2 °C and precipitation up by 7% by 2020–2050. We created three storylines based on the interaction between climate change and economical and societal development. The realization of the storylines was based on existing development trends and plans in the Loimijoki catchment. In Finland, the North Atlantic Oscillation (NAO) index is observed to correlate with winter discharge in Finnish rivers (Arvola et al., 2002). Thus, in the NAO positive years, winter discharges and air temperatures were higher than in the NAO negative years. The years when air temperature and precipitation corresponded with future climate were selected based on positive NAO index in autumn and spring months (years 1995, 1999, 2000, 2007, 2008 and 2009). In NAO positive years annual mean precipitation was 667 mm and annual mean temperature 5.6 °C at the Jokioinen Observatory weather station. In NAO negative years annual mean precipitation was 615 mm and mean temperature 4.7 °C. Further, in NAO positive years both monthly mean temperature and monthly mean precipitation was higher in autumn and winter than in NAO negative years (Supplementary material 2). The three storylines containing the described climate change scenarios and the economical and societal changes are the following: 2.4.1. Storyline1 — Techno world This storyline is based on high awareness but poor regulation of environmental protection. Most actions are the result of individual or commune interest on protecting the environment and they are based on technical solutions. Cultural services like recreation opportunities are locally important. In the Loimijoki River basin this storyline is based on the realization of the new central sewage treatment plant, which collects sewage water from all the small sewage treatment plants in the river basin. With activated sludge technology it is assumed to have capacity to remove 95% of the indicator bacteria (Leino, 2008). 2.4.2. Storyline2 — Consensus world In the storyline 2 the main objective of the government and citizens is to stimulate economic activity but also to promote sustainable and efficient use of resources. The current guidelines and policies are continued. As future climate is assumed to favour agricultural production in Finland due to expected longer growing period and milder winters

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

6

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

(Peltonen-Sainio et al., 2009). In the Loimijoki River basin, we assumed field percentage to increase up to 58% of the area, limited only by human settlements and soil types which are not suitable to agriculture. Animal husbandry is not assumed to intensify, as the manure use as fertilizer is still regulated by Nitrates directive and FAEP.

These qualitative storylines were included into the model application by making a quantitative change in the relevant parameter value.

2.5. Indicators of ecosystem services 2.4.3. Storyline3 — Fragmented world The focus of this storyline is to survive as a country instead of as part of Europe. National institutions focus on economic development and no attention is paid to the preservation of the ecosystems. In this storyline field area is assumed to increase up to 58% of the area as future climate favour agricultural production. Animal husbandry is assumed to intensify up to 5 animal units (AU) per ha. Agricultural or environmental policy does not regulate manure use as fertilizer and thus the third spreading to the fields in late autumn became a common practice to empty manure storages before winter (15th November).

Multiple pressures and their changes can result in the alteration of both the status and the services of aquatic ecosystems. To structure the analysis of ecosystem services and select appropriate indicators, we used the conceptual framework proposed by Grizzetti et al. (2016), based on the cascade model. The framework includes the capacity of the ecosystem to deliver the service, the actual flow of the service, and the benefits. Capacity refers to the potential of the ecosystem to provide ecosystem services, while flow is the actual use of the ecosystem services.

Fig. 2. Monthly simulated and observed faecal coliform counts on calibration period (a, c) and validation period (b, d) in the high pollution level (tributary 35.92) and low pollution level sampling points (tributary 35.99). Notice different scales between the figures.

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

For water purification we considered the rate of faecal coliforms removal (bacteria/km2 yr−1), which is an indicator of the actual flow of the service and is estimated by the model INCA-Pathogens. To assess the capacity of the ecosystem to provide clean water for drinking and recreational purposes we referred to the number of days faecal coliforms are below 1 bacteria 100 ml−1 and below 1000 CFU 100 ml−1 per year, respectively.

7

3. Results and discussion 3.1. Calibration and validation of the INCA-Pathogens model There are clear spatial gradient in the counts of faecal coliforms along the River Loimijoki. The highest pollution level is in the middle reach of the main stream immediately below small industries and

Fig. 3. Daily simulated and observed faecal coliform counts in the middle reaches (a) and at the outlet (b), and simulated and observed discharge at the outlet (c).

