1
METHODOLOGY TO QUANTIFY CLOGGING COEFFICIENTS FOR
GRATED INLETS. APPLICATION TO SANT MARTI CATCHMENT
(BARCELONA)
Manuel Gómez1, Joan Pa és1, Beniamino Russo2, Edua do Ma ínez-Goma iz1
1 Flumen Resea ch Ins i u e. Technical Uni e si y o Ca alonia. Spain
2 G oup o Hyd aulic and En i onmen al Enginee ing. Technical School o La Almunia
(EUPLA). Uni e si y o Za agoza. Spain
ABSTRACT:
Wi hin he d ainage sys em o a ci y, he se o inle s is in cha ge o aking he uno
p oduced by local s o ms o he s o mwa e /sewe . In he d ainage sys em design he
selec ion o app op ia e inle models and hei loca ion is one o he undamen al aspec s.
The hyd aulics o hese inle s has ecei ed g ea a en ion wi hin he las yea s, howe e ew
inle make s p o ide he hyd aulic capaci y o hei p oduc s. In addi ion, hese da a usually
conside clean wa e , while in eali y, nume ous inle s can be ei he o ally o pa ially
clogged. This aspec should be kep in mind wi hin he design p ocess. In his pape , a
me hodology o conside he hyd aulic e ec s o clogging phenomena is p esen ed. The
wo k s a ed om a isual inspec ion o he g a ed inle s h oughou he u ban ca chmen o
San Ma í (Ba celona), as a means o iden i ying clogging pa e ns, hei epe i i e o ms
and hei associa ed equency. A e ha , clogged pa e ns we e ep oduced in labo a o y
es ing o ypical inle s ypes, he eby ob aining he eal quan i y o wa e ha could be
cap u ed by each o hem. I was shown ha he same exp ession employed o desc ibe he
e iciency o clean inle s can be used o assess he e iciency o hose clogged. A educ ion
ac o in e ms o hyd aulic capaci y and ela ed o each clogging pa e n has been de ined
o use in hyd aulic s udies o uno along s ee s. Finally, he pape compa es he ob ained
esul s in e ms o clogging coe icien wi h ano he expe imen al campaign ca ied ou in
o he ca chmen o he ci y.
KEYWORDS: Inle , clogging, su ace wa e , ca chmen
2
INTRODUCTION
S ee s cons i u e a su ace hyd aulic ne wo k du ing ain e en s when uno lows along
he s ee s. Flow on he s ee s in a ainy day depends on se e al ac o s like ain all
in ensi y, u ban a ea, slopes, s ee c oss-sec ions and inally he ype and numbe o inle s.
S ee s ha e an impo an ole associa ed o s o mwa e managemen and any s ee mus
ensu e sa e y o pedes ians (Gómez e al., 2011; Russo e al., 2013; Ma ínez e al., 2016)
and ehicula a ic acco ding o a le el o se ice associa ed o a e u n pe iod (UDFCD,
2016; Ma ínez e al., 2017 and 2018).
D ainage sys em can be seen as composed by wo sub-sys ems: he unde g ound wi h
s o mwa e /sewe sec ions and he su ace one, o med by s ee s. This concep o dual
d ainage sys em is he bes desc ip ion o wha eally occu s du ing a ain e en . Bu hese
wo sub-sys ems a e no independen . They a e connec ed by he so-called “inle s sys em”
composed by all he inle s, g a es, gullies, e c. placed in he ci y aiming o cap u e he uno .
A he same ime, in case o p essu e so mwa e /sewe low, hey beha e as a elie al e,
by allowing he excess o wa e no accep ed by he s o mwa e /sewe condui o eme ge o
he s ee le el. Hund eds o housands o inle s a e esponsible o link bo h sys ems whose
p ima y unc ion is o ake he uno o he s o mwa e /sewe s up o he igh place (A ina
S. e al., 2001; CDOT, 2000) and onwa d o ea men .
Nowadays, a la ge a ie y o inle s and g a es exis s in he ma ke , di e en om coun ies
and e en om ci ies, whe e local adap a ion o s ee s design and me eo ological condi ions
has p oduced such di e en shapes, dimensions and, as a consequence, di e en models.
