Coupling an HSPF Water Quality Model with an EFDC+ Model
Developing a hydrodynamic or a water quality model may require input of boundary conditions from other models. In a previous blog we described how to automate the data processing steps to loosely couple HSPF and EFDC+ models and run both the models using a Python script. In that script, we focused only on coupling flows from HSPF.
In this post, we describe how to couple water quality constituents from an HSPF model to an EFDC+ model. The general guidance here can be applied to other watershed and receiving water models as well (e.g. W2, ICM, SWAT, SWMM/SUSTAIN).
You can use the guidance and equations in this post to develop your own data processing scripts or do the data manipulation in the HSPF model itself. We assume that only the free-flowing reaches or mixed reservoir (RCHRES) from the HSPF model drain to the EFDC+ model. We recommend that if there is a local drainage to the EFDC+ model domain, the modelers should add a fake reach to the HSPF model and then connect it to the EFDC+ model for simplicity.
As experienced modelers know, HSPF and EFDC+ model organic and biological constituents differently. For example, HSPF simulates only one phytoplankton species, whereas EFDC+ can simulate unlimited phytoplankton species. To maintain the mass balance and ensure accurate coupling between the two models, it is important to properly convert the constituents simulated in HSPF to the constituents simulated in EFDC+. Table 1 provides a list of the constituents required by EFDC+, relevant output from the HSPF model, and the conversion equation or factors that may be needed. In some cases, users may need to manipulate more than one time series from HSPF to get the desired EFDC+ boundary time series. For example, users will have to add silt (SSED-2) and clay (SSED-3) time series from the HSPF model to calculate inorganic cohesive solids time series from HSPF.
Table 1. Linkage of constituents simulated in HSPF to the constituents simulated in EFDC+
| EFDC+ Input | Units | Conversion factor / Equation | HSPF Output1 | HSPF Output Description | Units |
1 | Flow | m3/s | 1 | HYDR:RO or HYDR:O | Total Outflow or Outflow from a Reach | m3/s |
2 | Water temperature | °C | 1 | HTRCH:TW | Water Temperature | °C |
3 | Inorganic cohesive solids | mg/l | 1 | SEDTRN:SSED:2, SEDTRN:SSED:3 | Silt Concentration, Clay Concentration | mg/l |
4 | Inorganic non-cohesive solids | 1 | SEDTRN:SSED:1 | Sand concentration | mg/l | |
5 | Algae concentration for each group | g Carbon /ml | (C/Chl)2 * fraction of the algae group | PLANK:PHYTO | Phytoplankton as Chlorophyll a | µg/l |
6 | Refractory Particulate Organic Carbon | mg/l | (CBOD3/(CVBO4/CDW5)+ORC6)*F_R7 | OXRX:BOD and PLANK:PKST3:3 | CBOD concentration, dead refractory carbon concentration | mg/l |
7 | Labile Particulate Organic Carbon | mg/l | (CBOD/(CVBO/CDW) + ORC)* F_L8 | mg/l | ||
8 | Dissolved Organic Carbon | mg/l | (CBOD/(CVBO/CDW) + ORC)* F_D9 | mg/l | ||
9 | Refractory Particulate Organic Phosphorus | mg/l | (CBOD/(CVBO/CDW) *P/C10 + ORP11)* F_R | OXRX:BOD and PLANK:PKST3:2 | CBOD, dead refractory phosphorus | mg/l |
10 | Labile Particulate Organic Phosphorus | mg/l | (CBOD/(CVBO/CDW) *P/C + ORP)* F_L | mg/l | ||
11 | Dissolved Organic Phosphorus | mg/l | (CBOD/(CVBO/CDW)*P/C + ORP)* F_D | mg/l | ||
12 | Total Phosphate | mg/l | 1 | NUTRX:DNUST:4 | Orthophosphorus as phosphorus | mg/l |
13 | Refractory Particulate Organic Nitrogen | mg/l | (CBOD/(CVBO/CDW)*N/C12 + ORN13)* F_R | OXRX:BOD and PLANK:PKST3:1 | CBOD, dead refractory nitrogen | mg/l |
14 | Labile Particulate Organic Nitrogen | mg/l | (CBOD/(CVBO/CDW)*N/C + ORN)* F_L | mg/l | ||
15 | Dissolved Organic Nitrogen | mg/l | (CBOD/(CVBO/CDW)*N/C + ORN)* F_D | mg/l | ||
16 | Ammonia as Nitrogen | mg/l | 1 | NUTRX:DNUST:2 | Total ammonia as nitrogen | mg/l |
17 | Nitrate as Nitrogen | mg/l | 1 | NUTRX:DNUST:1 and NUTRX:DNUST:3 | Nitrate as nitrogen, nitrite as nitrogen | mg/l |
18 | COD | mg/l | n/a | NA | mg/l | |
19 | DO | mg/l | 1 | OXRX:DOX | Dissolved oxygen | mg/l |
1These constituents names show the group name, constituent name and the subscript from the HSPF model. For their complete definitions, refer to the HSPF help document.
