All programs › BMPFast › Questions answered
Dry retention basins in series
Two dry retention basins in series came out below the same two basins standing alone. How BMPFast Plus calculates retention in series today, why that happens, and a volume balance proposed in its place, with the program's own numbers.
Download file 1: Basin 2 is 0.8 ac-ft (.bmpf) · Download file 2: Basin 2 is 6.5 ac-ft (.bmpf)
Under review, November 11, 2026. This page shows how BMPFast Plus 2.0.4 calculates retention basins in series, and a change proposed for the next version. Dr. Marty Wanielista and the review team are checking the proposal. This page will be updated with their decision.
1The question
A beta tester modelled two identical catchments, each with a dry retention basin. The only difference was the basins' volumes: 0.5 ac-ft in Catchment 1 and 0.8 ac-ft in Catchment 2. They set up two routings. In one, Catchment 1 drains to Catchment 2. In the other, each catchment drains straight to the outlet (the routing named Individual).
Since Catchment 2's BMP has additional unused treatment volume not needed for its own runoff, it uses this surplus retention volume to attenuate Catchment 1's overflow. As such, the annual TP and TN discharging through the system outlet should decrease for the In Series Routing compared to Stand-alone (individual) routing … This is NOT what BMPFast Plus outputs though. It shows more Annual removal when each BMP only receives nutrients from its basin.
A second file (file 2) made the gap larger. With Catchment 2's basin raised to 6.5 ac-ft, the series removed 82.48 % and the basins alone 82.89 %. The tester expected about 99 %.
Short answer: the tester is right. A series of dry retention basins can never remove less than the same basins standing alone. The downstream basin captures everything it would capture alone, plus some of the upstream overflow.
BMPFast Plus's default method today is Treatment Depth Allocation. It shrinks the downstream basin's treatment depth to make room for the upstream water, but it never credits the downstream basin with removing any of that water. Section 5 sets out a volume balance to replace it. Today, file 1 gives 69.82 % in series against 73.76 % for the basins alone, and file 2 gives 82.48 % against 82.89 %. The volume balance gives 73.76 % on file 1 (the same as the basins alone) and 99.08 % on file 2.
2The example
| Input | Catchment 1 | Catchment 2 |
|---|---|---|
| Project (General Site Information) | ||
| Meteorological zone · mean annual rainfall | Florida Zone 2 · 52 in | |
| Catchment (post-development) | ||
| Area (ac) | 20 | 20 |
| Non-DCIA curve number · DCIA | 35 · 30 % | 35 · 30 % |
| EMC N / P (mg/L) | 1.85 / 0.31 | 1.85 / 0.31 |
| Annual runoff (ac-ft/yr) | 21.32 | 21.32 |
| Load N / P (kg/yr) | 48.632 / 8.149 | 48.632 / 8.149 |
| Dry retention basin | ||
| Retention volume (ac-ft) | 0.5 | 0.8 (file 2: 6.5) |
| Treatment depth over its own catchment, volume × 12 ÷ area (in) | 0.30 | 0.48 (file 2: 3.90) |
| Removal standing alone, N and P | 65.88 % | 81.65 % (file 2: 99.90 %) |
With the same EMCs everywhere, every percentage on this page is the same for nitrogen and phosphorus, and the same as the percentage of runoff volume captured. Removals are read from the Zone 2 retention tables at CN 35 and DCIA 30 %. In the files, Basin 2's recovery time is 0 hours and Basin 1's is 72; that does not change any number here.
3What BMPFast Plus does today: Treatment Depth Allocation
On a Catchment Routing's Entry tab, Retention basins in series starts on Treatment Depth Allocation. The program walks down the strand (the chain of catchments that drain one to the next, then to the outlet). At each catchment it finds the treatment depth of the basins so far, then removes that catchment's own load at that depth. Load arriving from upstream passes through with no further removal. Here are the steps on file 1, in the program's order.
