Jou nal o En i onmen al Managemen 345 (2023) 118510
A ailable online 28 June 2023
0301-4797/© 2023 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY-NC license (h p://c ea i ecommons.o g/licenses/by-
nc/4.0/).
Resea ch a icle
T ea ed and highly dilu ed, bu was ewa e s ill impac s di e si y and
ene gy luxes o eshwa e ood webs
Ioa de Guzman
a
,
*
, A u o Elosegi
a
, Daniel on Schille
b
, Jose M. Gonz´
alez
c
, Lau a E. Paz
d
,
e
,
Benoi Gauzens
,
g
, Ul ich B ose
,
g
, Al a o An ´
on
h
, Nu ia Ola e
h
, Jos´
e M. Mon oya
i
,
Ai o La a˜
naga
a
a
Depa men o Plan Biology and Ecology, Facul y o Science and Technology, Uni e si y o he Basque Coun y (UPV/EHU), Ba io Sa iena S/n, 48940, Leioa, Spain
b
Depa men o E olu iona y Biology, Ecology and En i onmen al Sciences, Uni e si y o Ba celona, Diagonal 643, 08028, Ba celona, Spain
c
Depa men o Biology and Geology, Physics and Ino ganic Chemis y, Rey Juan Ca los Uni e si y, Tulip´
an S/n, 28933, M´
os oles, Spain
d
Ins i u o Mul idisciplina io Sob e Ecosis emas y Desa ollo Sus en able, Uni e sidad Nacional Del Cen o de La P o incia de Buenos Ai es, CONICET, Campus
Uni e si a io, Pa aje A oyo Seco S/n, Tandil, 7000, Buenos Ai es, A gen ina
e
Facul ad de Ciencias Na u ales y Museo, Uni e sidad Nacional de La Pla a. C.C 712-1900, La Pla a, A gen ina
Ge man Cen e o In eg a i e Biodi e si y Resea ch (iDi ) Halle-Jena- Leipzig, Leipzig, Ge many
g
Ins i u e o Biodi e si y, Uni e si y o Jena, Jena, Ge many
h
Depa men o Ma hema ics and Expe imen al Sciences Didac ics, Facul y o Educa ion o Bilbao, Uni e si y o he Basque Coun y (UPV/EHU), Ba io Sa iena S/n,
48940, Leioa, Spain
i
Cen e o Biodi e si y Theo y and Modelling, Theo e ical and Expe imen al Ecology S a ion, F ench Na ional Cen e o Scien i ic Resea ch, Moulis, F ance
ARTICLE INFO
Keywo ds:
BACI expe imen
Communi y size-spec a
Di e si y
Ene gy luxes
F eshwa e ood webs
T ea ed was ewa e s
ABSTRACT
Was ewa e ea men plan s (WWTPs) ha e g ea ly imp o ed wa e quali y globally. Howe e , ea ed e luen s
s ill con ain a complex cock ail o pollu an s whose en i onmen al e ec s migh go unno iced, masked by
addi ional s esso s in he ecei ing wa e s o by spa io empo al a iabili y. We conduc ed a BACI (Be o e-A e /
Con ol-Impac ) ecosys em manipula ion expe imen , whe e we di e ed pa o he e luen o a la ge e ia y
WWTP in o a small, unpollu ed s eam o assess he e ec s o a well- ea ed and highly dilu ed e luen on
i e ine di e si y and ood web dynamics. We sampled basal ood esou ces, ben hic in e eb a es and ish o
sea ch o changes on he s uc u e and ene gy ans e o he ood web wi h he e luen . Al hough e luen
oxici y was low, i educed di e si y, inc eased p ima y p oduc ion and he bi o y, and educed ene gy luxes
associa ed o e es ial inpu s. Al oge he , he e luen dec eased o al ene gy luxes in s eam ood webs,
showing ha ea ed was ewa e can lead o impo an ecosys em-le el changes, a ec ing he s uc u e and
unc ioning o s eam communi ies e en a high dilu ion a es. The p esen s udy shows ha cu en p ocedu es
o ea was ewa e can s ill a ec eshwa e ecosys ems and highligh s he need o u he e o s o ea
pollu ed wa e s o conse e aqua ic ood webs.
1. In oduc ion
The implemen a ion o was ewa e ea men plan s (WWTP) led by
en i onmen al egula ions such as Eu opean Wa e F amewo k Di ec-
i e (2000/60/EC) o he No h Ame ican Clean Wa e Ac has u ned
in o signi ican imp o emen s in wa e quali y by educing he amoun
o con aminan s eaching aqua ic ecosys ems (B ion e al., 2015; B auns
e al., 2022). Howe e , ea ed sewage s ill consis s o a complex mix u e
o pollu an s, nu ien s, and pa hogens (Pascual-Beni o e al., 2020;
Wei e e e al., 2021). Some o hese compounds a e always oxic
(Vasilachi e al., 2021; Ku wadka e al., 2022) and can ha e de imen al
e ec s on bio a, whe eas o he s, like nu ien s, can also subsidize bio-
logical ac i i y bu become haza dous abo e ce ain concen a ions
(Wang e al., 2019). The composi ion o he pollu an s in e luen s a y
depending on he design and pe o mance o WWTPs, which can include
p ima y (physical) and seconda y (biological) ea men s, o e en a
mo e ad anced e ia y ea men o educe nu ien s, o ganic ma e
and me als (Rocca o, 2018). In Eu ope o ins ance, 69% o he popu-
la ion is connec ed o a WWTP pe o ming some kind o e ia y ea -
men , and ano he 13% o a WWTP wi h a seconda y ea men (EEA,
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (I. de Guzman).
Con en s lis s a ailable a ScienceDi ec
Jou nal o En i onmen al Managemen
jou nal homepage: www.else ie .com/loca e/jen man
h ps://doi.o g/10.1016/j.jen man.2023.118510
Recei ed 23 Feb ua y 2023; Recei ed in e ised o m 22 June 2023; Accep ed 23 June 2023
Jou nal o En i onmen al Managemen 345 (2023) 118510
2
2021). These WWTPs ha e signi ican ly con ibu ed o he eco e y o
eshwa e di e si y in Eu opean i e s, bu wo yingly, his eco e y
a e has s opped du ing he las wo decades (Haase e al. Accep ed),
sugges ing he need o u he imp o emen s in he ea men o
sewage. In addi ion o hei composi ion, he e ec s o e luen s on i e s
s ongly depend on dilu ion capaci y (Bü ne e al., 2022), which is
globally being educed due o clima e change and human ac i i ies (Shi
e al., 2019). Ope a ing laws o low discha ges in ecei ing wa e
bodies a e no he only challenges ha wa e depu a ion su e s om,
po en ially explaining he obse ed pause in he imp o emen o wa e
quali y. The e ec s o well- ea ed (i.e. e ia y ea ed) and highly
dilu ed e luen s, al hough weake , could be signi ican , since e en low
p essu es can esul in clea e ec s on ecological esponses i hey a e
main ained in ime (Hilleb and e al., 2020). These mo e sub le e ec s
could emain unno iced by common moni o ing su eys, such as he
ones compa ing ups eam and downs eam eaches om WWTPs
(Combe e al., 2022), as ecei ing wa e masses a e o en subjec o
o he co-occu ing s esso s o because o di icul ies o isola e he e -
ec s o in e es om con ounding ac o s inhe en o spa io empo al
a iabili y. This di icul y o p ope ly es ima ing e ec s o WWTPs calls
o ca e ully designed expe imen al s udies, such as BACI (Be o -
e-A e /Con ol-Impac ) expe imen s ha con ol spa io empo al a i-
a ions (Downes e al., 2002).
T ea ed was ewa e e luen s can igge bo om-up ophic cascades
by s imula ing algal g ow h (Pe eda e al., 2020), and a ec he
composi ion and s uc u e o ood webs by al e ing densi y and di e si y
o eshwa e communi ies (Gonz´
alez e al., 2023). Thus, as biodi e si y
has been posi i ely linked o ecosys em unc ioning (Lo eau e al., 2021)
and ecological se ices (Isbell e al., 2011), e luen s can ul ima ely in-
luence ecosys em unc ions (B auns e al., 2022). In his ega d, ood
web s udies p o ide a quan i a i e amewo k ha combine communi y
ecology wi h ecosys em ecology, me ging bo h, species-cen ed ap-
p oaches and ene gy lux-cen ed app oaches (Thompson e al., 2012).
