Data collection NAPSEA

DELIVERABLE 3.1 DATA COLLECTION Work Package 3 Measures & Pathways 31-05-2023 www.napsea.eu Page 2 of 17 Deliverable 3.1 Grant Agreement number 101060418 Project title NAPSEA: the effectiveness of Nitrogen and Phosphorus load reduction measures from Source to sEA, considering the effects of climate change Project DOI 10.3030/101060418 Deliverable title Data collection Deliverable number D3.1 Deliverable version 1 Contractual date of delivery 31.05.2023 Actual date of delivery 31.05.2023 Document status Prepared Document version 1 Online access Yes Diffusion Public (PU) Nature of deliverable Report Work Package WP3. Measures & Pathways Partner responsible UFZ Contributing Partners Deltares Author(s) Seifeddine Jomaa and Andreas Musolff Editor van der Heijden, L.H. Approved by van der Heijden, L.H. Project Officer Blanca Saez-Lacava/Christel Millet Abstract Data required for modelling including climate, land use, soil type, discharge, observed nutrient concentrations, agricultural practices, and point source. Keywords Elbe, Rhine, mQM, hydrological model, water quality, Nitrogen, Phosphorus, diffuse source, point source, mitigation measure. Page 3 of 17 Deliverable 3.1 Contents 1. ACRONYMES ..................................................................................................................................... 4 2. EXECUTIVE SUMMARY .................................................................................................................... 5 3. INTRODUCTION ................................................................................................................................. 6 3.1 Work package description ....................................................................................................... 6 3.2 Spatio-temporal data requirements for the envisioned modelling approach .......................... 6 3.3 Data availability and access .................................................................................................... 8 3.4 Data availability per case study ............................................................................................... 8 3.4.1. Rhine river basin ............................................................................................................. 8 3.4.2. Elbe river basin ................................................................................................................ 8 3.4.3. Hunze river basin ............................................................................................................ 9 4. REFERENCES .................................................................................................................................. 17 Page 4 of 17 Deliverable 3.1 1. ACRONYMES AMSL Above mean sea level CnANDY Coupled Complex Algal-Nutrient Dynamics FAIR Findability, Accessibility, Interoperability, and Reusability FGG Flussgebietsgemeinschaft/ Riverine Commission GRQA Global River Water Quality Archive MOSES Modular Observation Solutions for Earth Systems mQM multiscale water Quality Model mHM mesoscale Hydrological Model PE Population equivalent TERENO Terrestrial Environmental Observatories WP Work package WWTP wastewater treatment plant Page 5 of 17 Deliverable 3.1 2. EXECUTIVE SUMMARY Deliverable 3.1 summarizes the data availability, data sources and their spatiotemporal extent for the NAPSEA project basins draining into the Wadden Sea with a specific focus on the case sites Elbe, Rhine and Hunze. The data will be used to setup the water quantity and quality models in task 3.2 (mHM, mQM and CnANDY) to backcast concentrations and loads and capture spatial patterns of nutrients (Nitrogen and Phosphorus). These models will serve for testing the efficiency of different stakeholders-selected mitigations measures to ensure the safe ecological boundaries of both case studies. Data include climatic forcing, land use properties, soil type, point-sources (sewage stations and their discharge), agricultural practices (manure, mineral fertilizer) and measured stream discharge and nutrients (Nitrogen and Phosphorus) concentrations. Observed data covering the longest time series were prioritized in the analysis to ensure a good historical coverage of nutrient-pollution background conditions. Table 1 and table 2 summarize the spatiotemporal coverage of data availability and sources. The deliverable was planned as a “data collection” in the project proposal. This document lists and describe the data, their data source and enables