A graphical presentation of WQ datain time and space

Dr Peter G Stoks

RIWA/IAWR

Association of Rhine Water works

Dutch nat’l level: RIWAInternat’l Rhine basin level: IAWR

Mission: Source water quality should allow drinking water production using simple treatment only!

IAWR

Umbrella organizationof 3 Associations

RIWA: Netherlands

ARW: lower Germany

AWBR: upstream Germany, Switzerland

120 utilities30 million consumers

RIWA WQ monitoring network(part of IAWR basinwide network)

• Cooperation with Nat’l Dutch and German water authorities– Harmonized program (WQ variables, methods, data

exchange,…)

• Five locations in the Dutch part of the Rhine basin – German-Dutch border, intake sites

• Trend detection and compliance testing– “legal standards” & “emerging contaminants”

– Chemical & biological

WQMN Data presentation

• Historically as annual WQ reports– Text, graphs, tables

– Executive summaries containing demands and/or

recommendations

• Shift observed in target group’s WQ background

Need for simple WQ overview at a glance

Most current graphs plot 2 or 3 dimensions of a single WQ variable

- change over time at given location

- change over time in space

2 dimensions: conc versus timeMTBE at NL / G border

0

2

4

6

8

10

12

Jan 02 Jul 02 Feb 03 Aug 03 Mrz 04 Sep 04 Apr 05 Nov 05

MT

BE

in µ

g/L

3 dimensions: conc versus time in spaceA pharmaceutical in the Rhine

Basel

Karlsruhe

Mainz

Köln

Düsseldorf

Lobith

30.11.200317.12.2003

15.01.200410.02.2004

11.03.200405.04.2004

05.05.200404.06.2004

29.06.2004

0

50

100

150

200

250

300

c in

ng

/L

Venlafaxin

Solution:

Color palettes combining locations and variables

• allows for basinwide overview of multiple variables

• color indicates WQ development over time–WQ improvement or deterioration at a glance

• area coverage indicates accuracy of statements–The less data, the less area coverage

Area coverage indicates reliability based on number of data

Variable b

Variable c

Variable d

Variable a

Site 1 Site 2 Site 3 Site 6 Site 7Site 4 Site 5

number

of area

year quarter data

moderate

sufficient

few

< 4

1

1 - 5

> 5

4 - 7

8 - 19

> 20

< 1

WQ data in a given

minimum number of

coverage

inadequate

The procedure - 1

• for 5 consecutive years determine the compliance ratio – classify each year as “Good” (ratio < 0.5), “Moderate”

(ratio 0.5 – 1.0), or “Bad” (ratio > 1.0)

Compliance ratio

max value of variable

rmax= divided by

its WQstandard (or objective)

r max interpretation color

< 0.5 good0.5 - 1.0 moderate

> 1.0 bad

The procedure - 2

• determine trend over the 5 years’ period– reduce total dataset to quarterly averages

– calculate trend & classify as “Good” (negative trend*),

“Stable” (no trend) or “Bad” (positive trend)

*not for oxygen

Trend

determined using a STAT-package (modified linear regression model, 95% conf interval)

trend interpretation color

negative improvementnone stable

positive deterioration

Combining compliance ratio and trendyields Status

Good status:

- compliance ratio “good” and trend stable

- compliance ratio “moderate” and trend improving

Bad status:

- compliance ratio “bad” and trend stable

- compliance ratio “moderate” and trend deteriorating

Moderate status:

- all other combinations

Status on the Rhine(2000-2004 data)

> 4 data

4 – 8 data good

9 – 19 data moderate

> 19 data bad

Lobith NieuwegeinCologne Dusseldorf

Oxygen

Basle Karlsruhe Koblenz

Chloride

AOX

Lead

Diglyme

Iopromide

Diuron

Isoproturon

Summary

Standard violations and trends can be presented- for a single variable at different locations over time

- for several variables at a fixed location over time

Status (=WQ development over 5 year period) can be presented

- for several variables at different locations simultaneously

Summary (continued)

Use primarily for non-expert decision makers

Also: triggers WQ analist about curious events

- software built into database management system