21. LID, groundwater, snowmelt and water quality

The Project menu's second group of dialogs edits the sections that describe what happens on a subcatchment beyond runoff, and what a node does to the water passing through it: LID controls and their placement, aquifers and groundwater flow, snow packs, pollutant buildup and washoff, land use coverages and initial loadings, treatment, and the RDII unit hydrographs. Each opens from the Project menu; the per-object ones also open from the property sheet of the subcatchment or node they belong to.

Column names and meanings below are the SWMM 5.2 User's Manual (EPA/600/R-22/030), Appendix D, section by section; the physics is in the Reference Manual Volume I (Hydrology) for LID, groundwater, snowmelt and RDII and Volume III (Water Quality) for buildup, washoff and treatment. Where a rule is the engine's rather than the manual's (how many fields a row must have, which trailing fields may be left out) it comes from the EPA source: lid.c, gwater.c, snow.c, landuse.c, subcatch.c, treatmnt.c, rdii.c.

21.1 How these dialogs edit the file

Every dialog works on a draft of the rows that belong to one name — the LID control, the subcatchment, the land use, the node — as the raw fields of each row. OK (and Apply) writes the draft back as one command and one undo step; Cancel or closing the window drops it. A row the draft did not change is compared field for field with the file's and left alone, so:

• OK on a dialog you did not edit writes nothing, adds no undo step and leaves the file byte for byte as it was (the test suite opens every dialog on the EPA sample models LID_Model.inp, Groundwater_Model.inp and Site_Drainage_Model.inp, presses OK, and compares the text);
• an edited row is rewritten in place, keeping its line and any trailing ; comment; new rows go in after the object's last row, so an LID's layers stay together; a removed row is deleted;
• a trailing optional field you clear is dropped from the row rather than written empty, so the engine reads the row the way it did before.

The list dialogs (LID Controls, Aquifers, Snow Packs, Unit Hydrographs) have + and − on the left: + adds an object with the defaults below as its own undo step, − deletes the chosen one together with the rows that use it ([LID_USAGE] rows for an LID control, [GROUNDWATER] rows for an aquifer, [RDII] rows for a unit hydrograph set; a snow pack's subcatchments keep the name and the validator reports them). The Name field renames the object and every column that refers to it (the engine has no rename of its own for these; AquaSWMM rewrites each referring row). Choosing another object in the list while the draft has edits applies them first, as the Curves dialog does.

Cells are the same widgets as the property sheet: a number, a keyword list, or a combo over the names the model has (* for none). Units follow [OPTIONS] FLOW_UNITS: inches and in/hr for a CFS/GPM/MGD model, millimetres and mm/hr for CMS/LPS/MLD.

21.2 LID Controls

[LID_CONTROLS] (UM Appendix D; RM I §3.6, LID Controls). One process is a Name Type row followed by one row per layer: Name Layer p1 p2 …. Types: BC bio-retention cell, RG rain garden, GR green roof, IT infiltration trench, PP permeable pavement, RB rain barrel, VS vegetative swale, RD rooftop disconnection. The dialog shows, for the chosen type, each layer it uses with a tick box (ticked = the row exists) and the layer's fields:

Which layers a type uses, and which the engine insists on (lid.c validateLidProc, ERROR 184 when missing):

A layer row the type does not use is shown as ignored and kept (the engine reads and ignores it). Changing the Type keeps every row and adds, with defaults, the layers the new type requires. As the engine reads it at the bottom prints the rows as they will be written and names any required layer still missing. + adds a bio-retention cell named LID1, LID2, …; Duplicate copies the chosen control under a new name; − deletes it with the [LID_USAGE] rows that place it. The defaults for a new layer are: surface 0 / 0.0 / 0.1 / 1.0 / 5; soil 12 / 0.5 / 0.2 / 0.1 / 0.5 / 10.0 / 3.5 (the soil of the planter boxes in the EPA sample LID_Model.inp); pavement 6 / 0.15 / 0 / 100 / 0; storage 12 / 0.75 / 0.5 / 0 / NO; drain 0 / 0.5 / 0 / 6; drainage mat 3 / 0.5 / 0.1. Ticking REMOVALS adds an empty row and Add pollutant adds a pollutant / percent pair to it; a REMOVALS row still without a pair when you press OK is left out, because a two-field row is what the engine reads as the type row.