Many imes, aes he ic concep s in luence inle design as in many o he disciplines, hough
un o una ely se e al ci ies decide he inle selec ion pu ing i s hese “new” c i e ia wi h no
conside a ion o hei hyd aulic capaci y o he p oblems associa ed o hei no mal ope a ion
wi h clogging. Mo eo e , in se e al coun ies i is possible o ind ha mos o local supplie s
p o ide no da a conce ning he hyd aulic e iciency o hei p oduc s. Al hough se e al
p ocedu es like HEC-22 p o ide in o ma ion abou how o assess he wa e collec ed, hey
a e es ic ed o some inle shapes ha a e no ound all o e he wo ld (B own e al, 1996).
Recen s udies p opose some p ocedu es o cha ac e ize inle hyd aulic pe o mance. Mos
o imes hey a e based on eal scale expe imen al analysis and, as a esul , allow he
es ima ion o cap u ed low depending on s ee geome y, inle ype and gu e low
condi ions
(CCRFCD,
1999; Gómez e al., 2008; Gómez and Russo, 2011; Russo e al.,
2013). In hese s udies, labo a o y es s a e usually done in bes condi ions (i.e. clean wa e ,
inle wi h no clogged a ea, no obs acles, e c.) so, he ob ained esul s exp essed in e ms o
inle hyd aulic e iciency can be conside ed as an uppe limi ha can be un ealis ic in case
o inle clogging. In ac , in eal condi ions, dus , di , lea es o deb is, can educe he a ea
o g a ed inle holes, he eby educing he amoun o wa e o be cap u ed. In o de o ake
in o accoun he clogging phenomenon, Guo (2006) p oposed he concep o clogging
coe icien , which educes he hyd aulic e iciency wi h a ac o be ween 0 and 1 (Guo, 2006),
al hough he p oposed alues we e no ela ed o heo e ical o expe imen al s udies.
Clogging phenomenon depends on inle design, ype and densi y o ege a ion, ain all
pa e ns, s ee slope, cleaning equency, e c. Al hough he ep oduc ion o he e ec s o
his phenomenon is complex, labo a o y and ield s udies could allow a easonable
es ima ion o he po en ial educ ion in e ms o hyd aulic e iciency due o clogging.
So inle s es ing unde a ious clogging condi ions should be equi ed, being
cha ac e iza ion o clogging beha io and i s se e i y an impo an p e- equisi e o a co ec
design o su ace d ainage sys ems. Howe e , i migh be wonde ed whe he inding a
cha ac e is ic way o he inle o be clogged is possible. In p e ious s udies, Gómez e al.
(2012) in oduced his idea h ough he “clogging pa e n” concep whe e, depending on he
ype o inle , some cha ac e is ic o ms whe e holes a e co e ed by lea es, deb is, e c. we e
3
obse ed. The s udy was de eloped in an u ban ca chmen o Ba celona (Rie a Blanca)
including s eep and la slopes. Howe e , one doub could appea : a e hose clogging
pa e ns ep esen a i e o he whole ci y? In o de o answe his ques ion, he au ho s o
his wo k selec ed ano he ca chmen in Ba celona wi h di e en cha ac e is ics, close o
he beaches, la slopes, and po en ially mo e p one o be a ec ed by clogging, and maybe
o de elop di e en pa e ns.
As s a ed be o e, clogging e ec s a e speci ic o local clima e condi ions, land use, s ee
geome y and inle models. The me hodology de eloped could be applied wi h no loss o
gene ali y anywhe e hough. In his wo k, he ollowing ea u es we e conside ed:
Field isi s in d y and we wea he condi ions, o obse e h oughou he ca chmen
be o e and a e ain e en s he way inle s become clogged
Compa ison o he clogged inle s o iden i y common aspec s in o de o de ine
clogging pa e ns, and assess he equency associa ed o each one
Quan i ica ion o he educ ion o hyd aulic capaci y associa ed o each one o he
clogging pa e ns, h ough expe imen al s udies ca ied ou in he hyd aulic labo a o y
o he Technical Uni e si y o Ca alonia ( ollowing UPC labo a o y)
P oposal o clogging coe icien s a he end o he p ocess, o be used o each
pa e n, and make i ex ensi e o o he case s udies.