access to the data. The minimum data to run the model is available in FAIR repositories and open access publications such that everyone within the project consortium but also outside can access the data. Page 6 of 17 Deliverable 3.1 3. INTRODUCTION 3.1 Work package description Measures and Pathways WP3 aims to evaluate the connection between nutrient concentration and load reduction measures, considering changed terrestrial inputs as well as enhanced retention processes (such as instream retention) and the safe ecological boundaries in the receiving waters of three case studies (Figure 1). An integral approach of testing stakeholders-approved pathways of nutrients reduction from the sources via streams and rivers to estuaries and coastal waters of the Wadden Sea will be adopted. To this end, scenarios for testing the efficiency of different nutrients reduction mitigations and enhanced retention measures will be conducted using process-based mHM (Samaniego et al., 2010, Kumar et al., 2013) 1 , 2 , mQM (e.g., Nguyen et al, 2022) 3 and CnANDY (Yang et al. 2021) 4 models. This allows to prioritise nutrient reduction mitigation measures under shortand long-term perspectives and under different scenarios of climate change. Also, measures considering co-benefits aspects beyond the waterborne and airborne will be further explored. To this end, the integrated and time-variant modelling approach to nutrient transport and retention across nested scales from source to sea will be implemented (task 3.2). Modelling will be used for quantifying the effectiveness of proposed solutions under current and future climate change scenarios (task 3.3, 3.4 and 3.5). The modelling results will also be used as knowledge hubs for motivating stakeholders’ solutions uptake of proposed actions. The model will be initially set up for the three case studies to capture the current observed state of nutrient concentrations in surface waters and exports to the Wadden Sea and to implement the measures and climate scenarios. Results from the case studies will be transferred to other basins draining into the Wadden Sea (task 3.6, e.g., Weser/ Ems basin, catchments north of the Elbe estuary) using data-driven methods, such as machine learning methods. To broaden the benefits of the modelling approach, other co-benefits aspects have also been considered, such as the capability of the developed modelling approach to offer further insights into the impacts of recently experienced climate-change-related problems (droughts after 2018 and connected long travel times of water in the main streams) on algae blooms and ecological state of the estuary. To this end, model usages’ capabilities and its internal processes will be explored as additional sources of information to enhance our physical understanding. This requires further consideration of spatial and temporal resolution of model results and parameters to capture the instream processes at sufficient resolution (land-stream transfer). 3.2 Spatio-temporal data requirements for the envisioned modelling approach A modelling approach is one of the most cost-effective tools for testing mitigation measures scenarios compared to any other approach before a real implementation. However, facilitating rigorous testing of mitigation measures requires adequate process detail in the modelling and adequate spatiotemporal resolution. Additionally, the model needs to acknowledge the spatial dimension of the modelled basins and the availability of data to run and calibrate the model. Consequently, not all small-scale measures 1 Samaniego L., R. Kumar, S. Attinger (2010): Multiscale parameter regionalization of a grid-based hydrologic model at the mesoscale. Water Resour. Res., 46, W05523, doi:10.1029/2008WR007327. 2 Kumar, R., L. Samaniego, and S. Attinger (2013): Implications of distributed hydrologic model parameterization on water fluxes at multiple scales and locations, Water Resour. Res., 49, doi:10.1029/2012WR012195. 3 Nguyen TV, Sarrazin FJ, Ebeling P, Musolff A, Fleckenstein JH, Kumar R. Toward Understanding of Long-Term Nitrogen Transport and Retention Dynamics Across German Catchments. Geophysical Research Letters 2022; 49: e2022GL100278. 