Application of the Ecosystem Model to the Mathematical Simulation of Water Environment Dynamic under Anaerobic State in the Organically Polluted Agricultural Reservoir

Application of the Ecosystem Model to the Mathematical
Simulation of Water Environment Dynamic under Anaerobic
State in the Organically Polluted Agricultural Reservoir
Laboratory of Water Environment Engineering
Graduate School of Bioresourse and Bioenvironment Sciences
Kyushu University
HOANG QUANG DUONG
2nd Master’s course


BACKGROUND


BACKGROUND
Closed Water Body in an Organically Polluted Reservoir
Hot
Low
Transparence

Thermal
Stratification

Anoxic Condition
Increase of Toxic
Substance Sulfide

Increase of PO4P and NH4-N
due to Elution
Cold

Sediment

N

P


BACKGROUND
Closed Water Body in an Organically Polluted Reservoir
Hot

Eutrophication
Low
Transparence

Thermal
Stratification

/>Anoxic
Condition

PO4-P
and NH4-N:
Reason
PO4
Increase
of Toxic
Increase
of PO4Substance
andNH4NH4-N
PP

and
of eutrophication
Sulfide Toxic substance
Ndue to Elution
Sulfide:
Cold
Sediment

N

P

Deterioration
of Water
Environment

It is necessary to
evaluate quantitatively
PO4-P, NH4-N and
sulfide under anaerobic
state


BACKGROUND
Schematic Diagram of Ecosystem Model
reaeration

nitrification

NH4-N

NO2-N
NO3-N

DO

DO

It is necessary to
settling

mortality Phytoplankton
defecation/natural death evaluate quantitatively
respiration

PO4-P, NH4-N and
DO
grazing
sulfide underconsumption
anaerobic
Bottom Sedimentstate

Zooplankton

Elution

excretion

respiration

mineralization

decomposition

DO

POM

settling

PO4-P

mineralization

DOM

It is important to
predict and analyze
PO4-P, NH4-N and
sulfide by using
ecosystem model

denitrification

Photosynthesis

DO


BACKGROUND

of PO4-P and NH4-N

② No simulation of sulfide variation

0.3
PO4-P (mg/l)

However …
In current ecosystem model:
① There are large errors in simulation

Large
error

0.2
0.1
0
4

5

6

7

8

9

Month

blem
o
r
P

Simulation of ecosystem model is not accurate due to
internal load under anaerobic state
It is impossible to evaluate and predict variations of
water quality
d
Nee

To reflect dynamic characteristic of water
quality under anaerobic state

10 11 12


BACKGROUND

My

Study

An one-dimensional vertical ecosystem model
that could simulate correctly NO3-N, NH4-N,
PO4-P and sulfide under anaerobic state
To

Implement

To modify oneTo analyze biochemical
dimensional vertical
processes
based
on
d
To reflect dynamic characteristic
waterby
Nee
ecosystemofmodel
observed data
quality under anaerobic Fortran
state


FIELD OBSERVATION & WATER DYNAMIC
CHARACTERISTICS


INTRODUCTION
No.5 Reservoir

RESEARCH AREA

ch
ar
se y
Re o r

o at
Bi or
ri- ab
Ag L

No.5 Regulation Reservoir:
• Location: Ito campus – Kyushu
University
• Purpose: supply water for cultivation
activities at the downstream
• Maximum depth: 8 m
• Catchment area: 31.3 ha
• Surface area: 19,300 m2
• Total storage capacity: 63,000 m3
• In summer, anoxic condition occurs
due to heavy organic pollution
/>

FIELD OBSERVATION

Location: Fix point at the center of the
No.5 Reservoir
Period:
2015/04/01 – 2015/12/09

Classification
Outdoor
analysis
Indoor
analysis

Water quality items

Notes

Transparence
Water temperature, DO, ORP, 0.5 m
etc
interval
Chl.a, NO3-N, NH4-N, PO4- 1 m
P, Sulfide, SO42-, etc
interval


WATER QUALITY CHARACTERISTIC

DO (mg/l)

WT (oC)

Ds (m)

6
4

Average of Transparency
is 2.0 m

Ave = 2.0 m

2
0
40
30
20
10
0
6

:0m
:8m

DO was
exhausted

4
2
0
4

5

6

7

8

9 10 11 12

Month

Fig. Transparency, Water temperature
and DO in 2015

Strong thermal
stratification (ca. 15-200C
difference between the
surface and the bottom)