In his line, a use ul way o in e how ood webs espond o s esso s is o
analyse he dis ibu ion o indi idual body sizes (i.e. communi y
size-spec a), which is a ein e p e a ion o he ophic py amids (T e-
bilco e al., 2013) and in o ms abou how e icien ly ene gy is ans-
e ed along he ood web o he ecological py amid (Whi e e al., 2007;
Collye e al., 2023), and he e o e abou he capaci y o he communi y
o sus ain species om highe ophic le els. This app oach has been
p oposed as a ool o ecological assessmen (Pe chey and Belg ano,
2010) bu has no been used o de e mine he e ec s o WWTP impac s.
In his con ex , WWTP e luen s a e expec ed o ha e complex and
en angled e ec s o communi y size-spec a. The nu ien inpu om
WWTP e luen s o ins ance, can inc ease ophic ans e e iciency by
educing he s oichiome ic gap be ween consume s and esou ces
(Mulde and Else , 2009; O e al., 2014) inc easing he slope o he
communi y size spec a. Besides, oxic compounds can dec ease he
abundance o a speci ic size class (Baho e al., 2019) o inc ease he
abundance o ole an axa (Pe al a-Ma a e e al., 2019). In addi ion,
changes in composi ion, di e si y and s uc u e o communi ies can lead
o changes in he amoun o ene gy luxes ans e ed along ood webs.
These al e a ions can also p o ide in o ma ion abou modi ica ions in
ecosys em unc ions such as he bi o y, de i i o y o ca ni o y (Ba nes
e al., 2018).
In he p esen s udy, we aim a add essing he e ec s o well ea ed
and highly dilu ed e luen on communi ies and ecosys em unc ional-
i ies. Fo ha , a whole-ecosys em manipula ion expe imen ollowing a
BACI design was pe o med, whe e pa o he e luen o a WWTP was
di e ed o a nea by, unpollu ed s eam o assess he esponse o he
ecosys em o he e luen a di e en ecological scales. Mo e p ecisely,
we quan i ied (i) he esponse o biodi e si y using classical me ics, (ii)
communi y s uc u e h ough he use o size spec a and (iii) ecosys em
unc ions h ough he calcula ion o ene gy luxes in ood webs. P e ious
labo a o y and ield s udies showed li le oxici y o he e luen o in-
e eb a es and mic obes (Solagais ua e al., 2018; Gonz´
alez e al.,
2023). S ill, we expec ed he mos sensi i e axa o dec ease in abun-
dance, inducing de imen al e ec s on di e si y a di e en scales. We
also p edic ed he e luen addi ion o p omo e ene gy lux h ough he
g een ood web (i.e. bio ilm-based), inc easing he bi o y and he de-
pendency o he ood web on bio ilm. We inally p edic ed ha enhanced
nu ien inpu s would p omo e he e iciency o he ene gy ans e
along he ood web, leading o shallowe slopes o he size spec a.
2. Ma e ials and me hods
2.1. S udy si e and expe imen al design
The expe imen was conduc ed in he Ap ai z S eam (N-Ibe ian
Peninsula, 43◦13
′
41.1
″
N 2◦23
′
56.3
″
W), a small unpollu ed s eam wi h a
mean discha ge o 0.12 m
3
s
-1
d aining a 7 km
2
ca chmen o e sand-
s one and shale. In he lowe mos 300 m, i s i e bed is domina ed by
bed ock and cobbles. The young ipa ian o es su ounding i is mainly
composed o black alde (Alnus glu inosa (L.) Gae n.), hazel (Co ylus
a ellana L.) and ash (F axinus excelsio L). This each uns nex o he
WWTP o Ap ai z which eleases he ea ed sewage in o he Deba Ri e
(mean ±SE discha ge du ing ou s udy 10.9 ±0.7 m
3
s
-1
, h p://gipu
zkoa.eus/) h ough a egula ly pulsed discha ge (20–40 min e e y 2
h). The WWTP ecei es he sewage o app oxima ely 90,000 popula ion
equi alen s om u ban and indus ial a eas, ea ing on a e age 29,904
m
3
o was ewa e pe day in sequen ial biological eac o s (h ps://www
.acciona-agua.com). A e mixing he sewage wi h ac i a ed sludge and
subjec ing i o ae obic and anae obic p ocesses o educe he load o
o ganic ma e and ni ogen, i ecei es e ia y ea men (p ecipi a ion
wi h e ic sulpha e) o educe phospho us concen a ions.
The expe imen ollowed a BACI design (Fig. S1), which allows
con olling bo h empo al and spa ial a iabili y o isola e he e ec o
in e es (Downes e al., 2002). We di e ed pa o he WWTP e luen
in o he lowe mos 150 m o he Ap ai z S eam o ge a inal dilu ion
a e simila o ha o he e luen discha ged in o he Deba Ri e
(0.2–4% and 0.1–9% o ange o e luen concen a ion, espec i ely;
Pe eda e al., 2020). We s udied wo 100-m long eaches: a Con ol
each, ups eam om he e luen addi ion poin and an Impac each
jus below. Reaches we e sepa a ed by a 20 m-long bu e each ha was
composed by a man-made conc e e la su ace ha was able o educe
wa e dep h o less han 5 cm and ha ended in an o e hung wa e all o
abou 50 cm making ups eam mig a ion o ish i ually impossible
du ing low low pe iods. Bo h eaches o med a con inuum only on he
s onges spa es. Fo his s udy, one sampling was conduc ed a each
each be o e (Sep embe 2016) and a e (Oc obe 2017) he s a o he
e luen addi ion (May 2017), bo h low- low pe iods, en ailing a low
dilu ion o he e luen du ing he a e pe iod. The e luen elease
changed signi ican ly wa e physico-chemical cha ac e is ics a he
impac each (Pe eda e al., 2020): dissol ed oxygen sa u a ion and pH
dec eased by 10% bu empe a u e emained una ec ed. Nu ien con-
cen a ions signi ican ly inc eased, especially ammonium and soluble
eac i e phospho ous (5.2 and 2.4 imes espec i ely). Howe e , o he
nu ien s such as dissol ed o ganic ca bon o sulpha e we e no al e ed
(Pe eda e al., 2020). Ye , physico-chemical cha ac e is ics in he impac
each du ing he addi ion o he e luen we e wi hin he e y good
s a us ca ego y desc ibed in Mendigu en e al. (2021).
2.2. Sampling and sample p ocessing
2.2.1. Bio ilm
Be o e he s a o he expe imen , we deployed a i icial subs a a
(g ani e pa ing s ones o 20 ×10 ×8 cm) along each each (co e ing
less han 0.5% o he s eambed) o allow o bio ilm coloniza ion. Ac-
co ding o Pe eda e al. (2020), bio ilm was o med by a simila amoun
o pho oau o ophic and he e o ophic mic oo ganisms, wi h a endency
o ne au o ophy. In each sampling pe iod and each, we sc apped he
whole su ace o i e andomly chosen pa ing s ones and p ocessed
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
3
hem o ob ain ash- ee d y mass pe su ace uni (g m
−2
) (comple e
me hodology in Pe eda e al., 2020).
2.2.2. Ben hic o ganic ma e and mac oin e eb a es
We collec ed nine ben hic Su be samples (su ace o 0.09 m
2
, mesh
o 0.5 mm) andomly along each each o he Ap ai z S eam in each
sampling pe iod and each. F om each sample, he o ganic ma e
e ained on an 8-mm sie e was ga he ed, lea es we e sepa a ed om he
es o he coa se de i us and bo h ca ego ies we e sepa a ely p ocessed
o ob ain hei ash- ee d y mass in each sample (see Pe eda e al., 2020).
Since consump ion o wood and o he ecalci an ma e ials by mac o-
in e eb a es is negligible compa ed o lea es (Díez e al., 2002) we only
e ained lea es (he ea e coa se de i us). Be o e ge ing he ash- ee
d y mass o coa se de i us, a subsample was kep o he assessmen
o ca bon and ni ogen concen a ions. F om hese Su be samples,
mac oin e eb a es collec ed in a 0.5-mm sie e we e p ese ed in 96%
e hanol. In he labo a o y hey we e so ed, iden i ied o he lowes
possible axonomic le el ollowing Tache e al., (2010) (mos ly o
genus-le el, excep o some Dip e a iden i ied o sub amily le el and
Annelida o subclass le el) and coun ed o ob ain popula ion densi ies.
In addi ion, we measu ed he body leng h o up o 30 andomly selec ed
indi iduals o each axon in e e y sample (excep o oligochae es,
plana ians and leeches, which we e no measu ed) wi h a binocula
mic oscope (Leica M165FC, We zla , Ge many) equipped wi h a Leica
DFC310FX came a using “Leica Applica ion sui e V4
″
so wa e p og am.
