Question on Time-Varying Channel External Load for Nonpoint Source TP in SWAT+

[최혜빈]

Hello SWAT+ Group members,

I would like to ask for confirmation regarding the structural capability of SWAT+ to handle time-varying external loads at the channel level, specifically for nonpoint source (NPS) phosphorus (TP).

Background

In my study, I am working with already-estimated nonpoint source TP loads at the administrative level (읍·면·동), based on official total maximum load (TMDL) data.
Because these loads are already defined, my intention is:

• NOT to re-calculate TP generation using HRU management (fertilizer, manure, soil P cycling),

• but to use SWAT+ only for routing, attenuation, and storage processes after generation.

Using the Korean TMDL technical guideline, I applied a rainfall delivery ratio concept to split annual TP loads into:

• baseflow (constant daily load), and

• rainfall-driven load (applied only on rainy days).

This resulted in subbasin-level daily TP loads (kg/day) that vary by day, with higher values on rainfall days and near-zero values on dry days.

---

Planned Approach

My original plan was:

1. Allocate administrative-level TP loads to HRUs by area,

2. Aggregate HRU loads to subbasin level,

3. Input the resulting daily TP loads (kg/day) as external loads to the corresponding channels, and

4. Let SWAT+ simulate downstream routing and attenuation.

---

Issue Encountered

After inspecting the SWAT+ SQLite database structure, I found that channel external loads are defined through:

• dr_path_del

• dr_path_val

These tables define one constant value per path, and do not include any time dimension (no year, day, or timestep fields).

As a result, it appears that:

• Channel external loads are applied as constant daily values over the entire simulation period, and

• Daily time-varying external loads (e.g., rainfall-only NPS loads) cannot be represented at the channel level using this structure.

This limitation seems to persist even if the loads are defined at the subbasin scale and routed through channels.

---

My Question

Could you please confirm whether the following understanding is correct?

In SWAT+, time-varying (daily) external nutrient loads cannot be applied at the channel level using the dr_path_del / dr_path_val structure, and channel external loads are inherently constant over time unless nutrients are generated internally at the HRU level.

If there is any supported method (e.g., alternative routing structures, time-series external paths, or project settings) that allows daily varying channel external loads for nonpoint sources, I would greatly appreciate any guidance or references.

Thank you very much for your time and help.

[Natalja C.]

Hello,

It seems there might be a mix-up in the conceptual model setup regarding how SWAT+ handles different load types.

The SWAT+ routing mechanism is essentially a 0D model. It is designed to "pass" information from one discrete point (a channel segment) to another point (the next downstream segment). Because routing is a point-to-point process, it is inherently structured to handle point sources, which can be defined with various temporal resolutions: daily, monthly, annual, or average annual.

When you attempt to incorporate distributed nonpoint source data directly into the channel via dr_path_del or dr_path_val, you encounter the "constant value" limitation because those tables are not designed for time-series inputs. They represent a static connection rather than a dynamic loading interface.

If your TP loads are already estimated as daily values at the subbasin level, you might consider treating them as point sources (in-stream inlets) rather than trying to force them through the HRU/landscape delivery structure. This would allow you to bypass the HRU-level generation while still utilizing the daily variability you've calculated.

Otherwise, if you intend to use the landscape-to-channel "pathway," SWAT+ expects that variability to be driven by the HRU's response to weather, not a pre-defined daily time series at the delivery point.

Best,

Natalja