WATER QUALITY CHARACTERISTIC
Ave = 2.0 m

DO (mg/l)
2 4 6

20

8

0

0

40
1
30
2
20

3
10
4
0
56

1
2
Depth (m)

DepthWT
(m)

(oC)

00

DO (mg/l)

NH4-N (mg/l)
0 0.5 1 1.5 2

:0m
:8m

3
4
5

Average of Transparency

P O4-P (mg/l) is 2.0 m Sulfide (µg/l)
0.1

0.2

0

0

0

1

1

5

5

6

6

72

DO6was
7
exhausted

7

7

8

8

8

8

64

0
4

5

6

7

8

9 10 11 12

a

Fig.Month
DO, NH4-N, PO4-P, sulfide on October 7, 2015

Fig. Transparency, Water temperature
and DO in 2015

200 400 600

Strong thermal
2
stratification (ca. 15-200C
3
3
difference
between the
Elution of nutrient
salts
4
4
surface
and the bottom)
and generation
of sulfide
2

Depth (m)

4

Depth (m)

Ds (m)

6


WATER QUALITY CHARACTERISTIC

DO (mg/l)

WT (oC)

Ds (m)

6
4

Average of Transparency
is 2.0 m

Ave = 2.0 m

2
0
40
30
20
10
0
6

:0m

:8m

Strong thermal
stratification (ca. 15-200C
difference between the
surface and the bottom)

4
2
0
4

5

6

7 8 9 10 11 12
Month

Fig. Transparency, Water temperature
and DO in 2015

DO increases when
thermal stratification
disappears


NO3-N (mg/l)

NH4-N (mg/l)

WATER QUALITY CHARACTERISTIC
At water surface, NH4-N
is almost no change.
However, NO3-N
decreases quickly

1
0.5
0
0.3
0.2
0.1
0
4

5

6

7

8

9 10 11 12

Fig. Dissolved inorganic nitrogen (DIN)
at water surface (0m)

Preferential ingestion of

NO3-N by
phytoplanktons


WATER QUALITY CHARACTERISTIC
6
5
4
3
2
1
0

DO (mg/l)

DO (mg/l)

6
5
4
3
2
1
0

DO=0.5

0.4
0.3
0.2

0.1
0

DO=0.5

NO3-N (mg/l)

NO3-N (mg/l)

0.4
0.3
0.2
0.1
0

4

5

6

7

8

9

10 11 12

Fig. DO and NO3-N at 7m

4

5

6

7

8

9

10 11 12

Fig. DO and NO3-N at 8m

Denitrification starts when DO ≈ 0.5 mg/l


WATER QUALITY CHARACTERISTIC

NO3-N (mg/l)

NO 3-N=0.1

1
0
4

5

6

7 8 9
Month

10 11 12

Fig. DO, NO3-N and NH4-N at
7m

DO=0

0.4
0.3
0.2
0.1
0

NH4-N (mg/l)

NH4-N (mg/l)

DO (mg/l)

DO=0

0.4
0.3

0.2
0.1
0
2

5
4
3
2
1
0

NO3-N (mg/l)

DO (mg/l)

6
5
4
3
2
1
0

NO 3-N=0.1

3
2
1
0

4

5

6

7 8 9
Month

10 11 12

Fig. DO, NO3-N and NH4-N at
8m

NH4-N starts to increase when NO3-N ≈ 0.1 mg/l


DO (mg/l)

DO=0

NO3-N (mg/l)

0.4
0.3
0.2
0.1
0

NO 3-N=0

800
600
400
200
0

: PO 4-P
: Sulfide

0.1
0
4

5

6

7 8 9 10 11 12
Month

Fig. DO, NO3-N and Sulfide at
7m

DO=0

NO 3-N=0

0.3

800

600
400
200
0

: PO 4-P
: Sulfide

0.2
0.1
0
4

5

6

7 8 9 10 11 12
Month

Fig. DO, NO3-N and Sulfide at
8m

PO4-P and Sulfide starts to increase when NO3-N ≈ 0 mg/l

Sulfide (µg/l)

0.5
0.4
0.3

0.2
0.1
0
0.2

5
4
3
2
1
0

PO4-P (mg/l)

6
5
4
3
2
1
0

Sulfide (µg/l)

PO4-P (mg/l)

NO3-N (mg/l)

DO (mg/l)

WATER QUALITY CHARACTERISTIC


ONE-DIMENSIONAL VERTICAL
ECOSYSTEM MODEL


MODEL DEVELOPMENT
Input by
diffusion flux

Input by
incident light flux

=

A( z − ∆z 2 )