To al body leng h was conside ed as he dis ance om he an e io pa
o he head o he pos e io pa o he las abdominal segmen excluding
an ennae and ails (Meye , 1989). Fo gas opod molluscs, we measu ed
he maximum leng h o he shell (Meye , 1989), and o c us aceans o
he genus Echinogamma us, he do sal leng h o he i s abdominal
segmen was measu ed o pos e io ly ob ain body leng h (Flo es e al.,
2014). Indi idual body mass (mg d y mass) was de i ed using published
leng h-mass ela ionships (Meye , 1989; S o els e al., 2003; La a˜
naga
e al., 2009).
2.2.3. Fish
We conduc ed ish samplings along he 100-m long eaches by
deple ion elec o ishing wi h a backpack-elec o ishing uni wi h a i-
able ou pu cu en (MARTIN PESCADOR III, Albo lan S.L.). S op-ne s
we e se a he ups eam and downs eam ends o he eaches and up
o h ee uns we e made. All ish we e anaes he ized wi h MS-222,
iden i ied, coun ed and weighed ( o he nea es g). The s udy was
app o ed by he E hics Commi ee o Resea ch In ol ing Animals a he
Uni e si y o he Basque Coun y (M20/2016/177) and was conduc ed
in con o mance wi h all applicable laws. We con e ed we mass in o
ash- ee d y mass h ough con e sion ac o s published in www. i
shbase.se.
2.2.4. Quali y o basal ood esou ces
Six samples o bio ilm we e collec ed in each each by sc aping he
en i e su ace o andomly chosen cobbles and collec ing he slu y in
il e ed i e wa e (0.7
μ
m po e size, Wha man GF/F). Six ine de i us
samples we e andomly collec ed pe each and pe iod using a sedimen
co e (su ace 81.7 cm
2
). Bio ilm and ine de i us samples we e ozen
(−20 ◦C) and eeze-d ied (Vi Tis Bench op 2 K) ( om 12 o 72 h
depending on hei wa e con en ). Each subsample o coa se de i us
was o en d ied (72 h, 70 ◦C). F eeze-d ied and o en-d ied ma e ial was
g ounded (Vib a ion mill MM301, Fishe Bioblock Scien i ic) and
weighed (app oxima ely 10 mg o ine de i us and 2 mg o he o he
basal esou ces) in o in capsules (Lüdiwiss Sn 98, 5 ×8 mm) o ca bon
and ni ogen analysis. Ca bon and ni ogen analyses o bio ilm and ine
de i us we e pe o med a he Helmhol z-Zen um ü Umwel o -
schung (Magdebu g, Ge many) by a lash combus ion on a Flash 2000
elemen al analyze (The mo Fishe Scien i ic, B emen, Ge many) and
o coa se de i us a he S able Iso ope Facili y o he Uni e si y o
Cali o nia – Da is on a PDZ Eu opa ANCA-GSL elemen al analyze
(Se con L d., Cheshi e, UK).
2.3. Da a ea men
2.3.1. In e eb a e di e si y
We assessed in e eb a e axa di e si y a di e en scales h ough
Hill numbe s (i.e. numbe equi alen ,
q
D (Jos , 2006)) wi h he en o-
pa package o R (Ma con and H´
e aul , 2015). We used Hill numbe s o
o de 0 (
0
D, species ichness, which is insensi i e o he abundance o
indi iduals o each axon, highligh ing he esponse o a e axa), 1 (
1
D,
he exponen ial o Shannon’s en opy, which weighs each axon ac-
co ding o i s log- ans o med abundance), and 2 (
2
D, in e se o Simp-
son concen a ion, which weighs each axon acco ding o i s abundance,
highligh ing he esponse o dominan axa) (Jos , 2006). We compu ed
alpha-di e si y pe sample in each pe iod and each o he h ee Hill
numbe o de s and be a-di e si y among samples (i.e. pai wise dissim-
ila i y (Socola e al., 2016)) wi hin each pe iod and each o o de s
0 and 1 o di e si y measu es. Be a-di e si y (D
β
) o he di e si y o de s
(q) 0 and 1 was ans o med om be a en opy (H
β
) as desc ibed in
Ma con and H´
e aul (2015):
qDβ=eq
q Hβ
1− (q−1)qH
α
(1)
2.3.2. Food webs, ene gy lux and dependency on basal esou ces
Following he numbe o eplica es o Su be samples pe pe iod and
each we cons uc ed 9 local ood webs. In o ma ion abou o ganic
ma e and in e eb a es was ob ained om he Su be samples. In he
case o bio ilm, we used he a e age biomass pe pa ing-s one su ace in
each pe iod and each o es ima e o al biomass in each ood web. Fine
de i us was a sca ce basal esou ce wi h a he e ogeneous dis ibu ion
along he each and was no quan i ied du ing he sampling campaign,
so i was equalled o he mean bio ilm biomass alues eco ded. In he
case o ish, assuming a homogeneous dis ibu ion o ish along he
each, we es ima ed he o al biomass ela ed o each Su be sample
(0.09 m
2
). Fo e e y in e eb a e axon in each sample, we calcula ed
mean body mass om he indi idual masses measu ed and es ima ed he
o al biomass pe node. We also es ima ed o al me abolic a e o each
in e eb a e and ish nodes based on indi idual me abolic a e, calcu-
la ed o each indi idual using an allome ic equa ion de i ed om
B own e al. (2004):
X=exp((a⋅ln(BM) + x0) + E/kT)(2)
whe e X is he me abolic a e (in wa s, W), a is he allome ic exponen
(0.71), BM is he body mass (g), E is he ac i a ion ene gy (0.63 eV), k is
he Bol zmann’s cons an (8.62 ⋅ 10
−5
eV K
−1
), T is he empe a u e (K)
and x
0
is a no maliza ion ac o (17.17 o in e eb a es and 18.47 o
ish). All hese pa ame e s we e ex ac ed om B own e al. (2004). We
ga he ed mean daily empe a u e o he 190 days be o e he sampling
da e in each each.
To es ima e ene gy luxes be ween nodes o local ood webs, we used
an adap ed ood-web ene ge ics app oach (Ba nes e al., 2018; Gauzens
e al., 2019) by means o he “ luxweb” package (Gauzens e al., 2019).
This app oach uses allome ic scaling laws o quan i y me abolic a es
(B own e al., 2004). The model assumes a s eady-s a e sys em, whe e
he ene ge ic losses o nodes in each ood web, es ima ed by me abolic
a e o consume j (X
j
) and p eda ion on consume j by highe ophic
le els (k), need o be balanced by he ene ge ic gains de ined h ough
esou ce consump ion and assimila ion (Ba nes e al., 2018). The lux o
ene gy F
i,j
om esou ce i o consume j was calcula ed as:
∑
i
ƐijFij =Xj+∑
k
WjkFk(3)
whe e Ɛ
ij
is he e iciency in which consume j assimila es he ene gy
consumed om esou ce i. Ene gy luxes o each consume a e de ined
as Fij =WijFj, whe e F
j
is he sum o all he ingoing luxes o consume j
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
4
and W
ij
is he p opo ion o F
j
ob ained om esou ce/p ey i, a e
scaling consume p e e ences w
ij
o he biomass (B) o he di e en e-
sou ces/p eys as:
Wij =wij Bi
∑kwkj Bk
(4)
Fo ha , an adjacency ma ix wi h ophic links among all axa
p esen in ou s udy and eeding p e e ences o each ood esou ce was
c ea ed based on he li e a u e (Tache e al., 2010; G ay e al., 2015)
and ou own gu con en indings (see p o ided da ase in de Guzman
e al., 2023). Based on eeding p e e ences h ee ophic g oups we e
conside ed: p ima y consume s ( eeding on basal ood esou ces), om-
ni o es ( eeding on basal ood esou ces and p eys), ca ni o es ( eeding
on p eys). Fo ca ni o e axa we assumed ha p e e ences we e equally
dis ibu ed amongs p ey species. Fo omni o e in e eb a es and p i-
ma y consume s w alues we e gi en ollowing p e e ences in Tache
e al. (2010), whe e ai s ela ed o consumed ood a e quan i ied using
a ini y sco es be ween 0 and 5. Fo omni o es, a ini y sco es ela ed o
p eda ion we e equally dis ibu ed amongs p ey species. Fo canni-
balis ic species, we se he p e e ence o cannibalism o 0.01 in he
adjacency ma ix o minimize he amoun o ene gy a consume could
inges om i s own biomass pool. Assimila ion e iciencies (Ɛ) o he
consump ion o ood esou ces we e calcula ed de i ing a o mula om
Lang e al. (2017):
Ɛ=eƐ
′
eE(T−T0)/kTT0
1+eƐ
′
eE(T−T0)/kTT0
(5)
whe e Ɛ
′
is no maliza ion cons an o assimila ion e iciency (−1.670
o de i i o y, 0.179 o he bi o y and 2.260 o ca ni o y), E is he
ac i a ion ene gy (0.164 eV), k is he Bol zmann’s cons an and T is
empe a u e (K) and T
0
empe a u e no malized o 20 ◦C. Pa ame e s
we e ex ac ed om Lang e al. (2017).