STORMWATER CAPTURED BY INLETS
An excess o wa e o e he pa emen can gene a e sa e y p oblems o pedes ians and
ehicles. Some imes i is esponsible o acciden s due o aquaplaning (i.e. loss o adhe ence
be ween i es and pa emen ) and loss o s abili y o pedes ians and ehicles. Many ci ies
ha e ini ia ed s udies o ien ed o quan i y he s ee hyd aulics and p ocedu es o de e mine
bes loca ion o inle s.
Hyd aulic capaci y o a single inle depends on di e en pa ame e s: s ee geome y, s ee
low and ype o inle (Gómez and Russo, 2011; Russo e al., 2013)). The low cap u ed by
an inle could be high o low gene ally speaking, al hough i depends essen ially on he
compa ison wi h he s ee low. Absolu e low cap u ed by he inle some imes is no enough
o ha e a clea idea o he hyd aulic pe o mance o an inle . Tha is why he concep o
hyd aulic e iciency (E) is de ined as he a io be ween he cap u ed low (Qc) o e he low
along he s ee and app oaching he inle (Q). This pa ame e p o ides mo e objec i e
in o ma ion abou he pe o mance o a single inle espec o he s ee low. E iciency (E)
anges om 0 o 1 and i is a dimensionless pa ame e .
𝐸 =𝑄𝑐
𝑄
(1)
As p e iously s a ed, es s o achie ing inle hyd aulic e iciency a e usually conduc ed
unde he bes condi ions wi h clea wa e and no deb is. Reali y wi hin a ci y is no like his,
no ma e he e o s om he municipal cleaning se ices. Al hough in ense s ee washing
and sweeping is ca ied ou , a ce ain laye o dus and o he deb is like lea es om ees
will be buil up. Du ing a ain e en , he u ban uno , he mos e ec i e cleaning p ocess in
a ci y, sweeps away mos o hese deb is and once he inle is eached wa e en e s in o he
s o mwa e /sewe and deb is co e s pa ially o o ally he inle . Tha is why in he u ban
s ee design, a kind o secu i y coe icien educing he e iciency should be conside ed.
In o de o assess he coe icien s o a ec he hyd aulic e iciency o he inle s, a ield s udy
in an u ban ca chmen o he ci y o Ba celona was de eloped. Visual inspec ions and
4
pic u es o e e y selec ed inle we e collec ed (Figu e 1). A comp ehensi e analysis o he
clogged inle s was ca ied ou , by de e mining se e al ypes o clogging pa e ns, hei
associa ed equency and hei spa ial dis ibu ion. Once he ield s udy was inished, a
quan i ica ion o he e iciency educ ion was de e mined h ough labo a o y es s, and
educ ion coe icien s o hyd aulic e iciency we e p oposed.
a)
b)
c)
Figu e 1. a) Clean and b,c) clogged inle s
TEST CASE: SANT MARTI CATCHMENT IN BARCELONA
The u ban ca chmen in he dis ic o San Ma i in Ba celona, wi h an a ea o 6.49 km2 was
selec ed as es case o implemen he me hodology (Figu e 2). The ca chmen is loca ed in
he lowe pa o he ci y, nex o he Medi e anean Sea wi h ele a ion om 1 o 18 m. The
easons (Pa és, 2014) o selec his ca chmen a e:
Mos o he ca chmen is loca ed on eclaimed land, ini ially below sea le el. G a els
and sand deposi s a e he usual geology in he a ea. Due o his, la mo phology is
he no mal p o ile wi h slopes om 0.1 o 5% and an a e age o 0.5%, hus he inle s
a e mo e p one o be clogged. Beaches a e close o he a ea so sands ca ied by he
wind can be ound all o e he a ea o in e es
A ailable GIS in o ma ion including posi ion and ype o e e y inle loca ed in he
ca chmen (Figu e 2). This in o ma ion was p o ided by o me CLABSA, now
BCASA, he public company in cha ge o he managemen o he ci y sewe sys em
The inle sys em is composed by 6946 inle s, wi h 6874 inle s ha can be g ouped in
98 di e en models, while 72 a e non-classi ied models. Mo eo e 7 models a e
enough o desc ibe 80% o all he inle s.