4 Yang S, Bertuzzo E, Büttner O, Borchardt D, Rao PSC. Emergent spatial patterns of competing benthic and pelagic algae in a river network: A parsimonious basin-scale modelling analysis. Water Research 2021; 193: 116887. Page 7 of 17 Deliverable 3.1 can be captured by a model running at the spatial scale of the entire multiscale case studies river basins (Elbe, Rhine and Hunze (Figure 2)). We argue that the mechanistic modelling approach applied in NAPSEA has advantages against established empirical and semi-empirical models in capturing the large-scale combined effect of measures to reduce nutrient inputs in the landscape and enhanced retention also under the changed boundary conditions of future climate developments. Within NAPSEA we will combine the hydrological model mHM1,2 (modelling discharge, evapotranspiration, soil moisture and river network water routing on a daily basis and in high spatial resolution) with the water quality model mQM3 (modelling travel time-based nitrogen transport and retention in soil, groundwater and the stream network at annual time step at the spatial scale of subcatchments) and CnANDY4 (modelling dissolved phosphorous as well as benthic and pelagic-bound phosphorous transport in the river network at various time step) model. Fundamentally different to other water quality modelling approaches is the travel time concept in mQM model - a unique mechanistic approach that can consider different time scales up to decades needed to capture different flow paths and connect retention processes to transport and turnover nutrients in soil and groundwater bodies before reaching streams. This will also allow us to quantify the time that a measure will take to impact stream water nutrient concentration and fluxes. Finally, the travel time concept will account for feedback on changed hydroclimatic drivers (e.g., prolonged drought periods) on the transport and retention of nutrients in the subsurface and the stream network. Also, the spatially distributed model implementation in sub-catchments is a unique feature for testing a spatially targeted mitigation measure locally and quantifying their effect on the overall nutrient export at the basin scale. In a scenario approach, we will be able to track back nutrients exported to the Wadden Sea to its source and test effectiveness of spatially targeted measures considering retention in the river network. To achieve this level of enhanced understanding, intensive model calibration and validation is required. To this end, model setup and validation for nutrient retention and transport from source to sea become a data-demanding process. More specifically, the models need a large amount of data on the (1) meteorological forcing, (2) on nutrient inputs from point and non-point sources, on (3) landscape properties and (4) on water quantity and quality observations. Meteorological forcing comprises precipitation, potential evapotranspiration and air temperature data that drives the hydrological model mHM and allows estimations of the soil temperature needed for the soil nitrogen reactions in the water quality model mQM. Nutrient inputs are used for nitrogen fluxes in mQM and for phosphorous fluxes in CnANDY. Landscape properties comprise soil databases, land cover and topography shaping hydrological transport and retention of nutrients. Observational water quantity data are used to calibrate mHM while the water quality observations are used to calibrate and validate mQM and CnANDY. Since mQM will be calibrated against these observations for each sub-catchment, the availability of water quality observations dictates the spatial resolution of the model. At the moment, we envision a spatial resolution of approximately 100 km2 matching 2nd to 3rd Strahler order streams (see chapter 3.4). This also matches the spatial resolution of the diffuse N inputs that will be used in the model (Batool et al. 2022). Observational data on water quality and quantity for the German part have been collected before and published following the FAIR data principle (QUADICA database) 5 . These data have been updated and extended considering the case studies requirement and project needs. In addition, spatiotemporally monitored water quantity and quality data during the last ten years with the monitoring activities of 5 Ebeling P, Kumar R, Lutz SR, Nguyen T, Sarrazin F, Weber M, et al. QUADICA: water QUAlity, DIscharge and Catchment Attributes for large-sample studies in Germany. Earth Syst. Sci. Data 2022; 14: 3715-3741. Page 8 of 17 Deliverable 3.1 TERENO 6 and MOSES 7 activities are available for internal processes understanding and validating the modelling results. Collected data at the first phase of the project, their spatial and temporal resolution and their repositories are summarized in Table 1. Furthermore, more detailed data will be gathered