We calcula ed whole- ood web ene gy lux as he sum o ene gy
luxes wi hin each local ood web (each Su be sample). To calcula e he
consump ion on bio ilm, de i us and p eys, we summed all he ou going
ene gy luxes om each ood esou ce. Thus, we quan i ied h ee con-
sump ion pa hways: he bi o y (consump ion o bio ilm), de i i o y
(consump ion o coa se and ine de i us) and ca ni o y (consump ion o
animals) in he en i e ood web and wi hin each ophic g oup (p ima y
consume s, omni o es and ca ni o es) o each local ood web. In addi-
ion, we calcula ed he dependency o each ophic g oup on basal ood
esou ces (bio ilm and de i us) wi h he “Ne Indices” package (Soe ae
and Kones, 2014).
2.3.3. Size spec a
We cons uc ed size spec a o he en i e communi y (including bo h
in e eb a e and ish assemblages), and o p ima y consume s, omni-
o es and ca ni o es, sepa a ely. We used body mass o he measu ed
(in e eb a es) and weighed ( ish) o ganisms. Since he log- ans o med
leng h alues ollowed a no mal dis ibu ion, we ob ained body mass o
he emaining non-measu ed in e eb a es by means o he “ uncno m”
package (Me smann e al., 2018), based on he mean, s anda d de ia-
ion, minimum and maximum alues o each axon in each sample. We
used animals wi h body mass highe han 0.1 mg o cons uc size
spec a, since o ganisms wi h lowe weigh s a e unde sampled as hey
a e mos ly washed h ough he 0.5 mm mesh ape u e sie es (G uene
e al., 2007). We di ided he o al ange o body mass (log
10
body mass)
alues in o 8 loga i hmic bins o he same wid h and eg essed densi y o
o ganisms (log
10
densi y) agains he cen e o he bin (Whi e e al.,
2007).
2.3.4. S a is ical analyses
We conduc ed linea models wi h he ac o s pe iod (Be o e and
A e ), each (Con ol and Impac ) and hei in e ac ion ( o assess he
e ec o he e luen ) as sou ces o a ia ion (as well as he co a ia e
body mass o size spec a compa isons). This kind o BACI models do
some imes pose some challenges when g aphically in e p e ing he e-
sul s, as he s esso can dissipa e ins ead o ampli y ini ial di e ences
be ween con ol and impac si es (see Fig. S2 in Suppo ing In o ma ion
o an illus a i e explana ion). Coe icien s o he in e ac ion e m
poin he di ec ion o he e ec o in e es . Some a iables we e log-
ans o med o ul il he equi emen s o linea models. We conduc -
ed all he analyses using R so wa e, e . 3.6.0. (R Co e Team, 2019). Fo
de ails ega ding he code and collec ed da a, see de Guzman e al.
(2023).
3. Resul s
3.1. S ock and quali y o basal esou ces
Coa se de i us s ock ended o be highe in he impac each (F
1,31
=
3.90, p =0.057, Fig. S3, Table S1). Bio ilm s ock di e ed be ween pe-
iods (F
1,16
=5.56, p =0.031, Fig. S3), and exhibi ed a ma ginally
signi ican inc ease as a consequence o he addi ion o he e luen
(F
1,16
=3.84, p =0.068, Fig. S3, Table S1). In addi ion, ca bon- o-
ni ogen a ios (Table S2) signi ican ly di e ed among basal ood e-
sou ces (F
2,75
=240.8, p <0.001), wi h bio ilm showing he lowes and
coa se de i us he highes a ios, bu he addi ion o he e luen
a ec ed none (Table S1).
3.2. In e eb a e and ish di e si y and abundance
Alpha-di e si y o mac oin e eb a e axa ichness signi ican ly
di e ed be ween eaches (Fig. 1, Table 1) and wi h he addi ion o he
e luen dec eased in he impac each espec ing he con ol each
(F
1,30
=16.07, p <0.001, Fig. 1, Table 1). Howe e , he e luen did no
signi ican ly a ec Shannon and Simpson’s di e si y (Fig. 1, Table 1).
Mac oin e eb a e be a-di e si y o axa ichness was ini ially highe in
he con ol each, bu he di e ence disappea ed wi h he addi ion o he
e luen (F
1,124
=27.89, p <0.001, Fig. 1, Table 1). We obse ed a
simila pa e n o Shannon a be a-di e si y scale: e luen addi ion led
o less homogeneous communi ies (F
1,124
=5.73, p =0.018, Fig. 1,
Table 1). Mo eo e , he e luen had a nega i e ne impac on in e e-
b a e densi y, as he ini ial di e ences ound be ween he wo expe i-
men al eaches dec eased (F
1,30
=4.76, p =0.037, Fig. 1, Table 1).
Rega ding ish assemblages, ou species we e p esen du ing he s udy,
and in bo h pe iods, mo e indi iduals and highe biomass we e ound in
he impac each, bu o e all, indi iduals a he con ol each we e
bigge (Table S3).
3.3. Ene gy luxes and dependency on basal ood esou ces
Mean body mass, o al biomass and me abolic a e o he h ee o-
phic g oups did no change wi h he addi ion o he e luen (Fig. S4,
Table S1). Howe e , ene gy luxes o he en i e communi y di e ed
be ween pe iods and eaches (Fig. 2, Table 1), and he addi ion o he
e luen caused a nega i e ne e ec on hem (F
1,30
=6.62, p =0.015,
Fig. 2, Table 1). Mo eo e , some luxes we e also sensi i e o he
e luen . He bi o y inc eased wi h he addi ion o he e luen (F
1,30
=
7.75, p =0.010, Fig. 2, Table 1), and in he a e pe iod di e ences in
de i i o y be ween eaches dec eased (F
1,30
=10.23, p =0.003, Fig. 2,
Table 1). Howe e , al hough ca ni o y signi ican ly a ied be ween
pe iods and eaches, i was una ec ed by he addi ion o he e luen
(Fig. 2, Table 1). Omni o es we e esponsible o he inc ease in he -
bi o y du ing he addi ion o he e luen , as ene gy luxes signi ican ly
inc eased wi hin his ophic g oup (F
1,30
=5.47, p =0.026, Fig. S5,
Table S1). On he o he hand, he dec easing di e ences in de i i o y
be ween eaches a e he addi ion o he e luen , we e d i en by he
dec ease in de i i o y among p ima y consume s (F
1,30
=58.62, p <
0.001, Fig. S5, Table S1). Ca ni o y was una ec ed in any ophic g oup
wi h he addi ion o he e luen (Fig. S5, Table S1). O e all, all ophic
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
5
g oups depended on de i us mo e han on bio ilm (Fig. S6). Howe e ,
dependency o ca ni o es on bio ilm signi ican ly inc eased wi h he
addi ion o he e luen (F
1,30
=7.66, p =0.010, Fig. S6; Table 1), while
dependency on de i us dec eased (F
1,30
=7.99, p =0.010, Fig. S6;
Table 1), wi h no o he g oup showing a a ia ion on hei dependency
on bio ilm o de i us.
3.4. Size spec a
The slope o he size spec um o he en i e communi y was una -
ec ed by he addi ion o he e luen (Fig. S7, Table 2: Body mass:BA:CI,
Fig. S8), as well as he slopes o he h ee ophic g oups o consume s:
p ima y consume s, omni o es and ca ni o es (Fig. S7, Table 2).
Ne e heless, he addi ion o he e luen signi ican ly educed he
in e cep o he p ima y consume s (Table 2: BA:CI; coe icien AI:
0.636).
4. Discussion
E ec s o ea ed was ewa e e luen s depend on e luen composi-
ion, dilu ion and he cha ac e is ics o he ecei ing ecosys em. Ou
whole-ecosys em manipula ion expe imen shows ha e en a well-
ea ed and highly dilu ed (mean dilu ion 3%, ange 0.2–4%) e luen
al e s di e si y and ene gy pa hways on ood webs.