Fo all hese easons, San Ma í ca chmen could be seen as an example o he wo s case
ega ding inle s clogging, due o i s mo phological clima ic condi ions (Medi e anean
clima e cha ac e ized by sho and in ense ain alls).
The i s pa o he s udy included he di ec isual inspec ion o e i y he da abase o inle s,
acco ding o GIS in o ma ion, classi ied by he UTM coo dina es and inle ype. Acco ding
o he exis ing ypes wi hin he da abase, 7 models we e selec ed as he mos ep esen a i e
ones. These models a e E5, R-121, Impu, Me idiana, Del a/Ba celona, Teide and
Onda/Eb o and, as p e iously men ioned, hey ep esen 80% o all he inle s in he
ca chmen . In able 1 inle dis ibu ion o e e y model is shown.
5
Figu e 2. Gene al iew o he basin and inle s. Inle s a e indica ed wi h do s.
Table 1. Numbe o he selec ed inle models in he ca chmen
Inle model
To al
% o inle s in
he basin
Inle nº 1: E5
257
3,70%
Inle nº 2: R-121
425
6,12%
Inle nº 3: IMPU
616
8,87%
Inle nº 4: Me idiana
1162
16,73%
Inle nº 5: Del a/Ba celona
2226
32,05%
Inle nº 6: Teide
233
3,35%
Inle nº 7: Onda/Eb o
620
8,93%
To al
5539
79,74%
To al S . Ma í ca chmen
6946
100%
F om he 5539 inle s, a g oup o 1300 inle s was selec ed. The selec ion c i e ia we e a
uni o m spa ial dis ibu ion o he inle s along he ca chmen and a minimum o 15% o e e y
6
model. One o he objec i es is o check whe he he clogging pa e ns a e uni o mly
dis ibu ed along he ca chmen o no .
Each o he 1300 inle s was isually inspec ed, h ee imes as a minimum, and a pic u e o
each one was aken o eco d he eal condi ion o he inle . In o de o ake in o accoun he
po en ial e ec o soil e osion due o ain all in he clogging phenomenon, wo we ield
obse a ions we e planned. These wo isi s we e conduc ed wi hin he days a e a ain
e en (1 o 3 maximum) du ing he ainy season in Ba celona (au umn), while he las hi d
isi was ca ied ou unde d y wea he condi ions (in sp ing), wi h a minimum o 2 weeks
wi h no ain all, o check i he clogging pa e ns a e modi ied by he ac i i y o he municipal
cleaning se ices.
CLOGGING PATTERNS
A he end o he ieldwo k, an ex ensi e g aphical da ase was pe o med and employed o
he analysis o he inle s. Due o he small di e ences, clogging s a es unde d y and we
wea he condi ions we e combined o simpli y he app oach. The analysis was unde aken
s aigh ly in he PC sc een, no conside ing ye compu e ision p ocesses ha could be
employed o accele a e he p ocess. Howe e , he pe cen age o holes co e ed by lea es
o deb is (i.e. clogging elemen s) is possible o be dis inguished as a gene al alue. The
c i e ia indica ed in he able 2 we e adop ed. Due o hese c i e ia, speci ic clogging pa e ns
we e de ined depending on he a io be ween he clogged a ea (Ac) and he o al a ea o
holes in he inle (Ah).