continuously over the course of the project implementation within each of the local case studies specifically focusing on the local issues at hand and depending on the further requests and needs. All required data to setup the models in all basins are available and freely accessible in different repositories (Table 1). These data are originally reviewed and published separately in international and highly ranked journals using the FAIR principal. These data are considered as the minimum data needed for our models’ setup. However, modelling results can be further analysed and improved when additional in-situ observations about the model’s internal processes are available. Additional data that can help constraining the models are soil organic and mineral N data, groundwater quality data (see Table 1) and water age measurements. Table 2 lists additional local data that is available to calibrate and validate the Hunze test basin. 3.3 Data availability and access All data used to run and calibrate the models mHM, mQM and CnANDY are available in FAIR open data repositories or part of open access publications that can be assessed by everyone. We refrain from duplicating this data in another repository to avoid redundancy and to account for the fact that some of the data is updated (e.g., meteorological drivers) on the data website. Additional data for local calibration and validation of the Hunze basin was provided from local authorities and is available on request for everyone in the consortium. 3.4 Data availability per case study 3.4.1. Rhine river basin The Rhine river basin drains an area of 220,000 km2 and is one of the most heavily used waterways in the world. Precipitation of the Rhine River ranges from less than 200 mm y-1 in the central part to 3500 mm y-1 in the mountains. The Rhine has an average discharge of about 2,900 m3 s-1. In the past, the Rhine experienced strong anthropogenic impacts with strong modification of the hydromorphology and heavy industrial pollution. While industrial pollution has been greatly reduced, the Rhine is still a major contributor of nutrients from diffuse agricultural sources and from wastewater discharges to the Wadden Sea. In recent years, the Rhine experienced a series of droughts (2018, 2022) with severe impacts on the instream ecosystem such as algal blooms in tributaries. 3.4.2. Elbe river basin The Elbe River basin, located in central Europe covers an area of 148.268 km2, where approximately one third is the Czech Republic. Less than 1% belong to Austria and Poland. About 50% of Elbe River basin are lowlands below 200 m AMSL, dominating the north landscape of the basin. Overall, the river basin is characterised by sandy plateaus with loam-covered riparian zones and wetlands in between. The average annual precipitation of the Elbe River basin is about 628 mm. However, the annual precipitation reached higher levels in the Giant and Jizera Mountains where precipitation can reach 1700 mm per year 8 . Precipitation shows a rather uniform intra-annual distribution due to the low slopes, sandy soils, and relatively low rainfall intensity, the hydrological behaviour is governed by groundwater dynamics. Major land uses are grassland, forestry, and agriculture, often on poor soils. The long-term 6 https://www.tereno.net/ 7 https://www.ufz.de/moses/ 8 https://www.ikse-mkol.org/fileadmin/media/user_upload/E/06_Publikationen/08_IKSE_Flyer/2016_IC PER-Flyer_The_Elbe_River_Basin.pdf Page 9 of 17 Deliverable 3.1 annual mean discharge at the river mouth is about 861 m3 s-1, which is equivalent to an average evapotranspiration of 519 mm y-1 (FGG Elbe, 2005). With the onset of industrial revolution, the Elbe has developed from an increased chemical pollution primarily from wastewater inputs originating from urban and industrial and mining activities (Figure 3). Later, point sources pollution has been controlled through intensive building of sewer systems across the country, leading to rapid improvement of water quality and autotrophic river systems. 