Fig. 1. Di e si y and densi y o mac oin e eb a e assemblage: alpha- and be a-di e si y o axa ichness and o Shannon di e si y, alpha-di e si y o Simpson
di e si y and densi y. The box plo s show he median, he in e qua ile ange and he ails o he dis ibu ion. Do s ep esen ou lie s and dashed lines mean alues.
As e isks indica e he e ec o he e luen (Pe iod:Reach in e ac ion) was signi ican .
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
6
Table 1
Linea model esul s compa ing in e eb a e densi y and di e si y, ene gy luxes h oughou ood webs and dependencies o ophic g oups on basal esou ces be ween
con ol (C) and impac (I) eaches be o e (B) and a e (A) he s a o e luen addi ion o assess i s e ec s (BA:CI).
BA CI BA:CI
F p Coe . F p Coe . F p Coe .
In e eb a e communi y desc ip o s
Taxa ichness Alpha 1.6 0.216 7.83 0.009 10.47 16.07 <0.001 −12.33
Be a 52.21 <0.001 −0.17 11.76 0.001 −0.12 27.89 <0.001 0.14
Shannon di e si y Alpha 4.31 0.046 2.25 1.06 0.311 0.32 0.574
Be a 3.35 0.069 4.81 0.03 −0.09 5.73 0.018 0.1
Simpson di e si y Alpha 2.76 0.107 0.12 0.734 0.38 0.542
In e eb a e densi y* 2.11 0.156 15.65 <0.001 0.51 0.51 0.037 −0.36
Ene gy luxes
En i e communi y* 16.46 <0.001 0.35 71.6 <0.001 0.55 6.62 0.015 −0.26
He bi o y* 8.66 0.006 0.04 11.62 0.002 0.06 7.65 0.010 0.53
De i i o y* 8.83 0.006 0.47 50.86 <0.001 0.73 10.23 0.003 −0.45
Ca ni o y* 6.2 0.019 0.26 8.44 0.007 0.28 1.45 0.238
Dependencies on esou ces
P ima y consume s Bio ilm 0.29 0.591 0.36 0.555 2.36 0.135
De i us 0.29 0.591 0.36 0.555 2.36 0.135
Omni o es Bio ilm 0.45 0.507 0.23 0.633 3.2 0.084
De i us 0.45 0.507 0.23 0.633 3.2 0.084
Ca ni o es Bio ilm 0.32 0.579 0.61 0.440 7.66 0.01 0.21
De i us 0.49 0.490 0.34 0.566 7.99 0.01 −0.23
Foo no e: As e isks indica e log
10
ans o med a iables. Coe icien s a e shown o signi ican esponses (p <0.05) and conside B and C as e e ence in all cases. In
dependencies on basal esou ces F and p alues a e iden ical wi hin each ophic g oup because he sum o dependencies on bio ilm and de i us equals 1.
Fig. 2. Ene gy luxes o s eam ood webs and unc ions: luxes h oughou he en i e ood web and he bi o y, de i i o y and ca ni o y ela ed luxes. The box plo s
show he median, he in e qua ile ange and he ails o he dis ibu ion. Do s ep esen ou lie s and dashed lines mean alues. As e isks indica e he e ec o he
e luen (Pe iod:Reach in e ac ion) was signi ican .
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
7
4.1. The e luen p omo ed bio ilm biomass
Al hough he e ec o he e luen addi ion on bio ilm biomass was
ma ginally signi ican o he sampling campaigns in which ood web
ene gy luxes and size spec a we e quan i ied, Pe eda e al. (2020) e-
po ed a highly signi ican inc ease when conside ing samples collec ed
e e y second mon h du ing he yea he e luen was pou ed in he same
expe imen al se up. Bio ilm g ow h is known o inc ease unde
enhanced nu ien concen a ion (Dodds and Smi h, 2016) and unde
sewage inpu s (Pe eda e al., 2021). Many s udies ha e also epo ed
ha inc eased concen a ions o dissol ed nu ien s p omo e li e
quali y (C oss e al., 2003) and decomposi ion (Fe ei a e al., 2014).
Pe eda e al. (2020) also obse ed an inc ease in de i us decomposi ion
in he same expe imen al se up, al hough we did no obse e changes in
he quali y o de i us in he p esen s udy. Su p isingly, in p esence o
e luen , we did no obse e al e a ions in he ca bon- o-ni ogen a io o
bio ilm ei he , which con as ed wi h o he s udies add essing e ec s o
nu ien en ichmen e en downs eam om e luen discha ge poin s
(Sco e al., 2008).
4.2. E ec s on di e si y we e scale-dependen
The e luen educed in e eb a e ichness and densi y, bu Shannon
and Simpson a alpha-di e si y scale we e una ec ed. S udies add ess-
ing e ec s o ea ed was ewa e e luen s ha e epo ed inc eased
in e eb a e abundance and dec eased di e si y as a consequence o he
loss o he mos sensi i e axa and he s imula ion o he esis an ones
(Mo e al., 2019; Hamdhani e al., 2020; Wei e e e al., 2021), indi-
ca ing ha hese e luen s ac as en i onmen al il e s ha selec
pollu ion ole an axa. In his ega d, Gonz´
alez e al. (2023) also
obse ed a sensi i i y dependan esponse o he abundance, i.e. he
mos sensi i e species educed hei abundance and he mos ole an
ones inc eased i . In he cu en s udy, he highly dilu ed e luen did
s ill nega i ely impac axa ichness a alpha-di e si y scale. Con a ily,
we obse ed ha he e luen signi ican ly inc eased be a-di e si y,
indica ing inc eased spa ial u no e ac oss local communi ies. Mo
e al. (2019) also epo ed an inc ease in be a-di e si y be ween samples
ob ained in eaches downs eam om WWTP e luen s and a ibu ed i
o inc eases in local habi a he e ogenei y, o oppo unis ic esponses o
bio a o inc eased p oduc i i y o o basal en i onmen al di e ences
be ween si es. In ou s udy, we could associa e he inc ease in
be a-di e si y o he s onge ole o s ochas ic assembly e en s caused
by he inc ease in p oduc i i y (Chase, 2010), which p omo es he
oppo unis ic esponses o some species (Langenhede e al., 2012), and
o he une en elimina ion o some sensi i e species o pollu an s (T u-
bina and Vo obeichik, 2012). T ea ed was ewa e e luen s ha e been
epo ed o educe he abundance and biomass o ish (Wei e e e al.,
2021) as well as ish di e si y (Hamdhani e al., 2020). In con as , in
ou s udy he e luen had no clea e ec s on ish di e si y and densi ies,
likely due o he high dilu ion a e. O e all, mo e and smalle in-
di iduals we e ound in he impac each, bu he e luen did no a ec
abundance no body mass.
4.3. He bi o y was s imula ed, bu o al ene gy lux educed
We expec ed inc eased bio ilm o p omo e he ene gy lux along he
g een ood web, as i is o highe nu i ional quali y han de i us (C oss
e al., 2005) and can sus ain a high p oduc ion o p ima y consume s
(McCu chan and Lewis, 2002). Se e al s udies epo ed ha nu ien
addi ion inc eased seconda y p oduc ion as i inc eased he quali y o
basal esou ces (Demi e al., 2018; Ecke e al., 2020). We obse ed no
inc emen in mean body mass, o al biomass o es ima ed me abolic a e
o any ophic g oup. Howe e , we obse ed a ne dec ease in he o al
ene gy lux o he ood web and a change in he amoun o ene gy
p o ided o consume s h ough di e en pa hways, wi h inc eased
he bi o y and educed de i i o y. A simila shi om b own o g een
ood-web pa hways was epo ed h ough he inc ease o he bi o e
abundance in a nu ien en ichmen expe imen whe e de i us was also
he mos abundan basal esou ce (Bumpe s e al., 2017). In ou
expe imen he ene gy luxes owa ds omni o es and p ima y con-
sume s shi ed, al hough none o hese g oups inc eased in biomass o
abundance. Omni o es, able o consume de i us, bio ilm and p eys,
inc eased he consump ion o high-quali y bio ilm, whe eas p ima y
consume s educed he consump ion o low-quali y de i us. Ca ni o y
did no inc ease, bu ca ni o es became mo e dependen on bio ilm
h ough he highe consump ion o he bi o es, which shows ha he
e ec s o he e luen eached he op o he ood web.