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

8

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

municipal sewage treatment plants (sampling point Loimijoki 92). The bacterial count decreases towards the outlet of the river basin, although the lowest pollution level is at the outlet of the lake in upper reaches. There is also clear seasonality so that the numbers of faecal coliforms tend to be highest during low flow season in summer (Fig. 2). In point source dominated parts of the river high faecal coliform numbers are observed also in low-flow periods in mid-winter reflecting the constant input which is mixed with small amount of river water. In agricultural tributaries high numbers are found in spring and autumn, typically when manure is spread on fields. Uusi-Kämppä and Heinonen-Tanski (2008) concluded that surface runoff water from fields always contains relatively high numbers of total coliforms when manure is used as fertilizer. They measured for faecal coliform counts between 1 and 10,000 CFU 100 ml−1 depending on the runoff, amount of manure applied and application method. The simulated faecal coliform counts in runoff waters from fields were at the same magnitude, though the highest simulated levels were up to 90,000 CFU 100 ml−1. That high numbers were measured from runoff from exercise yards for cattle with high AU ha−1 (Uusi-Kämppä et al., 2007). Seasonality is seen also in the simulated time-series of faecal coliform counts (Fig. 3) that follows in reverse order the seasonality of temperature and global radiation. The model simulations overestimate the lowest observed counts, especially at the outlet of the river. On the other hand, measurement of indicator bacteria is known to have large uncertainty (Harmel et al., 2016). The INCA-Pathogen model was able to produce faecal coliform counts and seasonality in the low pollution level sampling point (tributary 35.99) and in the high pollution level sampling point (tributary 35.92) (Fig. 3). In the high pollution level sampling point the simulated counts were lower than measured ones during summer, especially in July. One reason may be that surface spreading of manure during growing season has become more common in 2000's, and that was not taken into account in the model set-up. In the low pollution level sampling point, the measured counts were higher than simulated ones during snow melting period in April. This high number of faecal coliforms may originate from wild animals whose influence may be seen on areas with low anthropogenic loading, but which are not included into the model set-up. The FC/FS ratio was over one but below four at all of the measurement stations indicating that faecal contamination originated from both humans and animals. The highest ratios (N 3) were in the middle reaches and at the outlet of the main river, but also in one

site at the upper reaches where no point sources were reported. As this area is outside the communal sewage treatment, the reason may be leaking septic tanks of private houses. The lowest ratios (b2) were in the tributaries with high animal density but no point sources (Table 1). 3.2. Sensitivity analysis of the INCA-Pathogens model The INCA-Pathogen model performance was found to be particularly sensitive to the parameters for light controlled pathogen decay in river water and in soil compartment (Fig. 4). Also, simulations were sensitive to the number of faecal coliforms in manure spread on the fields. The model performance showed slightly lower sensitivity for parameters defining the decay rate in river water and in the soil compartment, and processes in sediment. Optimized decay rate (0.908 day−1) in the river was in the range of observed decay rates of E. coli (0.78–1.28 day−1) from outdoor microcosm trials performed in southern Sweden in August (Sokolova et al., 2012). In the neighboring river of the River Loimijoki, Wessels (2014) measured decay rates of E. coli (0.211–0.401 day−1) in dark and in temperature of + 4 °C. This was about the same rate that Sokolova et al. (2012) observed in November in dark conditions (0.22–0.56 day−1). Optimized light decay rate (0.049 day−1) was found to be between the light decay rates Caslake et al. (2004) measured for total coliforms (0.083 day−1), and Kadir and Nelson (2014) measured for E. coli (0.004–0.08 day− 1). Further, Sokolova et al. (2012) measured decay rate of 0.55 day−1 in light and 0.5 day−1 in dark in March which is a cool but light month. 3.3. Indicator bacteria counts in the river Despite the recent improvements in the waste water treatment, the River Loimijoki is still significantly impacted by faecal pollution. Indicator bacteria counts were highest from the middle reaches to the outlet of the main river. At the measurement station Lojo 54 average faecal coliform count exceeded 1000 CFU 100 ml−1 in every year. The daily value exceeded 1000 CFU 100 ml−1 in 40% of the days. Lowest faecal coliform counts were in the water of the tributary Niinijoki (35.99) despite of its high animal density. In all sub-catchments 1000 CFU 100 ml− 1 was exceeded in 5% of the days. In the tributaries 100 CFU 100 ml−1 was exceeded in 20% of days. We have no pathogen data from the River Loimijoki area as the microbiological water quality monitoring is solely based in the indicator approach. 3.4. Vulnerability of ecosystem functioning for climate change

Fig. 4. Sensitivity of parameters (20 most influential).

