Further
improving
the
quality
of
surface
water
is
becoming
more
difficult
after
control
main
point-sources,
especially
in
complex
pollution
area
with
mixed
industrial
and
agricultural
productions,
whereas
source
apportionment
might
be
key
to
quantify
different
sources
developing
some
effective
measures.
In
this
study,
a
technical
framework
for
based
on
three-dimensional
fluorescence
microbial
traceability
model
developed.
Based
screening
environmental
factors
their
spatiotemporal
characteristics,
potential
have
been
tentatively
identified.
Then,
are
further
tested
analysis
excitation-emission
matrix
(EEM)
similarity
components
sources.
At
same
time,
correlation
between
species
constructed
by
analyzing
characteristics
composition
response
factors.
Therefore,
quantified
using
PCA-APCS-MLR,
Fast
Expectation-maximization
Microbial
Source
Tracking
(FEAST),
Bayesian
community-wide
culture-independent
tracking
(SourceTracker).
Overall,
results
these
three
methods
comparable
extent,
shows
accurate
identification
ability
environments
than
better
reflects
differences
at
various
points.
FEAST
exhibits
sensitive
shorter
calculation
time
SourceTracker,
thus
used
guide
precise
regional
control.
To
explore
how
environmental
factors
affect
the
structure
of
plankton
in
urban
rivers,
we
analyzed
Caowangjing
River,
an
river
that
passes
through
Wuxi,
to
survey
water
and
population
different
seasons.
We
identified
103
phytoplankton
species
belonging
eight
phyla,
with
Chlorophyta,
Bacillariophyta,
Cyanobacteria
being
dominant
groups.
A
total
45
zooplankton
belonged
three
classes,
Rotifera
class.
Phytoplankton
density
was
highest
autumn,
followed
by
spring,
lowest
summer.
biomass,
along
exhibited
seasonal
declines.
The
average
values
Shannon–Wiener
index,
Pielou’s
evenness
Margalef
richness
index
were
3.58
±
0.50,
0.72
0.76,
2.03
0.31,
respectively,
indicating
River
mildly
polluted
based
on
a
quality
evaluation.
Redundancy
analysis
showed
turbidity,
temperature,
ammonia
nitrogen
are
key
community
distribution,
while
permanganate
distribution.
A
technical
framework
for
pollution
tracing
based
on
three-dimensional
fluorescence
and
microbial
traceability
model
is
proposed,
which
can
help
in
identifying
calculating
the
contributions
of
multiple
sources.
The
dissolved
organic
matter
(DOM)
analyzed
by
excitation-emission
matrix
(EEM)
coupled
with
parallel
factor
analysis
(PARAFAC),
Tucker
congruence
coefficient
fluorescent
components
between
surface
water
sources
compared,
thus
potential
are
listed.
Water
quality
parameters
further
using
Spearman
correlation
to
verify
main
sources:
aquaculture,
mechanical,
chemical,
textile
wastewater.
At
same
time,
influence
environmental
factors
microorganisms
determined
a
random
forest
analysis,
source
biomarkers
identified
linear
discriminant
(LEfSe).
Network
then
used
reveal
relationship
biomarkers,
impact
different
proposed
biomarkers-based
primary
coordinate
(PCoA).
Finally,
apportionment
quantified
PCA-APCS-MLR,
Fast
Expectation-maximization
Microbial
Source
Tracking
(FEAST),
Bayesian
community-wide
culture-independent
tracking
(SourceTracker).
shows
more
accurate
identification
ability
complex
environments
than
better
reflects
differences
at
various
points.
FEAST
has
sensitive
shorter
calculation
time
SourceTracker.
similar
results
indicate
credibility
identification,
be
guide
precise
regional
control
improve
effectiveness
resources
management.
Further
improving
the
quality
of
surface
water
is
becoming
more
difficult
after
control
main
point-sources,
especially
in
complex
pollution
area
with
mixed
industrial
and
agricultural
productions,
whereas
source
apportionment
might
be
key
to
quantify
different
sources
developing
some
effective
measures.
In
this
study,
a
technical
framework
for
based
on
three-dimensional
fluorescence
microbial
traceability
model
developed.
Based
screening
environmental
factors
their
spatiotemporal
characteristics,
potential
have
been
tentatively
identified.
Then,
are
further
tested
analysis
excitation-emission
matrix
(EEM)
similarity
components
sources.
At
same
time,
correlation
between
species
constructed
by
analyzing
characteristics
composition
response
factors.
Therefore,
quantified
using
PCA-APCS-MLR,
Fast
Expectation-maximization
Microbial
Source
Tracking
(FEAST),
Bayesian
community-wide
culture-independent
tracking
(SourceTracker).
Overall,
results
these
three
methods
comparable
extent,
shows
accurate
identification
ability
environments
than
better
reflects
differences
at
various
points.
FEAST
exhibits
sensitive
shorter
calculation
time
SourceTracker,
thus
used
guide
precise
regional
control.