4.4. Communi y size spec a we e una ec ed
Wa e nu ien en ichmen can inc ease he quali y (i.e. educe
Ca bon:nu ien a io) o au o ophs and de i us (E ans-Whi e and
Hal o son, 2017) na owing he s oichiome ic gap be ween basal e-
sou ces and p ima y consume s and inc easing he e iciency o ophic
ans e (Mulde and Else , 2009). We did no see a shi in
ca bon- o-ni ogen a io o basal esou ces, al hough we obse ed an
inc ease o he ela i e abundance o bio ilm o e de i us, which
enhanced he o al a ailabili y o nu ien s o consume s. The e o e, we
expec ed a shallowe slope in he body mass-abundance size spec a
(Mulde and Else , 2009; O e al., 2014), as nu ien subsidy a ou s
o ganisms wi h highe pe capi a me abolic equi emen s (C oss e al.,
2003) and inc eases he capaci y o he sys em o suppo la ge in-
di iduals (Ju igny-Khena ou e al., 2021). Howe e , he e luen did no
change he slope o size spec um. This inding could sugges ha he
nu ien en ichmen was no high enough o igge a change in ene gy
ans e e iciency. Addi ionally, oxici y o he e luen migh no ha e
been s ong enough o modi y he slope o he size spec a ei he ,
ollowing o he s udies epo ing size speci ic esponses o con aminan s
(Ge gs e al., 2015; Taddei e al., 2021). The only sign o he e ec o he
oxici y o he e luen on size spec a was he lowe in e cep o p i-
ma y consume s, which is in line o he sub le e ec s o ea ed and
dilu ed e luen on ecophysiology o consume s (Solagais ua e al.,
2018) and ecosys em unc ioning (Pe eda e al., 2020).
Table 2
Linea model esul s compa ing size spec a be ween con ol (C) and impac (I) eaches be o e (B) and a e (A) he s a o e luen addi ion o assess i s e ec s (BA:CI)
and conside ing body mass.
En i e communi y P ima y consume s Omni o es Ca ni o es
F p Coe . F p Coe . F p Coe . F p Coe .
Body mass 304.20 <0.001 −0.95 164.55 <0.001 −0.78 245.34 <0.001 −0.79 66.70 <0.001 −0.28
BA 6.03 0.022 0.58 11.58 0.002 0.54 13.04 0.001 0.19 13.34 0.001 0.45
CI 0.02 0.888 0.13 0.722 21.15 <0.001 0.39 0.00 0.954
Body mass:BA 0.16 0.695 6.00 0.022 0.13 0.86 0.364 0.00 0.968
Body mass:CI 0.66 0.423 1.71 0.203 3.99 0.057 −0.27 0.87 0.359
BA:CI 0.89 0.356 10.48 0.004 −0.64 1.54 0.227 0.91 0.350
Body mass:BA:CI 0.04 0.844 2.07 0.163 0.13 0.720 0.10 0.757
Foo no e: Coe icien s a e shown o signi ican esponses (p <0.05) and conside B and C as e e ence in all cases.
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
8
5. Conclusion
Ou whole-ecosys em manipula i e expe imen shows ha e en well
ea ed and highly dilu ed WWTP e luen s can s ill modi y he s uc u e
and unc ioning o s eam communi ies. The educ ion o alpha- and he
inc ease o be a-di e si y show ha he e luen educed he numbe o
species, bu inc eased he dissimila i y among he di e en pa ches o
he i e bed. The o e all size s uc u e o he communi y did no
change, ene gy luxes h ough he bi o y inc eased and hose h ough
de i i o y dec eased, bu a nega i e e ec on he o al lux along he
en i e communi y was igge ed. Ou expe imen unde lines ha cu en
p ocedu es o ea con amina ed wa e s migh no be enough o p e-
se e na u al p ope ies o ood webs and ha u he e o s migh be
needed when dealing wi h sensi i e o highly aluable aqua ic
ecosys ems.
Au ho s a emen
AE, AL, D S and JMG designed he s udy. IdG, AE, AL, D S, JMG,
LEP, AA and NO con ibu ed o collec ing he da a. IdG and LEP con-
duc ed labo a o y wo k. IdG, AL, JMM, BG and UB discussed and ana-
lysed da a. IdG, AL and JMM d a ed he manusc ip and all he au ho s
p o ided eedback, con ibu ed o he w i ing and ga e inal app o al
o publica ion.
Open esea ch s a emen
The o iginal da a ha suppo s he indings o his s udy and he code
a e openly a ailable in Figsha e (h ps://doi.o g/10.6084/m9. igsh
a e.21185224. 1) and Gi Hub.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing inancial
in e es s o pe sonal ela ionships ha could ha e appea ed o in luence
he wo k epo ed in his pape .
Da a a ailabili y
The o iginal da a ha suppo s he indings o his s udy and he code
a e openly a ailable in Figsha e (h ps://doi.o g/10.6084/m9.
igsha e.21185224. 1) and Gi Hub
Acknowledgemen s
This esea ch was suppo ed by he EU Se en h F amewo k P o-
g amme (GLOBAQUA; 603629-ENV-2013-6.2.1). Au ho s also
acknowledge he inancial suppo om he Basque Go e nmen
(Consolida ed Resea ch G oup IT951-16) and he p e-doc o al ellow-
ship om he Basque Go e nmen (I. de Guzman). D. on Schille is a
Se a Hún e Fellow. J. M. Mon oya is unded by he FRAGCLIM
Consolida o G an (numbe 726176) om he Eu opean Resea ch
Council unde he EU Ho izon 2020 Resea ch and Inno a ion P og am
and by he “Labo a oi es d’Excellences (LABEX)" TULIP (ANR-10-LABX-
41). Au ho s a e especially g a e ul o Libe Solagais ua, Ola z Pe eda,
Vicki Pe ez, Mi en A is ain, Jani e Diez, Madalen Goi ia, Lau a Sie a
and Kelly Huaman umba o hei assis ance du ing ield campaigns and
labo a o y analyses, and o he s a o he Ap ai z WWTP o con inuous
suppo du ing he expe imen . Ma io B auns and Ines Locke a e also
g a e ully acknowledged o p o iding da a on esou ce quali y.
Appendix A. Supplemen a y da a
Supplemen a y da a associa ed wi h his a icle can be ound, in he
online e sion, a h ps://doi.o g/10.1016/j.jen man.2023.118510.
Re e ences
Baho, D.L., Poma i, F., Leu, E., Hessen, D.O., Moe, S.J., No be g, J., Nizze o, L., 2019.
A single pulse o di use con aminan s al e s he size dis ibu ion o na u al
phy oplank on communi ies. Sci. To al En i on. 683, 578–588. h ps://doi.o g/
10.1016/j.sci o en .2019.05.229.
Ba nes, A.D., Jochum, M., Le check, J.S., Eisenhaue , N., Sche be , C., O’Conno , M.I., de
Rui e , P., B ose, U., 2018. Ene gy lux: he link be ween mul i ophic biodi e si y
and ecosys em unc ioning. T ends Ecol. E ol. 33, 186–197.
B auns, M., Allen, D.C., Bo¨
echa , I.G., o he s, 2022. A global syn hesis o human impac s
on he mul i unc ionali y o s eams and i e s. Global Change Biol. 28, 4783–4793.
h ps://doi.o g/10.1111/gcb.16210.
B ion, N., Ve banck, M.A., Bauwens, W., Elskens, M., Chen, M., Se ais, P., 2015.
Assessing he impac s o was ewa e ea men implemen a ion on he wa e quali y
o a small u ban i e o e he pas 40 yea s. En i on. Sci. Pollu . Res. 22,
12720–12736.
B own, J.H., Gillooly, J.F., Allen, A.P., Sa age, V.M., Wes , G.B., 2004. Towa d a
me abolic heo y o ecology. Ecology 85, 1771–1789. h ps://doi.o g/10.1890/03-
9000.
Bumpe s, P.M., Rosemond, A.D., Mae z, J.C., Bens ead, J.P., 2017. Expe imen al nu ien
en ichmen o o es s eams inc eases ene gy low o p eda o s along g eene ood-
web pa hways. F eshw. Biol. 62, 1794–1805. h ps://doi.o g/10.1111/ wb.12992.
Bü ne , O., Jawi z, J.W., Bi k, S., Bo cha d , D., 2022. Why was ewa e ea men ails
o p o ec s eam ecosys ems in Eu ope. Wa e Res. 217, 118382 h ps://doi.o g/
10.1016/j.wa es.2022.118382.
Chase, J.M., 2010. S ochas ic communi y assembly causes highe biodi e si y in mo e
p oduc i e en i onmen s. Science 328, 1388–1391. h ps://doi.o g/10.1126/
science.1187820.