Table 2. P oposed c i e ia o classi y clogging pa e ns ollowing he ield inspec ion
No hing (clean inle )
Ac/Ah< 10%
Pa e n C1
10 % ≤ Ac/Ah< 25 %
Pa e n C2
25 % ≤ Ac/Ah< 50 %
Pa e n C3
50 % ≤ Ac/Ah< 80 %
Inle ou o o de
Ac/Ah ≥80 %
Fo wo di e en inle s o same model and clogged a ea, likely he clogging does no always
ha e he same shape. To simpli y he app oach, some clogging shapes ela ed o e e y
pa e n ha e been p oposed (i.e. desc ibing he mos gene al manne he inle is clogged),
acco ding o he ieldwo k obse a ions. Shapes associa ed o he di e en clogging pa e ns
we e mo e easily ecognized o some inle s (i.e. Eb o, Teide o IMPU), while o o he inle s
(E5 and 4 Me idiana) ew o no clogging p oblems we e obse ed, hus no pa e ns we e
p oposed o hose inle s. Clogging pa e n C1 is associa ed o he co e age o na ow
spaces, many imes nex o he ex e nal pe ime e o he inle . Some o he inle s show a
mo e in ense clogging e ec , in case ha cleaning p ocess is no enough, hus a C2 pa e n
is p oposed. Di e ences be ween C1 and C2 pa e ns a e mo e in ense o inle s 3, 5, 6 and
7 (i.e. IMPU, Ba celona, Teide and Eb o).
Finally, i he clogging p ocess p og esses, ano he pa e n, called C3, can be obse ed,
whe e a signi ican a ea o holes is co e ed. I was obse ed ha no all he inle s each C3
pa e n, while inle s 1, 2, and 4 (E5, R-121 and Me idiana) a e hen less suscep ible o
de elop his pa e n. In Figu e 3 he p oposed shapes o e e y clogging pa e n o he
selec ed 7 inle models a e p esen ed.
The inle s wi h highe suscep ibili y o be clogged a e inle s Eb o, Teide and IMPU, maybe
due ha hey a e old models, placed a ound 30 yea s ago. Inle s E5, Me idiana and
Ba celona ha e been ins alled mo e ecen ly, wi hin he las 5 o 10 yea s, and show o ha e
be e condi ions o a oid clogging e ec s. This di e ence could be explained by poo
main enance o inadequa e cleaning o hese g a ed inle s.
7
Figu e 3. Clogging pa e ns o selec ed models
Inle
C1 Clogging pa e ns
C2 Clogging pa e ns
C3 Clogging pa e ns
Id.
Name
Ske ch
Pic u e
Ske ch
Pic u e
Ske ch
Pic u e
1
E5
2
R-121
3
IMPU
4
Me idiana
5
Ba celona
6
Teide
7
Eb o
8
FREQUENCIES AND DISTRIBUTION OF CLOGGING PARTNERS
F equencies and spa ial dis ibu ion o clogging pa e ns we e assessed unde d y and we
wea he condi ions. F equency dis ibu ion can be conside ed qui e simila o bo h
si ua ions, hence he ela ed alues in all cases we e combined and only a single equency
was p esen ed. The e o e, i was obse ed ha clogging phenomenon occu s h oughou
he yea in we and ainy season, and i seems ha pa e ns do no depend on empo ali y.
Second ele an aspec is ha a e iewing he spa ial dis ibu ion o clogging pa e ns,
hose appea all o e he ca chmen . No signi ican di e ences be ween highe o lowe
slope a eas we e obse ed, hus i may indica e ha acco ding o he obse ed da a,
clogging pa e ns show a qui e uni o m spa ial dis ibu ion h oughou he ca chmen . Those
di e ences could be esponsible o he ime needed o mo e om one pa e n o ano he .
This aspec has no been e i ied hough.
The mo e common pa e ns a e C1 and C2, wi h equencies be ween 20 and 70%, while
pa e n C3 is less equen , wi h equencies om 2 o 9%. The equency o pa e ns C1
and C2 is app oxima i ely he same o all he inle s, excep o inle Eb o, whe e C1 pa e n
is mo e usual (75%) han C2 (15%).
In addi ion, an in e es ing issue is o compa e he numbe o inle s ha show some clogging
p oblem. This could be exp essed by combining he equencies associa ed o C1, C2 and
C3 pa e ns. The accumula ed equency (C1+C2+C2) is app oxima ely 95% o Eb o, 89%
o Teide, 60% o IMPU, 38% o R-121, 10% o Ba celona and 0% o Me idiana and E5
(Table 3).
