J. Jarkovský, L. Dušek, J. Rá ek
Institute of Biostatistics and Analyses, Masaryk University, Czech Republic
!
"
#$%
! "#
$ %&' ("#
& ' (
&
) * (
& +
) #
) +
) * , ( - .
& /#0'%*+
& 1
& ) ' /2%) 3
) (
& # 3 ) % 4-+"%/5..
& 4 6
) 0 0
) 708+
) 984
) %:';
& # ( 6
!"
& * ( % 4 %
) <===>?@@@ " (
. " -
. + !#+ .
) ?@@< % ) (
+ . . "
, - - .
) ?@@A>?@@B " $. 1
( . 4 , 6 -
!#+ - -
4 ' ' -
) ( - 6 ( .
& * ( ( 4
*6 / (
#$ %&
$
%
'$"
&
%
#
Expert knowledge (species
properties etc.)
Species abundances
Abiotic data (chemical,
geography, hydrology etc.)
Dynamic abiotic
parameters
Natural heterogeneity
(static parameters)
Pollution data
Measured data
Indices
Community "quality"
(biotic indices,
ecological scores etc.)
Community structure
(diversity indices)
Measured data
A) Calibration data set B) Environmental gradient
Contaminated
sites
Reference (clean)
sites
X1 X2 X3 ...... Xp X1 X2 X3 ...... Xp
X1 X2 X3 ...... Xp
New
problem
Searching
for similarities
X1 X2 X3 ...... Xp
New
problem Searching
for similarities
Gradient calibration
Increasing
contamination
Increasing
disturbance
RISK ASSESSMENT SCENARIORISK ASSESSMENT SCENARIO
! %
River system of the Czech Republic
! %
Abiotic river types
! %
Reference and standard data
(
& 4 6 (
) AC?
) <?@
& /
) ! AC@
&
) D@
& %
) B@><?@
& 4
) <?@
& %
& +
& +
& (
& 2 ( (
# " )! *
+
, -
Data of standard biomonitoirng Reference data
Stability of determination
Taxonomy database
Elimination of species with random occurrence
and problematic determination
Indicative power of taxa
Stability of determination vs.
indicative power
Stable determination and indicative
power
No indicative power
Problems with determination and
indicative power
Elimination of taxaExpert selection of important taxa
Pruning of taxonomy tree
Agreement of
response curves
Agreement of
species traits
Co-occurrence of
species
In the case of agreement of taxa
characteristics, the taxa are united
into so called functional taxa. The
algorithm is repeated until there
are no more branches to unite.
Indicative power
Final taxonomy database
Monitoring
Data analysis
Expert
knowledge
Inputs and
outputs
Splits of
taxonomy
database
( )
BiologicalcommunityIII
Standard dataStandard data ofof monitoringmonitoring
((unknownunknown sitessites))
Reference data
ClassificationClassification intointo referencereference groupsgroups
accordingaccording toto naturalnatural heterogeneityheterogeneity
Definition of reference categories
according to biological communities
DifferenceDifference betweenbetween observedobserved andand
expectedexpected ecologicalecological statestate
Reference model: step IReference model: step I
EnvironmentalEnvironmental
parametersparameters
CommunityCommunity
compositioncomposition
Description of reference categories
= reference model
Environmental
parameters
Communities
Reference model: step IIReference model: step II
I.
II.III.
IV.
V.VI.
PartialPartial
evaluationevaluation AA
PartialPartial
evaluationevaluation BB
Biological community I
BiologicalcommunityIIA, BA, B A, BA, B
FinalFinal
evaluationevaluation ofof
statestate
VII.
Final result is a
single metric with
straightforward
interpretation
based on different
communities,
possibly also on
environmental
evaluation
Natural
heterogeneity
Pollution
I. II. III. IV. V. VI. VII.
Independent problems to solve
"
" *+
Monitoring
Information
Systems
Standard Data
Formats
Data Collecting
XML Formats
Information Systems of
Report Readers
On-line
Generated
Reports
HTTP
E-mail
PDF
Central
Repository
Reporting Wizard
Workflow System
% ,
"
. . ( (
( ( .
" 6
-
Complex system for evaluation of ecological state of surface waters with
standardized procedure of data processing.