Collye , G., Pe kins, D.M., Pe sch, D.K., Siquei a, T., Sai o, V., 2023. Land-use
in ensi ica ion sys ema ically al e s he size s uc u e o aqua ic communi ies in he
Neo opics. Global Change Biol. h ps://doi.o g/10.1111/gcb.16720.
Combe , S.D.W., Ga dne , M.J., Ansell, L., Ello , B., 2022. Assessing he impac o
was ewa e ea men wo ks e luen on downs eam wa e quali y. Sci. To al
En i on. 845, 157284 h ps://doi.o g/10.1016/j.sci o en .2022.157284.
C oss, W.F., Bens ead, J.P., F os , P.C., Thomas, S.A., 2005. Ecological s oichiome y in
eshwa e ben hic sys ems: ecen p og ess and pe spec i es. F eshw. Biol. 50,
1895–1912. h ps://doi.o g/10.1111/j.1365-2427.2005.01458.x.
C oss, W.F., Bens ead, J.P., Rosemond, A.D., B uce Wallace, J., 2003. Consume - esou ce
s oichiome y in de i us-based s eams. Ecol. Le . 6, 721–732. h ps://doi.o g/
10.1046/j.1461-0248.2003.00481.x.
de Guzman, I., Elosegi, A., on Schille , D., 2023. Da ase : T ea ed and Highly Dilu ed,
bu Was ewa e S ill Impac s Di e si y and Ene gy Fluxes o F eshwa e Food Webs.
h ps://doi.o g/10.6084/m9. igsha e.21185224. 1.
Demi, L.M., Bens ead, J.P., Rosemond, A.D., Mae z, J.C., 2018. Li e P con en d i es
consume p oduc ion in de i us-based s eams spanning an expe imen al N:P
g adien . Ecology 99, 347–359. h ps://doi.o g/10.1002/ecy.2118.
Díez, J., Elosegi, A., Chau e , E., Pozo, J., 2002. B eakdown o wood in he Agüe a
s eam. F eshw. Biol. 47, 2205–2215. h ps://doi.o g/10.1046/j.1365-
2427.2002.00965.x.
Dodds, W., Smi h, V., 2016. Ni ogen, phospho us, and eu ophica ion in s eams. Inl.
Wa e s 6, 155–164. h ps://doi.o g/10.5268/IW-6.2.909.
Downes, B.J., Ba mu a, L.A., Fai wea he , P.G., Fai h, D.P., Keough, M.J., Lake, P.S.,
Maps one, B.D., Quinn, G.P., 2002. Moni o ing Ecological Impac s. Camb idge
Uni e si y P ess.
Ecke , R.A., Hal o son, H.M., Kuehn, K.A., Lamp, W.O., 2020. Mac oin e eb a e
communi y pa e ns in ela ion o lea -associa ed pe iphy on unde con as ing ligh
and nu ien condi ions in headwa e s eams. F eshw. Biol. 65, 1270–1287. h ps://
doi.o g/10.1111/ wb.13473.
EEA, 2021. U ban Was e Wa e T ea men in Eu ope. Eu opean En i onmen Agency.
E ans-Whi e, M.A., Hal o son, H.M., 2017. Compa ing he ecological s oichiome y in
g een and b own ood webs–a e iew and me a-analysis o eshwa e ood webs.
F on . Mic obiol. 8, 1184. h ps://doi.o g/10.3389/ micb.2017.01184/ ull.
Fe ei a, V., Cas agney ol, B., Ko iche a, J., Gulis, V., Chau e , E., G aça, M. a S., 2014.
A me a-analysis o he e ec s o nu ien en ichmen on li e decomposi ion in
s eams. Biol. Re . 90, 669–688. h ps://doi.o g/10.1111/b .12125.
Flo es, L., Banjac, Z., Fa ´
e, M., La a˜
naga, A., Mas-Ma í, E., Mu˜
noz, I., Ba cel´
o, D.,
Elosegi, A., 2014. E ec s o a ungicide (imazalil) and an insec icide (diazinon) on
s eam ungi and in e eb a es associa ed wi h li e b eakdown. Sci. To al En i on.
476, 532–541.
Gauzens, B., Ba nes, A., Giling, D.P., o he s, 2019. luxweb: an R package o easily
es ima e ene gy luxes in ood webs. Me hods Ecol. E ol. 10, 270–279.
Ge gs, A., Kulka ni, D., P euss, T.G., 2015. Body size-dependen oxicokine ics and
oxicodynamics could explain in a- and in e species a iabili y in sensi i i y.
En i on. Pollu . 206, 449–455. h ps://doi.o g/10.1016/j.en pol.2015.07.045.
Gonz´
alez, J.M., de Guzm´
an, I., Elosegi, A., La a˜
naga, A., 2023. Te ia y was ewa e
ea men combined wi h high dilu ion a es ails o elimina e impac s on ecei ing
s eam in e eb a e assemblages. Sci. To al En i on. 859, 160425 h ps://doi.o g/
10.1016/j.sci o en .2022.160425.
G ay, C., Figue oa, D.H., Hudson, L.N., Ma, A., Pe kins, D., Woodwa d, G., 2015. Joining
he do s: an au oma ed me hod o cons uc ing ood webs om compendia o
published in e ac ions. Food Webs 5, 11–20.
G uene , U., Ca , G., Mo in, A., 2007. Reducing he cos o ben hic sample p ocessing
by using sie e e en ion p obabili y models. Hyd obiologia 589, 79–90.
Haase, P., D. E. Bowle , N. J. Bake , and o he s. The eco e y o Eu opean eshwa e
biodi e si y has come o a hal . Na u e.
I. de Guzman e al.
Jou nal o En i onmen al Managemen 345 (2023) 118510
9
Hamdhani, H., Eppehime , D.E., Bogan, M.T., 2020. Release o ea ed e luen in o
s eams: a global e iew o ecological impac s wi h a conside a ion o i s po en ial
use o en i onmen al lows. F eshw. Biol. 65, 1657–1670. h ps://doi.o g/10.1111/
wb.13519.
Hilleb and, H., Donohue, I., Ha pole, W.S., o he s, 2020. Th esholds o ecological
esponses o global change do no eme ge om empi ical da a. Na . Ecol. E ol. 4,
1502–1509. h ps://doi.o g/10.1038/s41559-020-1256-9.
Isbell, F., Calcagno, V., Hec o , A., o he s, 2011. High plan di e si y is needed o
main ain ecosys em se ices. Na u e 477, 199–202. h ps://doi.o g/10.1038/
na u e10282.
Jos , L., 2006. En opy and di e si y. Oikos 113, 363–375. h ps://doi.o g/10.1111/
j.2006.0030-1299.14714.x.
Ju igny-Khena ou, N.P.D., Piggo , J.J., A kinson, D., Zhang, Y., Macaulay, S.J., Wu, N.,
Ma haei, C.D., 2021. Impac s o mul iple an h opogenic s esso s on s eam
mac oin e eb a e communi y composi ion and unc ional di e si y. Ecol. E ol. 11,
133–152. h ps://doi.o g/10.1002/ece3.6979.
Ku wadka , S., Dane, J., Kanel, S.R., Nadagouda, M.N., Cawd ey, R.W., Ambade, B.,
S uckho , G.C., Wilkin, R., 2022. Pe - and poly luo oalkyl subs ances in wa e and
was ewa e : a c i ical e iew o hei global occu ence and dis ibu ion. Sci. To al
En i on. 809, 151003 h ps://doi.o g/10.1016/j.sci o en .2021.151003.
Lang, B., Ehnes, R.B., B ose, U., Rall, B.C., 2017. Tempe a u e and consume ype
dependencies o ene gy lows in na u al communi ies. Oikos 126, 1717–1725.
h ps://doi.o g/10.1111/oik.04419.
Langenhede , S., Be ga, M., ¨
Os man, ¨
O., Sz´
ekely, A.J., 2012. Tempo al a ia ion o
β-di e si y and assembly mechanisms in a bac e ial me acommuni y. ISME J. 6,
1107–1114. h ps://doi.o g/10.1038/ismej.2011.177.
La a˜
naga, A., Basagu en, A., Elosegi, A., Pozo, J., 2009. Impac s o Eucalyp us globulus
plan a ions on A lan ic s eams: changes in in e eb a e densi y and sh edde ai s.
Fundam. Appl. Limnol. 175, 151–160. h ps://doi.o g/10.1127/1863-9135/2009/
0175-0151.
Lo eau, M., Ba bie , M., Filo as, E., o he s, 2021. Biodi e si y as insu ance: om concep
o measu emen and applica ion. Biol. Re . 96, 2333–2354. h ps://doi.o g/
10.1111/b .12756.