There were no substantial differences in hygienic water quality between NAO positive and NAO negative years. Only in the upper reaches and tributaries were the faecal coliform counts higher in NAO positive years compared to the NAO negative years. This is in accordance with observations of Uusi-Kämppä (2005) who found that bacteria counts in waters increase after heavy rainfall. Mean counts were slightly lower, as shown in Fig. 5. In the upper reaches, the water quality limit of 1000 CFU 100 ml−1 was exceeded in 10% of NAO positive days and 20% in NAO negative days (Fig. 3). In NAO positive year, the highest discharges were larger than in the NAO negative years which caused a dilution effect in the river (Fig. 5). Increased temperatures in future may accelerate the temperature dependent die-off of faecal microbes in the terrestrial and water environments. In this study the effect remained low. In NAO positive years the faecal bacteria die-off on barley fields receiving a high amount of manure was 319,900 CFU ha− 1 day− 1 and in NAO negative years 319,500 CFU ha−1 day−1. When comparing the discharge ranked average faecal coliform counts in July in NAO positive and NAO negative years, only very small differences occurred and mainly in low flow situations (Fig. 5). When

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

9

Fig. 6. Number of faecal coliforms in different discharge classes.

time. These results show that the current ecosystem services are vulnerable to climate change as the average faecal coliform counts may increase due to climate change. 3.5. Hygienic water quality according to different storylines In the storyline ‘Consensus world’ the peaks of faecal coliform counts were higher but otherwise there were no major difference compared to the current situation. In the storyline ‘Fragmented world’ hygienic quality of the river water was worst as the limit value of 1000 CFU 100 ml−1 was exceeded in 50% of days in the middle reaches and 20% of days in the tributary Punkalaitumenjoki (Fig. 9). In both storylines field area increased, but in the ‘Consensus world’ legislation limited the amount of manure to be spread on the fields. The storyline ‘Techno world’ yielded to the best hygienic quality of the river water. In the middle reaches 1000 CFU 100 ml−1 was exceeded in 10% of the days and in the tributary 9 in 5% of the days. In the middle reaches water quality was good in 30% of the days. In the both areas the existing municipal sewage water treatment plants were assumed to be replaced by one at the lowest reach. The hygienic quality improved also at the outlet of the Loimijoki River basin due to the improved purification efficiency of the new central waste water treatment plant (Fig. 7). In the storylines ‘Consensus world’ and ‘Techno world’ the highest faecal coliform counts occurred in May and in August when manure was spread on the fields (Fig. 8). In the storyline ‘Fragmented world’ the counts increased especially in November, but remained high throughout the winter. Precipitation was high both in August and November, but in November the photochemical decay is very low due to short days. In the storyline ‘Techno world’ water quality improves in every month, because sewage input was relatively constant. Current water protection legislation in agriculture (e.g. Nitrates Directive) is concentrated on reducing the harmful effects excess nutrients causes to water ecosystems, and the effect of legislation on pathogen loading has a secondary significance. This study shows that natural environment has a significant self-purification capacity which could better exploited also in environmental regulation and advising. Oliver et al. (2016) listed the integration of dynamic management practices and mitigation options as future challenges also in modelling context. 4. Conclusions Fig. 5. Count of faecal coliforms and discharge in NAO positive and NAO negative years.

changing date of manure spreading from August (average global radiation 14,400 W m− 2) to November (average global radiation 1070 W m− 2), the mean faecal coliform counts in river water in July the following year were about ten times higher (Fig. 6). This could be due to lower photochemical inactivation and temperature dependent decay when manure was spread in cold months with short daylight

The INCA-Pathogens model produced both the correct levels and seasonality of faecal coliform counts in high pollution level and low pollution level sites. The model behavior is not dominated by one process only, but there are a range of relevant processes that influence river water hygienic quality. Further, these processes are either relatively easy to measure or to find the corresponding parameter values from the literature.