3.4.3. Hunze river basin The Hunze river basin is an intensively farmed catchment of ~350 km2 draining into a freshwater lake (Zuidlaardermeer) with important recreational functions. Stakeholders around the lake, like holiday park and marina owners, are directly affected by harmful algal blooms. At the same time, the upstream lowreactive, sandy agricultural soils are susceptible to nutrient losses. The close link between the nutrient sources and the nearby eutrophication issues in lake Zuidlaardermeer make this case interesting. Further downstream, the Hunze river basin influences the channels of the city of Groningen before draining into the Wadden Sea. Figure 1. Case studies of NAPSEA project from sources to Wadden Sea: Hunze, Rhine and Elbe river basins, and Wadden Sea. Figure 2. Land cover maps for the Elbe and Rhine River basins. Page 16 of 17 Deliverable 3.1 Model evaluation data Measured nitrogen and phosphorus concentrations, surface water Nitrogen and Phosphorous species concentration Weekly to monthly observations/ Germany, Europe and global 1968-2020 Elbe and Rhine Entire Germany: Ebeling et al. (2022)5 ASCII table format, data from: https://doi.org/10.4211/hs.88254bd930d1466c85992a7dea6947a4 Europe and global: Virro et al. (2021)10 ASCII table format, data from: https://doi.org/10.5281/zenodo.5097436 Measured nitrogen concentration, groundwater Nitrogen species concentration bi-annual to annual observations/ entire Europe 1990-2017 EEA database containing observational raw data (Part 1: DisaggregatedData, ASCII table format) https://www.eea.europa.eu/data-and-maps/data/waterbase-waterquality-icm-2 Table 2: Additional data sources allowing for mQM and CnANDY model setup in the Hunze test catchment. Data type Variable Resolution/ Extent Period Source Measured nutrient concentrations, surface water TN, TN and other water quality parameters (such as Chloride, dissolved oxygen, pH, NO3, PO4 etc) Weekly to monthly observations/ nine stations 2000-2023 Biweekly measured concentrations data at nine gauging stations from upstream to downstream area of the Hunze catchment was collected from the responsible authorities. Hydrological data Measured discharge Daily observations at six gauging stations/ Hunze 2010-2022 Daily measured discharge data was collected from the responsible authorities. Page 17 of 17 Deliverable 3.1 4. REFERENCES 1. Samaniego L., R. Kumar, S. Attinger (2010): Multiscale parameter regionalization of a grid-based hydrologic model at the mesoscale. Water Resour. Res., 46,W05523, doi:10.1029/2008WR007327. 2. Kumar, R., L. Samaniego, and S. Attinger (2013): Implications of distributed hydrologic model parameterization on water fluxes at multiple scales and locations, Water Resour. Res., 49, doi:10.1029/2012WR012195. 3. Nguyen TV, Sarrazin FJ, Ebeling P, Musolff A, Fleckenstein JH, Kumar R. Toward Understanding of Long-Term Nitrogen Transport and Retention Dynamics Across German Catchments. Geophysical Research Letters 2022; 49: e2022GL100278. 4. Yang S, Bertuzzo E, Büttner O, Borchardt D, Rao PSC. Emergent spatial patterns of competing benthic and pelagic algae in a river network: A parsimonious basin-scale modelling analysis. Water Research 2021; 193: 116887. 5. Ebeling P, Kumar R, Lutz SR, Nguyen T, Sarrazin F, Weber M, et al. QUADICA: water QUAlity, DIscharge and Catchment Attributes for large-sample studies in Germany. Earth Syst. Sci. Data 2022; 14: 3715-3741. 6. https://www.tereno.net/ 7. https://www.ufz.de/moses/ 8. https://www.ikse-mkol.org/fileadmin/media/user_upload/E/06_Publikationen/08_IKSE_Flyer/2016 _ICPER-Flyer_The_Elbe_River_Basin.pdf 9. Büttner O, Jawitz JW, Birk S, Borchardt D. Why wastewater treatment fails to protect stream ecosystems in Europe. Water Research 2022; 217: 118382. 10. Virro H, Amatulli G, Kmoch A, Shen L, Uuemaa E. GRQA: Global River Water Quality Archive. Earth Syst. Sci. Data 2021; 13: 5483-5507. 11. Schrier, E.J.M. van den Besselaar, and P.D. Jones. 2018: An Ensemble Version of the E-OBS Temperature and Precipitation Datasets, J. Geophys. Res. Atmos., 123. 12. Samaniego L., R. Kumar, S. Attinger (2010): Multiscale parameter regionalization of a grid-based hydrologic model at the mesoscale. Water Resour. Res., 46, W05523. 13. Kumar, R., L. Samaniego, and S. Attinger (2013): Implications of distributed hydrologic model parameterization on water fluxes at multiple scales and locations, Water Resour. Res., 49. 14. Batool, M., Sarrazin, F.J., Attinger, S. et al. Long-term annual soil nitrogen surplus across Europe (1850–2019). Sci Data 9, 612 (2022). https://doi.org/10.1038/s41597-022-01693-9 15. EEA (2022). https://www.eea.europa.eu/data-and-maps/data/waterbase-uwwtd-urban-wastewater-treatment-directive-9 16. UBA (2010). Calculation of emissions into rivers in Germany using the MONERIS ModelNutrients, heavy metals and polycyclic aromatic hydrocarbons. The German Federal Environmental Agency (UBA). Retrieved from https://www.umweltbundesamt.de/en/publikationen/calculation-of-emissions-into-rivers-ingermany.