Ma con, E., H´
e aul , B., 2015. En opa : an R package o measu e and pa i ion
di e si y. J. S a . So wa e 67, 1–26. h ps://doi.o g/10.18637/jss. 067.i08.
McCu chan, J.H.J., Lewis, W.M.J., 2002. Rela i e impo ance o ca bon sou ces o
mac oin e eb a es in a Rocky Moun ain s eam. Limnol. Oceanog . 47, 742–752.
h ps://doi.o g/10.4319/lo.2002.47.3.0742.
Mendigu en, O., Vegas, X., Felipe, A., Rod iguez, S., Gimeno, E., Gil, E.M., Llo ca, J.,
Manzanos, A., 2021. Red de seguimien o del es ado químico de los íos de la CAPV.
Campa˜
na 2020.
Me smann, O., T au mann, H., S eue , D., Bo nkamp, B., 2018. T uncno m. T unca ed
No mal Dis ibu ion.
Meye , E., 1989. The ela ionship be ween body leng h pa ame e s and d y mass in
unning wa e in e eb a es. A ch. Hyd obiol. 117, 191–203.
Mo , J.-R., Dol´
edec, S., Acu˜
na, V., Saba e , S., Mu˜
noz, I., 2019. In e eb a e communi y
esponses o u ban was ewa e e luen pollu ion unde di e en hyd o-
mo phological condi ions. En i on. Pollu . 252, 483–492. h ps://doi.o g/10.1016/
j.en pol.2019.05.114.
Mulde , C., Else , J.J., 2009. Soil acidi y, ecological s oichiome y and allome ic scaling
in g assland ood webs. Global Change Biol. 15, 2730–2738. h ps://doi.o g/
10.1111/j.1365-2486.2009.01899.x.
O , D., Digel, C., Rall, B.C., Ma aun, M., Scheu, S., B ose, U., 2014. Uni ying elemen al
s oichiome y and me abolic heo y in p edic ing species abundances. Ecol. Le . 17,
1247–1256. h ps://doi.o g/10.1111/ele.12330.
Pascual-Beni o, M., Balles ´
e, E., Monle´
on-Ge ino, T., U mene a, J., Blanch, A.R., Ga cía-
Alja o, C., Lucena, F., 2020. Impac o ea ed sewage e luen on he bac e ial
communi y composi ion in an in e mi en medi e anean s eam. En i on. Pollu .
266, 115254 h ps://doi.o g/10.1016/j.en pol.2020.115254.
Pe al a-Ma a e , I., Possel , M., Pe kins, D.M., Robe son, A.L., 2019. Mapping mic o-
pollu an s and hei impac s on he size s uc u e o s eambed communi ies. Wa e
11, 2610. h ps://doi.o g/10.3390/w11122610.
Pe eda, O., on Schille , D., Ga cía-Baque o, G., Mo , J.-R., Acu˜
na, V., Saba e , S.,
Elosegi, A., 2021. Combined e ec s o u ban pollu ion and hyd ological s ess on
ecosys em unc ions o Medi e anean s eams. Sci. To al En i on. 753, 141971
h ps://doi.o g/10.1016/j.sci o en .2020.141971.
Pe eda, O., Solagais ua, L., A is ain, M., de Guzm´
an, I., La a˜
naga, A., on Schille , D.,
Elosegi, A., 2020. Impac o was ewa e e luen pollu ion on s eam unc ioning: a
whole-ecosys em manipula ion expe imen . En i on. Pollu . 258, 113719 h ps://
doi.o g/10.1016/j.en pol.2019.113719.
Pe chey, O.L., Belg ano, A., 2010. Body-size dis ibu ions and size-spec a: uni e sal
indica o s o ecological s a us? Biol. Le . 6, 434–437. h ps://doi.o g/10.1098/
sbl.2010.0240.
R Co e Team, 2019. R: A Language and En i onmen o S a is ical Compu ing.
Rocca o, P., 2018. T ea men p ocesses o municipal was ewa e eclama ion: he
challenges o eme ging con aminan s and di ec po able euse. Cu . Opin. En i on.
Sci. Heal. 2, 46–54. h ps://doi.o g/10.1016/j.coesh.2018.02.003.
Sco , J.T., Back, J.A., Taylo , J.M., King, R.S., 2008. Does nu ien en ichmen decouple
algal–bac e ial p oduc ion in pe iphy on? J. No h Am. Ben hol. Soc. 27, 332–344.
h ps://doi.o g/10.1899/07-108.1.
Shi, X., Qin, T., Nie, H., Weng, B., He, S., 2019. Changes in majo global i e discha ges
di ec ed in o he ocean. In . J. En i on. Res. Publ. Heal h 16, 1469. h ps://doi.o g/
10.3390/ije ph16081469.
Socola , J.B., Gil oy, J.J., Kunin, W.E., Edwa ds, D.P., 2016. How should be a-di e si y
in o m biodi e si y conse a ion? T ends Ecol. E ol. 31, 67–80. h ps://doi.o g/
10.1016/j. ee.2015.11.005.
Soe ae , K., Kones, J.K., 2014. Package Ne Indices, Ne wo k Indices and Food Web
Desc ip o s in.
Solagais ua, L., de Guzm´
an, I., Ba ado, M., o he s, 2018. Tes ing was ewa e ea men
plan e luen e ec s on mic obial and de i i o e pe o mance: a combined ield and
labo a o y expe imen . Aqua . Toxicol. 203 h ps://doi.o g/10.1016/j.
aqua ox.2018.08.006.
S o els, R.J., Ka be, S., Pa e son, R.A., 2003. Leng h-mass models o some common New
Zealand li o al-ben hic mac oin e eb a es, wi h a no e on wi hin- axon a iabili y
in pa ame e alues among published models. N. Z. J. Ma . F eshw. Res. 37,
449–460. h ps://doi.o g/10.1080/00288330.2003.9517179.
Tache , H., Richoux, P., Bou naud, M., Usseglio-Pola e a, P., 2010. In e ´
eb ´
es D’eau
Douce: Sys ´
ema ique, Biologie, ´
Ecologie. CNRS ´
edi ions Pa is.
Taddei, A., R¨
as¨
anen, K., Bu don, F.J., 2021. Size-dependen sensi i i y o s eam
amphipods indica es popula ion-le el esponses o chemical pollu ion. F eshw. Biol.
66, 765–784. h ps://doi.o g/10.1111/ wb.13677.
Thompson, Ross M., Dunne, Jenni e A., Woodwa d, Guy, 2012. F eshwa e ood webs:
Towa ds a mo e undamen al unde s anding o biodi e si y and communi y
dynamics. F eshw. Biol. 57 (7), 1329–1341. h ps://doi.o g/10.1111/j.1365-
2427.2012.02808.x.
T ebilco, R., Baum, J.K., Salomon, A.K., Dul y, N.K., 2013. Ecosys em ecology: size-
based cons ain s on he py amids o li e. T ends Ecol. E ol. 28, 423–431. h ps://
doi.o g/10.1016/j. ee.2013.03.008.
T ubina, M.R., Vo obeichik, E.L., 2012. Se e e indus ial pollu ion inc eases he
β-di e si y o plan communi ies. Dokl. Biol. Sci. 442, 17–19. h ps://doi.o g/
10.1134/S0012496612010036.
Vasilachi, I., Asiminicesei, D., Fe u, D., Ga ilescu, M., 2021. Occu ence and a e o
eme ging pollu an s in wa e en i onmen and op ions o hei emo al. Wa e 13,
181. h ps://doi.o g/10.3390/w13020181.
Wang, M., Xu, X., Wu, Z., o he s, 2019. Seasonal pa e n o nu ien limi a ion in a
eu ophic lake and quan i a i e analysis o he impac s om in e nal nu ien
cycling. En i on. Sci. Technol. 53, 13675–13686. h ps://doi.o g/10.1021/acs.
es .9b04266.
Wei e e, M., Al enbu ge , R., Anlange , C., o he s, 2021. Disen angling mul iple chemical
and non-chemical s esso s in a lo ic ecosys em using a longi udinal app oach. Sci.
To al En i on. 769, 144324 h ps://doi.o g/10.1016/j.sci o en .2020.144324.
Whi e, E.P., E nes , S.K.M., Ke kho , A.J., Enquis , B.J., 2007. Rela ionships be ween
body size and abundance in ecology. T ends Ecol. E ol. 22, 323–330. h ps://doi.
o g/10.1016/j. ee.2007.03.007.
I. de Guzman e al.