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

10

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx

Climate change does not cause a major threat to hygienic water quality as presumably higher precipitation increases runoff and hence dilution in the river. Intensification of agriculture is not a threat either, as long as animal density remains relatively low, and manure use is regulated by legislation. More intensive agriculture without environmental legislation causes a threat especially in tributaries with high field percentage and animal density. The results of the model show that the hygienic water quality in the River Loimijoki can best be improved by improving sewage treatment as other stressors have a lower effect on the improvement of the water quality. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2016.05.043.

Acknowledgements This work was financed by the projects MARS (Managing Aquatic ecosystems and water Resources under multiple Stress; EU 7th Framework Programme, Grant Agreement 603378) and CONPAT (Aquatic contaminants – pathways, health risks and management; Academy of Finland). References

Fig. 7. Number of faecal coliforms according to different storylines.

Fig. 8. Average monthly counts of faecal coliforms according to different scenarios.

Arvola, L., Hakala, I., Järvinen, M., Huitu, E., Mäkelä, S., 2002. Effect of weather conditions on water quality in two small boreal rivers. Large Rivers 13, 195–208. Blaustein, R.A., Pachepsky, Y., Hill, R.L., Shelton, D.R., Whelan, G., 2013. Escherichia coli Survival in waters: temperature dependence. Water Res. 47, 569–578. Caslake, L.F., Connolly, D.J., Menon, V., Duncanson, C.M., Rojas, R., Tavakoli, J., 2004. Disinfection of Contaminated Water by Using Solar Irradiation. Appl. Environ. Microbiol. 70, 1145-115. Chapra, S.C., 1997. Chapter 27. Pathogens. Surface Water-Quality Modelling. McGraw Companies, USA, pp. 503–517. Decree of the Ministry of Social Affairs and Health Relating to the Quality and Monitoring of Water Produced by Small Water Supplies 401/2001, Finlex Data Bank; Edita Publishing Oy: Helsinki, Finland, 2001. Available online: http://www.finlex.fi/fi/laki/ alkup/2001/20010401 (accessed on 20 August 2015). (In Finnish). Decree of the Ministry of Social Affairs and Health Relating to the Quality Requirements and Monitoring of the Household Water 461/2000, Finlex Data Bank, Edita Publishing Oy: Helsinki, Finland, 2000. Available online: http://www.finlex.fi/fi/laki/alkup/2000/ 20000461 (accessed on 20 August 2015). (In Finnish). Dorner, S.M., Anderson, W.B., Slawson, R.M., Kouwen, N., Huck, P.M., 2006. Hydrologic modeling of pathogen fate and transport. Environ. Sci. Technol. 