Hi, I'm an error. x

Example Model Simulation for Lake Erken from the EU Water JPI project PROGNOS


Authors:
Owners:
Resource type: Composite Resource
Storage: The size of this resource is 233.8 MB
Created: Aug 19, 2019 at 1:49 p.m.
Last updated: Aug 29, 2019 at 6:38 p.m.
DOI: 10.4211/hs.eace2a19e3134ba8a442d75a344d4885
Citation: See how to cite this resource
Content types: Multidimensional Content 
Sharing Status: Published
Views: 372
Downloads: 46
+1 Votes: Be the first one to  +1 this.  (You need to be logged in to rate this.)
Comments: No comments (yet)

Abstract

The European Union Water JPI (http://www.waterjpi.eu/) has funded the project PROGNOS (Predicting In-Lake Responses to Change Using Near Real Time Models http://prognoswater.org/). PROGNOS developed an integrated approach that couples high frequency (HF) lake monitoring data to dynamic lake water quality models to forecast short-term changes in lake water quality. Here we provide an archive that demonstrates model simulations that were run on Lake Erken as part of the PROGNOS project. Simulations were run using the GOTM (https://gotm.net/)and SELMA models that are coupled by the frame work for biogeochemical models (https://github.com/fabm-model). The lake model was calibrated using the program acpy (https://bolding-bruggeman.com/portfolio/acpy/) The measured data used for calibration in the format used by acpy are also included in this archive

All data were collected from Lake Erken the site of the Uppsala University Limnology field station (http://www.ieg.uu.se/erken-laboratory/). The simulation period is between 2004-2018, and we used a 5 year (1999-2003) model spin-up period prior to simulation and calibration. Model forcing data are obtained from meteorological stations at the Erken laboratory, and from Erken's routine monitoring program. To calibrate the model data sets of HF water temperature, chlorophyll fluorescence and dissolved oxygen are combined with laboratory measurement from the routine monitoring program. Calibration data for nutrients are only available from the Erken laboratories routine monitoring program All data files are formated as required by the lake models. Files that define the model parameterization (xml and yaml) are also included

Subject Keywords

  • No subject keywords have been added.
  • ${ k }

Duplicate. Keyword not added.

Error: ${ error }
Deleting all keywords will set the resource sharing status to private.

Resource Level Coverage

Spatial

Coordinate System/Geographic Projection:
WGS 84 EPSG:4326
Coordinate Units:
Decimal degrees
Longitude
18.6328°
Latitude
59.8358°

Temporal

Start Date:
End Date:

Content

README.MD

Files Defining Model Setup

hypsograph.dat

File containing data to define the depth vs area relationship for the lake basin, which in turn are used to calculate the water volume associated with a given depth. Data are in the format expected by the GOTM model. Line 1 contains two numbers the first is the number of depths at which the surface area is calculated the second defines the direction of the profile bottom to top = 1 or top to bottom =2 the remaining lines contain the surface area (m2) of the lake at the specified depth.

erken.xml

Parameter settings for the GOTM model These have been optimized to give the best match between simulated and measured water temperature. In xml format expected by GOTM. Parameter values in this file are default values for the mode

fabm.yaml

Parameter settings for the Selma model. These have been optimized to give the best match between simulated and measured values of chlorophyll oxygen NOX, PO4-P and Total P. This file is in yaml format expected by the GOTM/SELMA model Parameter values in this file are default values for the mode

GOTM Model Meteorological Forcing Data

Erken_MetFile_1960-2018.dat

These are data collected from an automated monitoring station located on Malma Island, a small island 500 m offshore from the Erken Laboratory (59.83909N 18.629558W). SMHI meteorological station at Svanberga (59.83232N 18.654550W)

More information on this data set is also available from Moras, S. (2019). Simulation of Lake Erken water temperature (1961-2017) with GOTM: Model configuration, input data, output data and observed water temperature, HydroShare, https://doi.org/10.4211/hs.7e5ec8c0e2b245199ab13cc9ae08b841

The data are formatted to work with the GOTM model

  • Date/Time yyyy-mm-dd hh:mm:ss
  • UWind U wind component (m s-1) For 1D GOTM only this component is used
  • VWind U wind component (m s-1) Not used
  • AirPressure_hPa Air pressure (hPa)
  • Air_Temp Air temperature (C)
  • RelHumidity_Avg Relative Humidity (percent)
  • Cloud Cover Fraction of sky (0-1)

Data in this file are at an hourly time step

ErkenSWR_1960-2018.dat

Short wave radiation data read in a separate input file, which has an hourly time step

  • Date/Time yyyy-mm-dd hh:mm:ss
  • SW_Rad_Avg Shortwave radiation (watts m-2)

ErkenDailyPrec_1960-2018.dat

Precipitation data in a separate input file, which has an daily time step

  • Date/Time yyyy-mm-dd hh:mm:ss
  • Precip Precipitaion (mm d-1)

ErkenWSE_1995-2018.dat

Lake water level elevation data in a separate input file, which has an daily time step