Checks: the type is one of the eight; each layer keyword is one of the seven; a layer row has at least the fields its reader needs (SURFACE and PAVEMENT 7, SOIL 9, STORAGE and DRAIN 6, DRAINMAT 5, REMOVALS 4, counting the name and keyword); parameters are numbers; VegFrac is below 1; REMOVALS name existing pollutants with percentages between 0 and 100; a control without a type row (ERROR 183) and a required layer missing (ERROR 184) are errors; an ignored layer is a warning.

21.3 LID Usage

[LID_USAGE] (UM Appendix D; RM I §3.6). One row per LID placement in a subcatchment: Subcatchment LID Number Area Width InitSat FromImp ToPerv [RptFile DrainTo FromPerv]. Pick the subcatchment at the top (the dialog opens on the selected one, or from its sheet), edit the rows, Add LID unit adds one.

The dialog totals Number × Area against the subcatchment's Area (acres × 43 560 ft², or hectares × 10 000 m²) and says so in red when the LID units are larger than the subcatchment; the engine refuses that model with ERROR 187. Checks: the LID exists; Number, Area, Width are numbers; InitSat, FromImp and FromPerv are between 0 and 100; DrainTo, when not *, is a node or subcatchment. The sheet's LID Usage… button shows the count (2 LID units).

21.4 Aquifers

[AQUIFERS] (UM Appendix D; RM I §5, Groundwater). One row per aquifer, fourteen columns:

+ adds Aquifer1, … with the values of the EPA sample Groundwater_Model.inp; the Name field renames the aquifer in every [GROUNDWATER] row. Checks (gwater.c, ERROR 109): porosity positive, field capacity below porosity, wilting point below field capacity, conductivity positive; the pattern exists.

21.5 Groundwater

[GROUNDWATER] and [GWF] (UM Appendix D; RM I §5). Tick Groundwater flow from this subcatchment to give the subcatchment a row (Subcatchment Aquifer Node Esurf A1 B1 A2 B2 A3 Dsw [Egwt Ebot Wgr Umc]):

Under it, two text boxes hold the [GWF] expressions: LATERAL replaces the lateral flow equation, DEEP the deep percolation; a blank box means the built-in equation and no row. The variables the engine allows (gwater.c): Hgw water table height, Hsw surface water height, Hcb channel bottom height, Hgs ground surface height, Ks saturated conductivity, K unsaturated conductivity, Theta upper-zone moisture, Phi porosity, Fi surface infiltration, Fu upper-zone percolation, A subcatchment area; and the parser's functions (sin, cos, tan, cot, abs, sgn, sqrt, log, exp, asin, acos, atan, acot, sinh, cosh, tanh, coth, log10, step). The dialog checks each identifier as you type and the validator reports an unknown one as an error (the engine's ERROR 233). An expression is written token by token; one you did not change keeps its original row. Checks: the aquifer exists, the node exists, the numbers are numbers, the [GWF] type is LATERAL or DEEP.

21.6 Snow Packs

[SNOWPACKS] (UM Appendix D; RM I §4, Snowmelt). A pack is up to four rows, one per keyword, each with a tick box in the dialog:

+ adds SnowPack1, … with all four rows at the EPA GUI's defaults (0.001 / 0.001 / 32 / 0.10 / 0 / 0 / 0; removal 1.0 and zeros); the Name field renames the pack in every subcatchment's SnowPack column. A subcatchment takes a pack through the SnowPack field on its property sheet, a combo over the packs. The Temperature and adjustments fold holds the [TEMPERATURE] and [ADJUSTMENTS] sections as text, written with the same OK: TIMESERIES name, FILE "name" [start] [C|F], WINDSPEED MONTHLY v1…v12, WINDSPEED FILE, SNOWMELT Stemp ATIwt RNM Elev Lat DTLong, ADC IMPERVIOUS f0…f9, ADC PERVIOUS f0…f9; and TEMPERATURE | EVAPORATION | RAINFALL | CONDUCTIVITY followed by twelve monthly values. Checks (snow.c): the keyword is one of the four; a PLOWABLE, IMPERVIOUS or PERVIOUS row has its seven numbers (nine fields), a REMOVAL row its six; a subcatchment's SnowPack names a pack that exists; Scatch names a subcatchment.