40, 4746–4753. Drummond, J.D., Davies-Colley, R.J., Stott, R., Sukias, J.P., Nagels, J.W., Sharp, A., Packman, A.I., 2015. Microbial transport, retention, and inactivation in streams: a combined experimental and stochastic modeling approach. Environ. Sci. Technol. 49, 7825–7833. EEC, 1991. Council Directive 91/676/EEC of 12 December 1991 concerning the protection of waters against pollution caused by nitrates from agricultural sources. Official Journal L 375, 1–8. FAO, 2014. World reference base for soil resources 2014. International soil classification system for naming soils and creating legends for soil maps. WORLD SOIL RESOURCES REPORTS, p. 193. Ferguson, C.M., Croke, B.F.W., Beatson, P.J., Ashbolt, N.J., Deere, D.A., 2007. Development of a process-based model to predict pathogen budgets for the Sydney drinking water catchment. J. Water Health 05 (2), 187–208. Finlex, 2014. Valtioneuvoston asetus eräiden maa- ja puutarhataloudesta peräisin olevien päästöjen rajoittamisesta (1250/2014). Frey, S.K., Topp, E., Edge, T., Fall, C., Gannon, V., Jokinen, C., Marti, R., Neumann, N., Ruecker, N., Wilkes, G., Lapen, D.R., 2013. Using SWAT, bacteroidales microbial source tracking markers, and fecal indicator bacteria to predict waterborne pathogen occurrence in an agricultural watershed. Water Res. 47, 6326–6337. Futter, M.N., Butterfield, D., Cosby, B.J., Dillon, P.J., Wade, A., Whitehead, P.G., 2007. Modeling the mechanisms that control in-stream dissolved organic carbon dynamics in upland and forested catchments. Water Resour. Res. 43 http://dx.doi.org/10.1029/ 2006WR004960. Grizzetti, B., Lanzanova, D., Liquete, C., Reynaud, A., Cardoso, A.C., 2016. Assessing water ecosystem services for water resource management. Environ. Sci. Pol. 61, 194–203. Haines-Young, R., Potschin, M. (2013) Common International Classification of Ecosystem Services (CICES): Consultation on Version 4, August–December 2012. EEA Framework Contract No EEA/IEA/09/003, Download at www.cices.eu. Harmel, R.D., Hathaway, J.M., Wagner, K.L., Wolfe, J.E., Karthikeyan, R., Francesconi, W., McCarthy, D.T., 2016. Uncertainty in monitoring E. coliconcentrations in streams and stormwater runoff. J. Hydrodyn. 534, 524–533. Hokajärvi, A.M., Pitkänen, T., Siljanen, H.M., Nakari, U.M., Torvinen, E., Siitonen, A., Miettinen, I.T., 2013. Occurrence of thermotolerant Campylobacter spp. and adenoviruses in Finnish bathing waters and purified sewage effluents. J. Water Health 11, 120–134. Jylhä, K., Ruosteenoja, K., Räisänen, J., Venäläinen, A., Tuomenvirta, H., Ruokolainen, L., Saku, S., Seitola, T., 2009. In: Laitos, I. (Ed.), Arvioita Suomen muuttuvasta ilmastosta

