| Saul and Thornton [53] and Gupta and Saul [10] |
2 |
CSS at Great Harwood and Clayton-le-Moors (Northwest of England). |
84
(47–121) |
36 and 31 |
– |
– |
– |
TSS, COD, BOD5, NH4-N, VSS |
The maximum rainfall intensity, maximum inflow, rainfall duration, and the antecedent dry weather period were identified as the most important parameters affecting the occurrence of FF.
Gupta and Saul established empirical multi-regression relationships between the pollutant mass transported in the FF and some characteristics of the rainfall event.
|
| Saget et al. [54] and Bertrand-Krajewski et al. [16] |
12 |
12 separate and combined SS (Stuttgart-Busnau and Munchen-Pullach, Germany). |
220.1
(25.6–1145) |
197 |
5.54
(0.2–79.6) |
– |
– |
TSS, COD, BOD5, TOC |
Studying FFC values shows significant variation from one event to another. This indicates that the curves from one catchment cannot be replaced by a single average curve without losing information.
The values of FFC are lower for separate sewer systems than for CSSs.
The characteristics of the mass-volume rate curves depend on different factors, including the type of pollutant, the site, rainfall event, and the overall operation of the sewer system.
No clear and general linear multi-regression relationships can be established to explain their shape and their variability. This is probably due to the complexity of the phenomena and the multiplicity of influencing factors and parameters.
|
| Lee and Bang [49] |
9 |
Taejon and Chongju, South Korea. |
218.31
(1.4–650) |
34 |
– |
– |
– |
BOD5, COD, SS, TKN, NO3-N, PO4–P, TP, n-Hexane
extracts,
Pb, and Fe |
In watersheds with areas smaller than 100 ha, where impervious surfaces exceeded 80%, the peak pollutant concentration occurred before the flow peak. Conversely, in watersheds larger than 100 ha, where impervious surfaces were less than 50%, the flow peak followed the pollutant concentration peak. |
| Lee et al. [39] |
13 |
Chongju, South Korea. |
34.47
(0.74–190) |
38 |
11.0
(2.3–33.1) |
4
(1.6–11.9) |
6.5
(1–13) |
COD,
SS,
TKN, PO4–P,
TP, HEM, Pb,
and Fe |
The FF magnitude was greater for some pollutants (e.g., SS from residential areas) and less for others (e.g., COD from industrial areas).
No correlation was observed between FF and the antecedent dry weather period, while the former was greater for smaller watershed areas.
|
| Ma et al. [36] |
9 |
Southern California. |
1.47
(0.39–4.81) |
52 |
26.51
(0.2–156) |
– |
15.55
(0–108) |
COD, TOC, Oil and Gas, and TKN |
Pollutants representing organic contaminants had the highest MFF ratios.
The range of the MFF ratios for organic pollutants (COD, TOC, Oil and Gas, and TKN) ranged from 0.7 to 4.5, while the median for all parameters was greater than 1.5 for both MFF10 and MFF20.
The occurrence of FF for small watersheds on BMP removal efficiency and design has yet to be fully explored.
|
| Lee et al. [55] |
Many sites |
Various datasets (California, Central Chile, the Mediterranean Basin, Southwest and South Australia, and South Africa). |
(0.0000464–14,771) |
6500 |
– |
– |
– |
COD,
SC, TOC,
TSS,
Al, Cu, Pb, Ni, and Zn |
A seasonal FF existed for most cases and was strongest for organics, minerals, and heavy metals, except Pb. It suggests that applying BMPs early in the season could remove several times more pollutant mass compared to randomly timed or uniformly applied BMPs. |
| Soller et al. [56] |
25 |
The Guadalupe River and Coyote Creek in the City of San Jose (California, USA). |
– |
8 |
– |
– |
– |
Total and dissolved metals, pesticides, polyaromatic hydrocarbons, anions, TSS, total organic
carbon, conductivity, gasoline and diesel, and volatile and semi-volatile organics |
There was no relation between metal concentrations in storm runoff and land use.
Sulfate showed a relation between stormwater concentrations and agricultural land use within the catchment.