Internal production or
consumption
∆z

A( z)
A( z + ∆z 2 )

Output by
diffusion flux

 Turbulent diffusion equation for water temperature:

 Turbulent diffusion equation for water qualities
=

Output by
incident light flux

: Water area at water depth z (m2)
: Vertical turbulent diffusion coefficient (cm2/s)
: Water depth (m)
: Density of the water (kg/m3)
: Water temperature (oC)
: Specific heat of the water (J/kg/K)
: Heat flux of incident light (J/m2)
: Water quality item (mg/l)


MODEL DEVELOPMENT
Process
Phytoplankton
ingestion

Previous Model
NH4-N and NO3-N
were determined by
the existing ratio

Modified Model
Preferential ingestion of NO3-N

Denitrification

DO ≤ x
x was determined by Denitrification under DO ≤ 0.5 mg/l
trial and error

Elution of
nutrient salts

Limiting condition:
Elution depends on
NH4-N: DO = 0 and NO3-N <
only DO
0.1
PO4-P: DO = 0 and NO3-N = 0

Temporal
Zero-order reaction
variation of
There is no
Limiting condition: DO = 0 and
sulfide due to simulation of sulfide
NO3-N = 0
sulfate reduction


MODEL DEVELOPMENT
• Meteorology data: Observation data
(10 minutes)
• Transparency: Observation data
• Inflow data: Lack of inflow data

• Calculation time: April to December
Inflow data from box culvert cannot
measure due to deposition of sediment

Fig. Current situation
of box culvert in 2015


MODEL DEVELOPMENT
• Meteorology data: Observation data
(10 minutes)
• Transparency: Observation data
• Inflow data: Lack of inflow data
• Calculation time: April to December
: Rainfall

7

0

6

100

5

200

4
4

5

6

7

8 9
Month

10 11 12

Fig. DOC and rainfall in 2015

300

Rainfall (mm)

DOC (mg/l)

: DOC

Fig. Current situation
of box culvert in 2015

DOC varies much when
heavy rainfall occurs


MODEL DEVELOPMENT

• Meteorology data: Observation data
(10 minutes)
• Transparency: Observation data
• Inflow data: Lack of inflow data
• Calculation time: April to December

DOC (mg/l)

7
6
5

200

4
4

5

6

7

8 9
Month

10 11 12

Fig. DOC and rainfall in 2015

300

Rainfall (mm)

It is necessary
2
: Rainfall to divide into Fig.
Current situation
0 including
calculation periods
of box culvert in 2015
spring-summer and
summer100
winter based on heavy rainfall

: DOC

DOC varies much when
heavy rainfall occurs


SIMULATION RESULT
: Simulation

40
20

NO3-N (mg/l)

0.2

0

0.02
0
4

5

6

7

8 9
Month

10 11 12

Fig. Simulation of Chl.a and nutrient
salts in the previous model at 0m

40
20
0
0.5
0.4
0.3
0.2
0.1
0

NH4-N (mg/l)

0.04

0.4

: Simulation

60

0.6

PO4-P (mg/l)

NH4-N (mg/l)

0
0.5
0.4
0.3
0.2
0.1
0
0.6

: Observation
Chl.a (µg/l)

60

PO4-P (mg/l)

NO3-N (mg/l)

Chl.a (µg/l)

: Observation

0.04

0.4
0.2
0

0.02
0
4

5

6

7

8 9
Month

10 11 12

Fig. Simulation of Chl.a and nutrient

salts in the modified model at 0m


SIMULATION RESULT
: Simulation

: Observation
Chl.a (µg/l)

60
40
20

0.04

NO3-N (mg/l)

40
20
0
0.5
0.4
0.3
0.2
0.1
0

Better
Simulation

0.4
0.2
0

0.02
0
4

5

6

7

8 9
Month

10 11 12

Fig. Simulation of Chl.a and nutrient
salts in the previous model at 0m

NH4-N (mg/l)

0.6

: Simulation

60

0.6

PO4-P (mg/l)

NH4-N (mg/l)

0
0.5
0.4
0.3
0.2
0.1
0

PO4-P (mg/l)

NO3-N (mg/l)

Chl.a (µg/l)

: Observation

0.04

0.4
0.2
0

0.02
0

4

5

6

7

8 9
Month

10 11 12

Fig. Simulation of Chl.a and nutrient
salts in the modified model at 0m