-
Catchment 1. Only Basin 1 is above this point:
Treatment depth (TD) = 0.5 ac-ft × 12 ÷ 20 ac = 0.30 in → removal 65.88 % N leaving = 48.632 × (1 − 0.6588) = 16.593 kg/yr P leaving = 8.149 × (1 − 0.6588) = 2.780 kg/yr -
Catchment 2. The volume so far is spread over the area so far. The removal at that depth is applied to Catchment 2's own load only:
TD = (0.5 + 0.8) ac-ft × 12 ÷ (20 + 20) ac = 0.39 in → removal 73.76 % Own N leaving = 48.632 × (1 − 0.7376) = 12.759 kg/yr; plus 16.593 from Catchment 1, untreated = 29.352 kg/yr Own P leaving = 8.149 × (1 − 0.7376) = 2.138 kg/yr; plus 2.780 from Catchment 1, untreated = 4.918 kg/yr -
At the outlet.
N: (97.264 − 29.352) ÷ 97.264 = 69.82 % P: (16.298 − 4.918) ÷ 16.298 = 69.82 % The basins alone (Individual routing): N leaving 16.593 + 8.925 = 25.518 kg/yr → 73.76 %
File 2 works the same way. Catchment 2's depth is (0.5 + 6.5) × 12 ÷ 40 = 2.10 in, which gives 99.08 % on its own load: 0.447 kg/yr of N leaves from it. The 16.593 kg/yr from Catchment 1 again passes through untreated. The outlet gets 17.041 kg/yr, or 82.48 %, against 82.89 % for the basins alone.
4Why the series comes out lower
- Basin 2 is charged for the upstream water. Treatment Depth Allocation spreads Basin 2's volume over Catchment 1's area as well. That lowers Basin 2's depth on its own runoff from 0.48 in to 0.39 in, and its removal from 81.65 % to 73.76 %.
- Basin 2 gets no credit for that water. None of the 16.593 kg/yr arriving from Catchment 1 is removed at Basin 2.
- So connecting the basins can only lower the result. The space Basin 2 is assumed to give to Catchment 1's water treats nothing. In file 2 this is plain to see: Basin 2 holds 3.9 in of its own catchment's runoff, far more than it needs, yet all of Catchment 1's overflow leaves the site.
5A volume balance (proposed)
A retention basin captures the first part of each storm's runoff, up to its volume. With equal EMCs, the mass removed is in proportion to the volume captured. Take one storm with R inches of runoff from each 20-acre catchment, and basin depths d1 = 0.30 in and d2 = 0.48 in, each over its own catchment.
-
Basin 1 captures up to its depth and overflows the rest to Basin 2:
captured = min(R, d1) overflow = max(0, R − d1) -
Basin 2 receives its own runoff plus the overflow, and captures up to its depth:
captured = min(d2, R + max(0, R − d1)) ≥ min(d2, R), what it captures aloneThis is why a series cannot remove less than the same basins standing alone.
-
The two basins together. When d1 ≤ d2, as here, add the two captures (in inches over 20 acres):
R ≤ d1: R + R = 2R (everything) R > d1: d1 + min(d2, 2R − d1) = min(d1 + d2, 2R) Both cases: min(d1 + d2, 2R)That is exactly what one basin of 1.3 ac-ft on 40 acres captures. So the annual removal is the retention curve at the combined treatment depth, applied to all the load reaching Basin 2:
File 1: TD = 1.3 × 12 ÷ 40 = 0.39 in → 73.76 % (N leaving 97.264 × 0.2624 = 25.518 kg/yr) File 2: TD = 7.0 × 12 ÷ 40 = 2.10 in → 99.08 % (N leaving 97.264 × 0.0092 = 0.895 kg/yr) -
When the upstream basin is the deeper one (d1 > d2), Basin 1 overflows only in storms larger than d1. By then Basin 2 is already full with its own runoff, so the overflow passes on untreated. The series then removes exactly what the basins remove alone.
