1 Dataset description In ocean model or Earth System model applications, the riverine freshwater inflow is an important flux affecting salinity and marine stratification in coastal areas. However, in climate change studies, the river runoff based on climate model output often has large biases on local, regional or even basin wide scales. If these biases are too large, the ocean model forced by the runoff will drift into a different climate state compared to the observed state, which is especially relevant for semi-enclosed seas like the Baltic Sea. In order to fulfil the demands for low biases in river runoff, a three-part bias correction was developed by Hagemann et al. (in prep.) that comprises different correction factors for low, medium and high percentile ranges of river runoff over Europe. First, we utilized the global hydrology model HydroPy (Stacke and Hagemann 2021) and the Hydrological Discharge (HD) model (Hagemann et al. 2020) to simulate daily discharge time series over the European domain at 1/12° horizontal resolution Sect. 1.1) from 1901-2019. Then, we bias-corrected these time series as described in Sect. 1.2 to generate bias-corrected discharges at coastal ocean boxes of the European HD model domain from 1901-2019. 1.1 Century-long high-resolution discharge simulation over Europe Analogous to Hagemann and Stacke (2022), the global hydrology model HydroPy (Vs. 1.0.2 Stacke and Hagemann 2021) and the Hydrological Discharge (HD) model (Vs. 5.2.0, Hagemann et al. 2023) were used to simulate daily discharge time series over the European domain at 1/12° horizontal resolution. Daily data of two atmospheric datasets were utilized to force HydroPy that provided the input to the HD model. The Global Soil Wetness Project Phase 3 (GWSP3; Dirmeyer et al. 2006; Kim 2017) dataset is available at 0.5° resolution from 1901-2014. Here, we used the data from 1901-1978, and then the simulated time series were continued by using the WFDE5 dataset (Cucchi et al. 2020; 0.5° resolution) from 1979-2019. 1.2 Generation of bias corrected HD discharge data In order to apply the bias correction of Hagemann et al. (in prep.) to the simulated time series of daily discharge from 1901-2019, two sets of bias correction factors were derived. The first set uses the WFDE5-based discharges and discharge station observations for the period 1979-2014. This set was used to bias-correct the simulated discharge at HD river mouths from 1979-2019. The second set uses a further discharge simulation where we continued the GSWP3-based simulation with GSWP3 forcing until 2014. Again, the set of bias-correction factors was derived for the period 1979-2014 using discharge station observations. Then, this set was applied to bias-correct the simulated discharge at HD river mouths from 1901-1978. Detailed information you can find in the specified sections of the attached PDF https://www.wdc-climate.de/ui/entry?acronym=Biasc_hr_riverro_Eu_AdI_v1_0 Recently, a bug has been discovered in the part of the bias correction procedure, which transfers the bias correction factors from the station locations to the river mouths. Here, accidentally the bias correction factors from a previous simulation, which had utilized GSWP3 data, HydroPy and the HD model, were transferred to the river mouths for the whole considered period from 1901-2019. It can be noted that these factors still have improved the simulated inflows for most of the basins compared to the uncorrected HD model discharges. However, fixing this bug (see Version 1.1: https://www.wdc-climate.de/ui/entry?acronym=Biasc_hr_riverro_Eu_v1_1) has led to general improvement for most of the basins. Note that the other datasets of this Version 1.0 did not change. Acknowledgments This dataset was generated within the CoastalFutures project that was funded by the German Federal Ministry of Education and Research under grant number 03F0911A-K.

For the Helsinki Commission (HELCOM), annual waterborne basin inflows of total nitrogen (N) and total phosphorus (P) were compiled for the seven main Baltic Sea sub-basins (Sect. 1.1). In order to allow the utilization within a regional Earth System or ocean modelling framework, we redistributed these nutrient loads spatially and temporally using a dataset of bias corrected discharge that was generated with the Hydrological Discharge (HD) model (Sect. 1.2). Following the spatial and temporal downscaling procedure described in Sect. 1.3, we generated a dataset of daily riverine and annual direct nutrient loads (N and P) into the Baltic Sea at 1/12° resolution from 1901-2019. Detailed information you can find in the specified sections of the attached PDF https://www.wdc-climate.de/ui/entry?acronym=HELCOM_HD_info In November 2023, the bias corrected discharges were improved after a bug was found (see "Bias corrected high resolution river runoff over Europe (Version 1.0)", https://doi.org/10.26050/WDCC/Biasc_hr_riverro_Eu). Consequently, the redistributed HELCOM loads of nitrogen (N) and phosphorus (P) were also updated using the improved bias corrected discharge dataset. However, changes in the N and P loads into the Baltic Sea are marginal. The annual basin sums are the same, only daily values may have slightly changed.