21.7 Buildup / Washoff

[BUILDUP] and [WASHOFF] (UM Appendix D; RM III §2, Pollutant Buildup and §3, Pollutant Washoff). The left list is the land uses; each has one buildup and one washoff row per pollutant. Land uses and pollutants themselves are added with Project → Land Uses… and Project → Pollutants… (chapter 8).

[BUILDUP]: LandUse Pollutant Function Coeff1 Coeff2 Coeff3 PerUnit.

Buildup is per PerUnit: AREA (per acre, or per hectare in SI) or CURB (per unit of curb length, in whatever unit the subcatchment's CurbLen was entered). Mass is in the pollutant's units (lb or kg; counts for a #/L pollutant).

[WASHOFF]: LandUse Pollutant Function Coeff1 Coeff2 SweepRmvl BmpRmvl.

SweepRmvl and BmpRmvl are the percentages removed by street sweeping and by BMPs. Add buildup row / Add washoff row add a NONE row for the first pollutant. Checks (landuse.c): the land use and pollutant exist; the function keyword is one of the list; a buildup row other than NONE has all seven fields and a PerUnit of AREA or CURB; an EXT row names an existing time series; a washoff row other than NONE has at least five; coefficients are numbers; the percentages are 0–100.

21.8 Land Use Coverages

[COVERAGES] (UM Appendix D; RM III §2). Pairs of land use and the percent of the subcatchment's area under it. Add land use adds a pair; the total is printed under the table and turns red above 100 %. The engine accepts a total below 100 % (the rest generates no buildup); above it, buildup and washoff are over-counted, which the validator reports as a warning. A file row may carry several pairs; the dialog unrolls them and, only when you change something, writes one pair per row. Checks: each land use exists; each percent is a number.

21.9 Initial Loadings

[LOADINGS] (UM Appendix D; RM III §2). Pairs of pollutant and the buildup present on the subcatchment when the run starts, in mass per unit area (lb/acre or kg/ha), overriding what [OPTIONS] DRY_DAYS would build up. Add pollutant adds a pair. Checks: the pollutant exists; the buildup is a number.

21.10 Treatment

[TREATMENT] (UM Appendix D; RM III §5, Treatment). One expression per pollutant at the node: Node Pollutant Expression, where the expression is either R = …, the fraction removed, or C = …, the outlet concentration. Add treatment adds R = 0 for the first pollutant. The engine (treatmnt.c) joins the tokens after the pollutant into one string, so an expression may contain spaces; the dialog writes it back token by token and reads it the same way. Variables: HRT hydraulic residence time (hours), DT the routing time step (seconds), FLOW the inflow (in the model's flow units), DEPTH the node's water depth, AREA its surface area, any pollutant's inflow concentration by name, and R_<pollutant> for another pollutant's removal fraction; plus the parser's functions listed under Groundwater. A line under a row says what is wrong with it (must start with R or C, needs an =, unknown variable) and the validator reports the same as an error (ERROR 233). Checks: the pollutant exists; the expression passes.

21.11 Unit Hydrographs

[HYDROGRAPHS] (UM Appendix D; RM I §6, RDII). A set is a Name RainGage row and any number of Name Month Response R T K [Dmax Drec D0] rows. Rain gage edits the first; the table holds the rest, one per month and response, with Add hydrograph row:

+ adds UH1, … on the model's first rain gage with one SHORT, MEDIUM and LONG row for ALL months, R = 0 (no RDII until you set it) and times to peak of 1, 4 and 24 hours; the Name field renames the set in every [RDII] row. Checks (rdii.c): the set has a gage row and the gage exists; the month and response keywords are known; R, T, K are numbers, R between 0 and 1, T not negative (ERROR 151); the responses of a month sum to at most 1.01 (ERROR 153), which the dialog also prints.