The rule proposed for BMPFast Plus. At each catchment, compare its basin's depth with the depth of the system above it:
- Its basin is at least as deep as the system above: the basins so far act as one basin. The treatment depth is the volume so far × 12 ÷ the area so far. The retention curve at that depth, with the area-weighted non-DCIA CN and DCIA, removes all the load that has entered.
- The system above is deeper: the upstream overflow passes untreated, and the catchment's own load is removed at its own depth.
This rule is exact for two catchments with the same runoff in each storm. Its form for longer strands, and for catchments with different curve numbers, is part of the review.
In file 1 the volume balance gives the same 73.76 % as the basins alone. The tables are interpolated in a straight line between their 0.25 in and 0.50 in rows, and 0.39 in is the average of 0.30 and 0.48 in, so sharing Basin 2's spare volume gains nothing measurable here. File 2's Basin 2 has a great deal of spare volume, and there the gain is large.
6Every method on the two files
The three methods BMPFast Plus offers today under Retention basins in series, the volume balance, and the basins alone. All are calculated by BMPFast Plus's own code, except the volume balance, which is the retention table read at the combined depth. Removal is shown at the outlet, the same for N and P.
| Method | File 1: removal | File 1: leaving N / P (kg/yr) | File 2: removal | File 2: leaving N / P (kg/yr) |
|---|---|---|---|---|
| Basins alone (Individual routing) | 73.76 % | 25.518 / 4.276 | 82.89 % | 16.642 / 2.789 |
| Treatment Depth Allocation (today's default) | 69.82 % | 29.352 / 4.918 | 82.48 % | 17.041 / 2.855 |
| Volume balance (proposed) | 73.76 % | 25.518 / 4.276 | 99.08 % | 0.895 / 0.150 |
| Treatment Depths added (BMPFast v1) | 92.13 % | 7.653 / 1.282 | 99.90 % | 0.097 / 0.016 |
| Efficiencies in series (Equation 9-5) | 87.69 % | 11.970 / 2.006 | 99.93 % | 0.065 / 0.011 |
- Treatment Depths added reads the curves at 0.30 + 0.48 = 0.78 in over all 40 acres. That is as if the site had 2.6 ac-ft of retention rather than 1.3, so it overstates the removal.
- Efficiencies in series (Equation 9-5 of the Applicant's Handbook, Vol. I) multiplies the fractions passing each BMP. It assumes the BMPs remove by independent mechanisms. Two retention basins capture parts of the same storm volume, so for them it overstates the removal too.
7What happens next
- The review team decides whether the volume balance replaces Treatment Depth Allocation as the default. Treatment Depth Allocation would stay a choice on the routing.
- If it is accepted, it goes into the next BMPFast Plus release candidate, tested on these two files. The test checks that a dry retention series never removes less than the same basins alone. The help pages below will be updated to match.
- Until then, for dry retention basins in series, the Individual routing's removal is a floor. The volume balance shows the series cannot do worse.
8Try it, and read more
- dry-retention-in-series.bmpf: file 1 (Basin 2 is 0.8 ac-ft). Save it, then in BMPFast Plus choose File ▸ Open. Select Compare routings under Catchment Routings to see the two routings side by side. Select Routing in series and change Retention basins in series on its Entry tab to compare the methods; its Report tab shows each catchment's treatment depth and removal.
- dry-retention-in-series-large-basin-2.bmpf: file 2 (Basin 2 is 6.5 ac-ft). It also has a 3.34 ac-ft basin in Catchment 1 (2 in; 99.00 % alone) for comparison.
- Strand Based Routing for Retention in Series (help: Treatment Depth Allocation, worked by hand)
- Comparison of Methods for Calculating Retention BMPs in Series (help)
- Routing (help)
- Retention Efficiency Lookup Tables (help) and the retention curves
- All questions answered
Calculated with the calculation code of BMPFast Plus 2.0.4 (BMPFast version 2), November 11, 2026. The two project files, opened and calculated, give the numbers on this page. Questions: info@bmpfast.com.