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043

K. Rankinen et al. / Science of the Total Environment xxx (2016) xxx–xxx sopeutumistutkimuksia varten. ACCLIM-hankkeen raportti 2009 4. RAPORTTEJA RAPPORTER REPORTS 2009, p. 114. Kadir, K., Nelson, K.L., 2014. Sunlight mediated inactivation mechanisms of Enterococcus faecalis and Escerichia coli in clear water versus waste stabilisation pond water. Water Res. 50, 307–317. Kauppinen, A., Martikainen, K., Matikka, V., Veijalainen, A.-M., Pitkänen, T., HeinonenTanski, H., Miettinen, I.T., 2014. Sand filters for removal of microbes and nutrients from wastewater during a one-year pilot study. J. Environ. Manag. 133, 206–213. Kay, D., Wyer, M., Crowther, J., Stapleton, C., Bradford, M., McDonald, A., Greaves, J., Francis, C., Watkins, J., 2005. Predicting faecal indicator fluxes using digital land use data in the UK's sentinel water framework directive catchment: the Ribble study. Water Res. 39, 3967–3981. Kay, D., Anthony, S., Crowther, J., Chambers, B.J., Nicholson, F.A., Chadwick, D., Stapleton, C.M., Wyer, M.D., 2010. Microbial water pollution: a screening tool for initial catchment-scale assessment and source apportionment. Sci. Total Environ. 408, 5649–5656. Kirschner, A.K.T., Kavka, G.G., Velimirov, B., Mach, R.L., Sommer, R., Farnleitner, A.H., 2009. Microbiological water quality along the Danube River: integrating data from two whole-river surveys and a transnational monitoring network. Water Res. 43, 3673–3684. Lazar, A.N., Butterfield, D., Futter, M.N., Rankinen, K., Thouvenot-Korppoo, M., Jarritt, N., Lawrence, D.S.L., Wade, A.J., Whitehead, P.G., 2010. An assessment of the fine sediment dynamics in an upland river system: INCA-Sed modifications and implications for fisheries. Sci. Total Environ. 408, 2555–2566. Leino, N., 2008. Puhdistetun jäteveden patogeenit ja desinfiointitarve (pathogens in treated effluent and the need for disinfection). Lappeenranta University of Technology. Levenberg, K., 1944. A method for the solution of certain non-linear problems in least squares. Q. Appl. Math. 2, 164–168. MA Millenium Ecosystem Assessment, 2005. Ecosystems and Human Well-being (Washington DC). Marquardt, D.W., 1963. Am algorithm for least-squares estimation of nonlinear parameters. J. Soc. Ind. Appl. Math. 11, 431–441. Meays, C.L., Broersma, K., Nordin, R., Mazumder, A., 2004. Source tracking fecal bacteria in water: a critical review of current methods. J. Environ. Manag. 73 (1), 71–79. Myers DN, Stoeckel DM, Bushon RN, Francy DS, Brady AMG (2014) Fecal Indicator Bacteria (Ver. 2.1). U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A7, section 7.1, May 2014', [online], available: http://water.usgs.gov/ owq/FieldManual/Chapter7/7.1.html [accessed 14/07/2014]. Niemi, R.M., Niemi, J.S., 1991. Bacterial Pollution of Waters in Pristine and Agricultural Lands. Journal of Environmental Quality 20, 620–627. Niskanen, O., Lehtonen, E., 2014. Maatilojen tilusrakenne ja pellonraivaus Suomessa 2000-luvulla. MTT Raportti 150, 28. Oliver, D.M., Porter, K.D.H., Pachepsky, Y.A., Muirhead, R.W., Reaney, S.M., Coffey, R., Kay, D., Milledge, D.G., Hong, E., Anthony, S.G., Page, T., Bloodworth, J.W., Mellander, P.-E., Carbonneau, P.E., McGrane, S.J., Quilliam, R.S., 2016. Predicting microbial water quality with models: over-arching questions for managing risk in agricultural catchments. Sci. Total Environ. 544, 39–47. Peltonen-Sainio, P., Jauhiainen, L., Hakala, K., Ojanen, H., 2009. Climate change and prolongation of growing season: changes in regional potential for field crop production in Finland. Agricultural and Food Science 18, 171–190. Peltonen-Sainio, P., Jauhiainen, L., Trnka, M., Olesen, J.E., Calanca, P., Eckersten, H., Eitzinger, J., Gobin, A., Kersebaum, K.C., Kozyra, J., Kumar, S., Dalla Marta, A., Micale, F., Schaap, B., Seguin, B., Skjelvag, A.O., Orlandini, S., 2010. Coincidence of variation in yield and climate in Europe. Agric. Ecosyst. Environ. 139, 483–489. Pitkänen, T., 2013. Review of Campylobacter spp. in drinking and environmental waters. J. Microbiol. Methods 95, 39–47. Pitkänen, T., Ryu, H., Elk, M., Hokajärvi, A.-M., Siponen, S., Vepsäläinen, A., Räsänen, P., Santo Domingo, J.W., 2013. Detection of fecal bacteria and source tracking identifiers in environmental waters using rRNA-based RT-qPCR and rDNA-based qPCR assays. Environ. Sci. Technol. 13611–13620. Pitkänen, T., Kauppinen, A., Räsänen, P.S., Kusnetsov, J., Ryu, H., Hokajärvi, A.-M., Siponen, S., Santo Domingo, J.W., Miettinen, I.T., 2014. Waterborne Pathogens and Their Sources at Kokemäki River Water Course. IWA World Water Congress & Exhibition, 21–26 September 2014, Lisbon, Portugal.