21.12 RDII Inflow

[RDII] (UM Appendix D; RM I §6). Tick RDII inflow at this node to give the node a row: UnitHydrograph, the [HYDROGRAPHS] set whose gage's rainfall drives the inflow, and SewerArea, the sewershed area (acres or hectares) contributing rainfall-derived infiltration and inflow at the node. Checks: the set exists; the area is a number and not negative (ERROR 155).

21.13 Property sheet

A subcatchment's sheet ends with four buttons and what the subcatchment has behind each: LID Usage… (no LID units, 1 LID unit, 2 LID units), Groundwater… (none, aquifer flow, aquifer flow, 1 GWF expression(s)), Land Use Coverages… (2 land use(s), 100 %), Initial Loadings… (1 pollutant(s)). Its SnowPack field is a combo over the [SNOWPACKS] names. A node's sheet has Treatment… (1 expression(s)) and RDII… (set UH1). Pollutant inflows and dry-weather pollutant rows are the node sheet's Inflow and Dry Weather Flow sub-sheets (chapter 5): their Constituent field lists FLOW and every pollutant.

21.14 Worked example: LID_Model.inp

Open swmm/tests/fixtures/epa-samples/LID_Model.inp (the EPA's LID sample: six LID processes on a nine-subcatchment site) and Project → LID Controls…. The list shows GreenRoof (BC), PorousPave (PP), Planters (BC), InfilTrench (IT), RainBarrels (RB), Swale (VS). Choose Planters: a bio-retention cell with SURFACE (StorHt 6, VegFrac 0.0, Rough 0.0, Slope 0.0, Xslope 5), SOIL (Thick 12, Por 0.5, FC 0.2, WP 0.1, Ksat 0.5, Kslope 10.0, Suct 3.5), STORAGE (Height 12, Vratio 0.5, Seepage 0.2, Vclog 0, Covrd NO) and DRAIN (Coeff 0, Expon 1, Offset 0.5, Delay 6, Hopen 0, Hclose 0). REMOVALS is unticked. Set Thick to 18 and OK: one undo step, Planters SOIL 18 0.5 0.2 0.1 0.5 10.0 3.5 in the file, every other row of the section as it was. Ctrl+Z puts 12 back.

Select subcatchment S4 on the map and click LID Usage… on its sheet: one row, Planters × 30 at 500 ft² each, Width 0, FromImp 80 (80 % of S4's impervious runoff is routed into the planters), ToPerv 0, RptFile and DrainTo *, FromPerv 0. The line under the table reads 1 LID unit(s) cover 15000 of 297079.2 (Number × Area) — S4 is 6.82 acres. Type 300000 into Area and the line turns red with ERROR 187; Cancel.

Project → Validate Model on the unedited sample reports no errors.

21.15 Limitations

• The dialogs write values as text and check what the engine's readers check at input; they do not run the process models, so a physically odd but syntactically valid layer (a soil with Ksat 0) passes.
• [LID_CONTROLS] rows for a type the engine does not know are shown and kept but not interpreted; a layer row with more fields than the layout shows its extra fields read-only.
• [COVERAGES] and [LOADINGS] rows carrying several pairs are unrolled for editing and written one pair per row only when you change the pairs; otherwise they are untouched.
• [GWF] and [TREATMENT] expressions are checked for their identifiers, not parsed for syntax: an unbalanced parenthesis is the engine's ERROR 233 at run time.
• [TEMPERATURE] and [ADJUSTMENTS] are edited as text; the validator does not check them.
• [EVAPORATION], [TRANSECTS], [STREETS], [INLETS], [EVENTS] still have no dialog (chapter 8, What has no dialog).
• The engine's ERROR 188 (LID capture area larger than the impervious area) is not pre-checked; the LID area total (ERROR 187) is.