2Ratio of carbon to chlorophyll a (mgC: µg Chl a). Typically 0.025 (Wells et al. 2008). Refer to the model configurations.
3CBOD = carbonaceous BOD or the output BOD time-series from the HSPF model. [OXRX:BOD]
4CVBO = Conversion from oxygen to dry weight of biomass. Typically 1.3. Refer to the HSPF model.
5CDW = ratio of carbon: dry weight. Typically 0.49. Refer to the HSPF model.
6 ORC = Dead refractory organic carbon time-series output from the HSPF model. [PLANK:PKST3:3]
7F_R = Fraction of organic carbon/nitrogen/phosphorus that is refractory in the EFDC model.
8 F_L = Fraction of organic carbon/nitrogen/phosphorus that is labile in the EFDC+ model.
9 F_D = Fraction of organic carbon/nitrogen/phosphorus that is dissolved in the EFDC+ model.
10P/C = Stoichiometric ratio of phosphorus to carbon-based on Redfield ratio (typically, 41.1 mg C/mg-P).
11ORP = Dead refractory organic phosphorus time-series output from the HSPF model. [PLANK:PKST3:2]
12 N/C = Stoichiometric ratio of nitrogen to carbon-based on Redfield ratio (typically, 5.7 mg C/mg-N)
13 ORN = Dead refractory organic nitrogen simulated in the HSPF model. [PLANK:PKST3:1]
In the case of organic materials, the labile fraction of organic material (as part of CBOD) and dead refractory portion of organic material simulated in HSPF are summed to calculate total organic material, and then they are distributed to labile, refractory, and dissolved portions (e.g. rows 6, 7, and 8 in Table 1) for the EFDC+ model. The organic material from living phytoplankton in the HSPF model is not provided to the EFDC+ model, as the biomass of phytoplankton from the HSPF model is separately input into the EFDC+ model (line 5 in Table 1). The fraction of refractory, labile, and dissolved splits or organic matter may be selected based on the study by Wells et al., 2008 (Table 2).
Table 2. Refractory, Labile and Dissolved Splits for Organic Matter (Wells et al., 2008).
| Refractory F_R | Labile F_L | Dissolved F_D |
| RPOM | LPOM | DOM |
TOC | 0.25 | 0.25 | 0.5 |
TOP | 0.25 | 0.25 | 0.5 |
TON | 0.25 | 0.25 | 0.5 |
Conclusions
Coupling HSPF models to EFDC+ models is not necessarily a complicated process but requires a complete understanding of the water quality module in HSPF and in EFDC+. A one-to-one relationship between parameters is not available. However, reasonable conversions can be made based on the guidance provided here. In the future, EEMS will be enhanced to make these conversions, thereby allowing automated linkage between HSPF and EFDC+.
References
Di Toro, D.M. 2001. Sediment Flux Modeling. Wiley Interscience, New York, NY.
Michael Baker Jr., Inc., Aqua Terra, Consultants, and Dynamic Solutions, LLC. 2015. Setup, Calibration, and Validation for Illinois River Watershed Nutrient Model and Tenkiller Ferry Lake EFDC Water Quality Model (Final Report). Report prepared for USA EPA Region 6. August 7, 2015
Wells, S.A., V.I. Wells, and C.J. Berger. 2008. Water Quality and Hydrodynamic Modeling of Tenkiller Reservoir. Expert report prepared for State of Oklahoma in Case No. 05-CU329-GKF-SAJ State of Oklahoma v. Tyson Foods, et al. (in the United States District Court for the Northern District of Oklahoma).