11

Pitkänen, T., Juselius, T., Isomäki, E., Miettinen, I.T., Valve, M., Kivimäki, A.-L., Lahti, K., Hänninen, M.-L., 2015. Drinking water quality and occurrence of giardia in Finnish small groundwater supplies. Resources 4, 637–654. Rankinen, K., Valpasvuo-Jaatinen, P., Karhunen, A., Kenttämies, K., Nenonen, S., Bärlund, I., 2009. Simulated nitrogen leaching patterns and adaptation to climate change in two Finnish river basins with contrasting land use and climatic conditions. Hydrol. Res. 40 (2–3), 177–186. Ruosteenoja, K., Jylhä, K., 2007. Temperature and Precipitation Projections for Finland Based on Climate Models Employed in the IPCC 4th Assessment Report. Third International Conference on Climate and Water, Helsinki, Finland, 3–6 September 2007. Proceedings. pp. 404–406. Schoonover, J.E., Lockaby, B.G., 2006. Land cover impacts on stream nutrients and fecal coliform in the lower Piedmont of West Georgia. J. Hydrodyn. 331, 371–382. Servais, P., Billen, G., Goncalves, A., Garcia-Armisen, T., 2007. Modelling microbiological water quality in the Seine river drainage network: past, present and future situations. Hydrol. Earth Syst. Sci. 11, 1581–1592. Sokolova, E., Åström, J., Pettersson, T.J.R., Bergstedt, O., Hermansson, M., 2012. Decay of bacteroidales genetic markers in relation to traditional fecal indicators for water quality modeling of drinking water sources. Environ. Sci. Technol. 46, 892–900. STM, 2008. Sosiaali- ja terveysministeriön asetus yleisten uimarantojen uimaveden laatuvaatimuksista ja valvonnasta 177/2008. Tallon, P., Magajna, B., Lofranco, C., Leung, K.T., 2005. Microbial indicators of faecal contamination in water: a current perspective. Water Air Soil Pollut. 166, 139–166. TEEB, 2010. Chapter 3. Measuring biophysical quantities and the use of indicators. In: Kumar, P. (Ed.), The Economics of Ecosystems and Biodiversity. Ecological and Economic Foundation. Tiwari, A., Niemelä, S.I., Rapala, J., Kalso, S., Pitkänen, T., 2015. Comparison of methods suitable for monitoring of Escherichia coli in bathing water in Finland. Journal of Water and Health in press. Uusi-Kämppä J. In: (eds) (ed) (2005) Faecal Bacteria and Coliphages in Run-Off From Dairy Farms. Uusi-Kämppä, J., Heinonen-Tanski, H., 2008. Evaluating slurry broadcasting and injection to ley for phosphorus losses and fecal microorganisms in surface runoff. J. Environ. Qual. 37, 2339–2350. Uusi-Kämppä, J., Hyvärinen, M., Kuisma, M., Nykänen, A., Jansson, H., Närvänen, A., June 17-21 2007. Faecal microorganisms in runoff from cattle and horse farms - quantification and mitigation. In: A, A (Ed.), Animal Health, Animal Welfare and Biosecurity. XIII International Congress in Animal Hygiene, Tartu, Estonia, pp. 909–914 Proceedings. Veijalainen, N., Lotsari, E., Alho, P., Vehviläinen, B., Käyhkö, J., 2010. National scale assessment of climate change impacts on flooding in Finland. J. Hydrol. 391, 333–350. Wade, A., Durand, P., Beaujoan, V., et al., 2002. Towards a generic nitrogen model of European ecosystems: new model structure and equations. Hydrol. Earth Syst. Sci. 6 (3), 559–582. Wessels, L., 2014. Faecal bacteria and contamination source identifiers in Kokemäki river watershed. National Institute for Health and Welfare, Water and Health Unit. Whitehead, P.G., Wilson, E.J., Butterfield, D., 1998. A semi-distributed integrated nitrogen model for multiple source assessment in catchments (INCA): part I-model structure and process equations. Sci. Total Environ. 210/211, 547–558. Whitehead, P.G., Leckie, H., Rankinen, K., Butterfield, D., Futter, M.N., Bussi, G., 2015. An INCA model for pathogens in rivers and catchments: Model structure, sensitivity analysis and application to the river Thames catchment in the UK. Sci. Total Environ. (In press). Wilkinson, J., Jenkins, A., Wyer, M., Kay, D., 1995. Modelling faecal coliform dynamics in streams and rivers. Water Res. 29, 847–855. Yakirevich, A., Pachepsky, Y.A., Guber, A.K., Gish, T.J., Shelton, D.R., Cho, K.H., 2013. Modeling transport of Escherichia coli in a creek during and after artificial high-flow events: three-year study and analysis. Water Res. 47, 2676–2688. Zacheus, O., Miettinen, I.T., 2011. Increased information on waterborne outbreaks through efficient notification system enforces actions towards safe drinking water. J. Water Health 9, 763–772.

Please cite this article as: Rankinen, K., et al., The INCA-Pathogens model: An application to the Loimijoki River basin in Finland, Sci Total Environ (2016), http://dx.doi.org/10.1016/j.scitotenv.2016.05.043