Urban runoff dissolved metal concentrations do not have a strong relationship with storm size. However,
it seems to have a relation with the antecedent dry weather period.
|
| Kim et al. [57] and Kang et al. [37] |
1 |
A rectangular-shaped highway in Los Angeles, near the UCLA campus (California, USA). |
0.39 |
41 |
17.8
(3–56.4) |
8.43
(1.08–19.38) |
16.6
(1–69.4) |
COD, Conductivity, Zn, Cu |
The EMCs (event mean concentrations) are negatively correlated to storm duration, total rainfall, total runoff volume, and average rainfall intensity. Large storms have smaller EMCs because of dilution effects or exhaustion of pollutant mass. |
| Barco et al. [11] |
1 |
Cascina Scala urban catchment in North Pavia (Lombardia, Italy). |
12.7 |
23 |
15.93
(2–39.8) |
5.39
(0.18–18.88) |
6.10
(0–29.9) |
BOD5, COD, SS, SetS, TP, TN, NH3–N, Pb,
Zn, SC, and HC |
The magnitude of the first flush was large with 40% of the masses, on average, contained in the first 20% of the runoff (MFF20 = 2.0).
The FF occurrence presents opportunities to select BMPs that favor treatment of the early runoff when the pollutant concentrations are the highest.
|
| Zushi and Masunaga [58] |
1 |
Hayabuchi River, i.e., a tributary of the Tsurumi River in Yokohama, Japan. |
2460 |
2 |
– |
– |
– |
pH, EC, short- to medium-chain-length PFCAs |
It was found that large loads of long-chain-length PFCAs are discharged into the Hayabuchi River during FF after the rainfall event. |
| Obermann et al. [59] |
1 |
Vène River, French Mediterranean coast. |
6700 |
2 |
115.2
(50.6–190.8) |
2.55
(1.32–3.42) |
10.5
(1–27) |
TSS, VSS, TP, SRP, TN, org-N, nitrate, NOx–N, and
NH4–N |
The most important FF could be detected for NH4–N (FF25 = 0.79), followed by TSS (FF25 = 0.72) and VSS (FF25 = 0.70). |
| Bach et al. [27] |
7 |
Melbourne, Australia. |
65.3
(0.05–186) |
281 |
– |
– |
– |
TSS, TN and E. coli |
This study demonstrates that specific catchment characteristics (e.g., age, septic cross-connections, etc.) appear to influence FF volumes and strength. |
| Athanasiadis et al. [48] |
1 |
The building of the Academy of Fine Arts in the centre of Munich, Germany. |
0.51 |
33 |
812.6
(472.5–1342.5) |
– |
2.36
(0–15) |
Cu |
No correlation was found between the FF effect and weather parameters, such as rain height, rain intensity, and antecedent dry weather period. |
| Perera et al. [60] |
6 |
Four residential catchments, namely, Coomera, Alextown, Birdlife and Gumbeel and two completely impervious surfaces, the international apron of the Brisbane airport and DFO Shopping Centre car park. |
2.165
(0.036–6.208) |
61 |
– |
– |
– |
SS |
The maximum rainfall intensity is the most influential variable.
For a relatively small rainfall event (<5 mm), an optimum value of the antecedent dry period exists that maximizes the EMC.
The results showed that the rainfall intensity and depth are more important in estimating EMCs, and small changes to these variables can change the EMCs significantly.
The percentage of impervious area surfaces also influences EMC. Therefore, it can be concluded that the total area and the surface characteristics of the catchment substantially influence EMC.
|
| Perera et al. [61] |
9 |
Queensland, Australia. |
3.37
(0.0212–8.6) |
39 |
75.7
(0.8–492) |
3.54
(0.07–13.07) |
8.69
(0.125–37.5) |
SS |
The Monte Carlo simulation revealed that most commonly, the FF runoff varies over the initial 30–50% of the runoff volume. |
| Li et al. [22] |
3 |
Nanning City, China. |
22.63
(14.6–30.5) |
6 |
119.2
(10–210) |
1.98
(0.17–3.5) |
2.97
(0.46–10) |
COD, NH3-N, TN, TP and TSS |
The EMC values reveal that drainage outlets inappropriately connected with sewage are 2–4 times higher than those of stormwater outlets (especially for NH3–N, TN, and TP), having pollution levels similar to CSOs.
COD and TSS have a stronger FF effect than other indicators.
The discharge pollution load is primarily caused by the inside of the sewer through sewer sediment erosion (more than 60%, with heavy rainfalls).
|