  • Date/Time yyyy-mm-dd hh:mm:ss
  • Water_Level_Avg Level of the lake surface water meters above sea level

Nutrient Loading Data

Kristine_InputQ1999-2018.dat

This is the file of the daily discharge and water temperature for the largest input to Lake Erken that comprises 50.7% of the total watershed area. Data are obtained from the SMHI S-HYPE model (http://www.smhi.se/en/research/research-departments/hydrology/hype-in-sweden-s-hype-1.7891) with the Kristineholm station having an outlet at, SWEREF99: 694802, 6640401. Note that these data begin 2004-01-01. To enable a 5 year spin-up of the model we have taken the period 2004-2008 and appended it to this file as the time period 1999-2003

  • Date/Time yyyy-mm-dd hh:mm:ss

  • Inflow Stream discharge (m3 s-1)

  • Temp C Stream water temperature (C)

Unguaged_InputQ1999-2018.dat

This file is for the discharge for the remaining area of the watershed that is not in the sub-basin measured at Kristineholm. Discharge is adjusted based on the areas of the Kristineholm watershed and the remaining ungauged area. Stream water temperature is assumed to be the same as that at Kristineholm. This file is in the same format as the Kristineholm file above.

InFlowChem1999-2018_Selma.dat

This is a file of water chemistry measurements made at Kristineholm as part of the Erken laboratory routine monitoring program. Sampling frequency is monthly to every two weeks. GOTM does a linear interpolation between the measured concentrations and combines these estimates with daily discharge data to calculate nutrient loads. Note that we assume the same nutrient concentrations would be representative for the ungauged portion of the watershed. All concentrations are in mmole per liter

  • Date/Time yyyy-mm-dd hh:mm:ss
  • PO4-P (mmole/l) PO4-P concentration in the stream water(mmole l-1)
  • TP (mmole/l) Total P concentration in the stream water(mmole l-1)
  • NOX-N (mole/l) Nitrate + Nitrite concentration in the stream water(mmole l-1)
  • NH4 (mole/l) Ammonia concentration in the stream water(mmole l-1)

Measurements used for Model Calibration

Files used for model calibration are not directly used force model simulations, but were used to calibrate the model in order to the best parameter values in the erken.xml and fabm.yaml files. They are included here to allow independent attempts to calibrate GOTM/SELMA or other models, and also provide and to provide data that can be compared against the model simulations. Model calibration was done using the Program ACPy (https://bolding-bruggeman.com/portfolio/acpy/). The file config_acpy.xml is the file that is used by the ACPy to define the model state variables and sources of data to be used for comparison in the model calibrations. Also in this file are all of the parameters that were optimized and the range of acceptable calibration values for each parameter.

Erken_DailyTempData_2004-2018_NoWinter.obs

During ice free conditions full profiles of water temperature are automatically recorded at the Eastern end of Lake Erken approximately 500 m to the NE of the Malma island meteorological station at a depth of 15.5 m (59.84297N 18.635433W). During the PROGNOS project 2 different systems were used both based on thermocouple sensors. Between 1989-2016 temperatures were measured every 30 min at 30 depths between 0.5 and 15 meters depth using a seasonally deployed floating system. Starting in 2016 an upgraded system measured temperatures at 50 depths between 1.0 and 15.5 m depth at hourly intervals. This system was designed to operate year round and is moored underwater below the depth of ice formation. Finally at the Malma Island Meteorological station water temperature were measured year round at 3 depths 1.0m, 3.0m, and 15.0 m. These three data sources were combined to create as complete a data record as possible. Prior to the development of the new temperature measurement system, the Malma Island temperature measurements were merged with data from the moored system in order to provide a continuous data record that covered the periods when the moored system was not in operation. Once the new system was in operation its results were append onto the end of the data file created from the other two data sources.

The GOTM model used for the simulations documented here and used in PROGNOS did not have a functioning ice model, but instead cut off surface heat exchange when the simulated surface water temperature became negative. This provided a very simple way to make continuous simulations that includes freezing conditions that would normally lead to the formation of ice. However, the temperature profiles during winter were not realistic, and could not be used for model calibration. For this reason we eliminated all data collected between 15 December - 31 March are excluded from the temperature data used for model calibration. The data file format is that expected by the ACPY program that we used to calibrate the mode.