1 Dataset description In ocean model or Earth System model applications, the riverine freshwater inflow is an important flux affecting salinity and marine stratification in coastal areas. However, in climate change studies, the river runoff based on climate model output often has large biases on local, regional or even basin wide scales. If these biases are too large, the ocean model forced by the runoff will drift into a different climate state compared to the observed state, which is especially relevant for semi-enclosed seas like the Baltic Sea. In order to fulfil the demands for low biases in river runoff, a three-part bias correction was developed by Hagemann et al. (in prep.) that comprises different correction factors for low, medium and high percentile ranges of river runoff over Europe. First, we utilized the global hydrology model HydroPy (Stacke and Hagemann 2021) and the Hydrological Discharge (HD) model (Hagemann et al. 2020) to simulate daily discharge time series over the European domain at 1/12° horizontal resolution Sect. 1.1) from 1901-2019. Then, we bias-corrected these time series as described in Sect. 1.2 to generate bias-corrected discharges at coastal ocean boxes of the European HD model domain from 1901-2019. 1.1 Century-long high-resolution discharge simulation over Europe Analogous to Hagemann and Stacke (2022), the global hydrology model HydroPy (Vs. 1.0.2 Stacke and Hagemann 2021) and the Hydrological Discharge (HD) model (Vs. 5.2.0, Hagemann et al. 2023) were used to simulate daily discharge time series over the European domain at 1/12° horizontal resolution. Daily data of two atmospheric datasets were utilized to force HydroPy that provided the input to the HD model. The Global Soil Wetness Project Phase 3 (GWSP3; Dirmeyer et al. 2006; Kim 2017) dataset is available at 0.5° resolution from 1901-2014. Here, we used the data from 1901-1978, and then the simulated time series were continued by using the WFDE5 dataset (Cucchi et al. 2020; 0.5° resolution) from 1979-2019. 1.2 Generation of bias corrected HD discharge data In order to apply the bias correction of Hagemann et al. (in prep.) to the simulated time series of daily discharge from 1901-2019, two sets of bias correction factors were derived. The first set uses the WFDE5-based discharges and discharge station observations for the period 1979-2014. This set was used to bias-correct the simulated discharge at HD river mouths from 1979-2019. The second set uses a further discharge simulation where we continued the GSWP3-based simulation with GSWP3 forcing until 2014. Again, the set of bias-correction factors was derived for the period 1979-2014 using discharge station observations. Then, this set was applied to bias-correct the simulated discharge at HD river mouths from 1901-1978. Detailed information you can find in the specified sections of the attached PDF (https://www.wdc-climate.de/ui/entry?acronym=Biasc_hr_riverro_Eu_AdI_v1_1). Recently, a bug has been discovered in the part of the bias correction procedure, which transfers the bias correction factors from the station locations to the river mouths. Here, accidentally the bias correction factors from a previous simulation, which had utilized GSWP3 data, HydroPy and the HD model, were transferred to the river mouths for the whole considered period from 1901-2019. It can be noted that these factors still have improved the simulated inflows for most of the basins compared to the uncorrected HD model discharges. However, fixing this bug has led to general improvement for most of the basins. Fig. 1 in the attached PDF (https://www.wdc-climate.de/ui/entry?acronym=Biasc_hr_riverro_Eu_AdI_v1_1) provides an example for the major Baltic Sea sub-basins and shows the inflow biases compared to HELCOM observational estimates. Note that the other datasets of Version 1.0 (https://doi.org/10.26050/WDCC/Biasc_hr_riverro_Eu) did not change.

For the Helsinki Commission (HELCOM), annual waterborne basin inflows of total nitrogen (N) and total phosphorus (P) were compiled for the seven main Baltic Sea sub-basins (Sect. 1.1). In order to allow the utilization within a regional Earth System or ocean modelling framework, we redistributed these nutrient loads spatially and temporally using a dataset of bias corrected discharge that was generated with the Hydrological Discharge (HD) model (Sect. 1.2). Following the spatial and temporal downscaling procedure described in Sect. 1.3, we generated a dataset of daily riverine and annual direct nutrient loads (N and P) into the Baltic Sea at 1/12° resolution from 1901-2019. Detailed information you can find in the specified sections of the attached PDF https://www.wdc-climate.de/ui/entry?acronym=HELCOM_HD_info In November 2023, the bias corrected discharges were improved after a bug was found (see "Bias corrected high resolution river runoff over Europe (Version 1.0)", https://doi.org/10.26050/WDCC/Biasc_hr_riverro_Eu). Consequently, the redistributed HELCOM loads of nitrogen (N) and phosphorus (P) were also updated using the improved bias corrected discharge dataset (see Version 1.1: https://www.wdc-climate.de/ui/entry?acronym=HELCOM_HD_v1_1). However, changes in the N and P loads into the Baltic Sea are marginal. The annual basin sums are the same, only daily values may have slightly changed.

Under the heading of the OSPAR convention (Sect. 1.1), the IGC-EMO database of daily freshwater inflows and nutrient loads was compiled by van Leeuwen and Lenhart (2021), which covers the major rivers discharging into the Baltic Sea, North Sea and Northeast Atlantic (Sect. 1.2). In this database, the data are distributed in separate ASCII files according to the respective country and river. In order to allow an easier utilization within a regional Earth System or ocean modelling framework, we mapped the IGC-EMO data onto the flow grid of the European 1/12° domain of the Hydrological Discharge (HD) model (Sect. 1.3). This mapping was done for daily time series of discharge, total nitrogen (N), total phosphorus (P) and Silicate for the period 1940-2022 following the procedure described in Sect. 1.4. Detailed information you can find in the specified sections of the attached PDF https://www.wdc-climate.de/ui/entry?acronym=IGC-EMO_HD_info