Table 3. F equencies associa ed o each one o he obse ed clogging pa e ns
Inle model
Pa e n
C1
Pa e n
C2
Pa e n
C3
Pa e n
C1+C2+C3
Ou o
o de
To al
E5
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
R-121
18.37%
17.35%
0.00%
37.76%
3.06%
40.82%
IMPU
22.37%
18.42%
9.21%
60.53%
1.32%
61.84%
Me idiana
0.00%
0.00%
0.00%
0.00%
1.50%
1.50%
Ba celona
0.00%
7.48%
2.55%
10.04%
0.73%
10.77%
Teide
40.00%
49.09%
0.00%
89.09%
0.00%
89.09%
Eb o
75.54%
15.11%
4.32%
94.97%
1.44%
96.41%
HYDRAULIC ANALYSIS OF CLEAN AND CLOGGED INLETS
So a , a desc ip ion o he clogging p oblems in he Ba celona’s ca chmen o San Ma í
has been p esen ed. No wi hs anding, one o he objec i es o his wo k is he quan i ica ion
o he educ ion o he hyd aulic e iciency due o clogging p oblems wi hin inle s.
In p e ious s udies de eloped a he UPC (Technical Uni e si y o Ca alonia –
Ba celonaTech) (Ma ínez, P., 2000; Gómez and Russo, 2011), i was p oposed ha
hyd aulic inle e iciency (E) can be ela ed o he app oach low (Q) and he immedia ely
ups eam wa e le el (y) as a po en ial unc ion as ollows:
𝐸 =𝐴∙(𝑄
𝑦)−𝐵
(2)
whe e:
E: hyd aulic e iciency o he inle
9
Q: App oaching low o he inle (l/s)
y: wa e le el immedia ely ups eam he inle (mm)
A, B: wo speci ic coe icien s o e e y inle
This exp ession was p oposed a e conduc ing labo a o y es s wi h clean wa e and no
clogging e ec s. Thus, he issue o be sol ed is whe he his exp ession is s ill alid o
clogged inle s, o ano he app oach o o mula is needed o desc ibe he hyd aulic e iciency
o a clogged inle .
Two inle s o he 7 selec ed ones we e es ed in he UPC labo a o y, and i was p o ed ha
he same po en ial unc ion p oposed o clean wa e s could be used o clogged inle s, jus
modi ying he A and B coe icien s depending on he clogged a ea.
The same es p o ocol used o clean wa e condi ions was adop ed (Gómez and Russo,
2011), wi h ci cula ing lows o 25, 50, 100, 150 and 200 l/s, longi udinal slopes o 0, 0.5, 1,
2, 4, 6, 8 and 10 % and ans e se slopes o 0, 1, 2, 3 and 4%.
Tes s we e de eloped by co e ing wi h gypsum pa o he holes o he inle , ep oducing
he pa e n o be s udied, as can be seen in Figu e 4 o Ba celona inle . Expe imen al da a
a e p esen ed in Figu e 5, and om he analysis o hese esul s i is possible o conclude
ha a po en ial law is good enough o ep esen he hyd aulic capaci y o he clogged inle s.
Pa ame e s A and B ha e a ied acco ding o clogged a ea (Table 4), al hough qui e high
co ela ion coe icien s we e ob ained o he wo inle s (i.e. 91.5% o Ba celona inle
(Pa e n C3), 77.2% o Teide inle (Pa e n C1), and 75.9% o Teide inle (Pa e n C2)).
This con i ms he applicabili y o he e i ied po en ial law o clean inle s o clogged inle s
oo.
Figu e 4. Rep oduc ion o C3 clogging pa e n o he Ba celona inle , and iews o he
expe imen al se -up
Table 4. Pa ame e s A, B de e mined a e labo a o y es s
TEIDE
Inle clean
Pa e n C1
Pa e n C2
Pa e n C3
A
0,425
0,340
0,327
-
B
0,714
0,800
0,790
-
BARCELONA
Inle clean
Pa e n C1
Pa e n C2
Pa e n C3
A
0,355
-
-
0,230
B
0,850
-
-
1,003