Each file line contains:

  • Time stamp
  • Depth
  • Water temperature (C)

Water Quality Data Files

HF water quality measurements collected by a YSI EXO2 sonde (https://www.ysi.com/exo2) that was deployed on a profiling system (https://www.ysi.com/Pontoon-Vertical-Profiling-System) were used to calibrate the GOTM/SELMA model. Data used were dissolved oxygen, and chlorophyll fluorescence. This system was also deployed near Malma Island in the main basin of the lake the exact location (and maximum depth) varied from year to year. The present location is at 59.84530N 18.624217W and at depth of 18.0 meters. Profiles are collected every hour at 0.5 meter intervals. There is also a long-term routine monitoring program which collects a volume weighted sample of the epilimnion and hypolimnion water at one to two week intervals. Laboratory analysis of these samples provided measurements of PO4-P, Total P, NH4, NOX (nitrate + nitrite) and Chlorophylla that could be used for model calibration.Separate yearly files are provided for each of the water quality parameters and they are in the format expected by ACPY

Comb_Chl_YSI_Selma.obs

Spring_Comb_Chl_YSI_Selma.obs

Remaining_Comb_Chl_YSI_Selma.obs

These are chlorophyll data files that are created by merging data from both the automated and manual measurements. The automated profiles are averaged to calculate a mean daily profile. For the laboratory measurements we assigned a depth of -3 meters to the epilimnion sample and a depth of -15 meters to the hypolimnion sample. Evaluation of the chlorophyll data from both sources showed that the two data sources matched well during most of the year, but that the YSI data underestimated the lab measured chlorophyll concentrations during summer when large colonial cyanobacteria were present , especially in the presence of Gloeotrichia.colonies This is most likely because the larger (500 ml -1000ml) water sample used by the lab chlorophyll measurements more effectively sampled the inhomogenously distributed chlorophyll rich cyanobacteria colonies. During periods when cyanobacteria were present and there were large discrepancies between the laboratory and automated measurements, the manual measurements were removed from the data set. There are also three data files used for calibration Comb_Chl_YSI_Selma.obs contains the entire data record and can be used to calibrate all of the modeled phytoplankton groups (diatoms, flagellates and cyanobacteria) simultaneously. The remaining two files are used in a different calibration stragegy. Spring_Comb_Chl_YSI_Selma is a file containing data collected between (1 March and 16 May) the period during which the spring bloom normally occurs and it is used to specifically calibrate the model parameters associated with diatoms. Remaining_Comb_Chl_YSI_Selma.obs is a file containing the measured chlorophyll data from the remainder of the year and is used to calibrate the flagellate and cyanobacteria groups.

Each file line contains:

  • Time stamp
  • Depth
  • Chlorophyll concentration (mg l-1)

MeanDaily_2004-2018_Erken_O2_mmole.obs

This file contains oxygen data from both the YSI automated profiling system and field measurements made during the routine sampling program. The hourly automated profiles are averaged to calculate a mean daily profile. During the routine sampling program measurements were made between the surface and 20 meters at 1 meter intervals using a temperature /oxygen probe at the deepest area of the lake. Oxygen profiles from both the routine monitoring program were merged in this file. The format is that expected by ACPY.

Each file line contains:

  • Time stamp
  • Depth
  • Oxygen concentration (mmole l-1)

Comb_TotP_Selma.obs

Comb_PO4_Selma.obs

Comb_NOX_Selma.obs

Comb_NH4_Selma.obs

These are Lake Erken nutrient data collected by the routine monitoring program. No automated measurements of these nutrient are collected. A depth of -3 meters is assigned to the volume weighted epilimnion sample and a depth of -15 is assigned to the volume weighted hypolimnion sample.

Each file line contains:

  • Time stamp
  • Depth
  • Nutrient concentration (mmole l-1)

Water Quality Data Files

These are the files that are used to set up the GOTM and SELMA model. The final optimized parameter values are contained in these files.

exken.xml

  • This is the file that contains all of the parameter values and other information such as input file locations that are needed to run the GOTM model. The final value of the calibrated parameters are contained in this file.

fabm.yaml

  • This is the file that contains the parameter values associated with the SELMA model. The final value of the calibrated parameters are contained in this file

Example Simulation Output

Output.nc

This is a file in NetCDF format and contains a large amount of simulated output as well as the model forcing data. The simulation that produced this output is are forced by the data described above and model parameterization is defined by the setup files described above. The simulation period was between 2004-2018 as defined in erken.xml. To simply view the data contained in the file it is recommended to use the program PyNcView (https://sourceforge.net/projects/pyncview/)

References

Sources

Derived From: Moras, S. (2019). Simulation of Lake Erken water temperature (1961-2017) with GOTM: Model configuration, input data, output data and observed water temperature, HydroShare, https://doi.org/10.4211/hs.7e5ec8c0e2b245199ab13cc9ae08b841

Credits

Funding Agencies

This resource was created using funding from the following sources:
Agency Name Award Title Award Number
European Union ERA-NET WaterWorks2014 Cofunded Call PROGNOS
FORMAS Predicting in-lake responses to change using near real time models (PROGNOS) 2016-00006
Swedish Infrastructure for Ecosystem Science (SITES)

How to Cite

Pierson, D. (2019). Example Model Simulation for Lake Erken from the EU Water JPI project PROGNOS, HydroShare, https://doi.org/10.4211/hs.eace2a19e3134ba8a442d75a344d4885

This resource is shared under the Creative Commons Attribution CC BY.

 http://creativecommons.org/licenses/by/4.0/
CC-BY

Comments

There are currently no comments

New Comment

required