Trents rural strategy Vertical flow wetlands for full wastewater - - PowerPoint PPT Presentation

www.cranfield.ac.uk

Bringing to life Severn Trent’s rural strategy –

Vertical flow wetlands for full wastewater treatment

• Dr. Yadira Bajón Fernández

29th January 2019

Motivation - Rural strategy

2

• Low carbon footprint (aeration / energy)
• Reduced intervention
• No/low sludge tankering
• Societal value to local communities
• Innovation

> 1000 sewage treatment plants ~ 750 small works (<2000 PE)

Chosen technology

3

• Proven technology (France)
• Simple process that works
• Lowest totex solution
• 10-12 years sludge accumulation
• Low / no energy
• Low maintenance

4

Chosen technology

• 1st stage (BOD, TSS removal): 3 beds, 3.5 days feeding / 1week resting
• 2nd stage (Nitrification, polishing): 2 beds, 1 week feeding / 1week resting

941 p.e. 1st stage 1.2 m2/p.e. 2nd stage 0.8 m2/p.e.

CONSENTS:

• TSS 50 mg/L
• BOD 30 mg/L
• Ammonium 15 mg/L

Main layer >30cm Transition layer: 10-20cm Drainage: 10-20cm 0.25 <d10 <0.40 mm sand 3-10 mm gravel 20-40 mm gravel 2-8 mm gravel 5-20 mm gravel 20-40 mm gravel

5

New inlet works Pumping to 1st stage 3 x 1st Stage vertical filters Designed @ 1.2 m2/PE – 0.37 m/d 3.5 d feed / 7 d rest Pumped New Siphon chamber 2 x 2nd Stage vertical filters Designed @ 0.8 m2/PE 7 d feed / 7 d rest Gravity New Storm system with existing 2 Storm tanks

Hulland Ward – New assets

6

Red: Expected results for two stage vertical flow reed beds based on a study of 169 full scale sites in France (Paing et

• al. 2015).

* Based on TKN **excluding storm events (HL > 0.6 m·d-1)

Treatment performance

1st year of operation – 24 h composite samples

7

Commissioning challenges

Winter 2015/16 – Attempts to operate side at FFT (11 l/s)

• 1st stage reduced infiltration rate lead to ponding
• Ponding caused overflows with bypass of first stage. Clog

second stage

• 2nd stage switched off – first stage + nSAF winter 2016/17
• Aerated second stage Dec. 2016

8

Commissioning chellenges

Reasons:

• Combined sewers + UK rainfall pattern VS in France. 1st stage undersized
• Not mature beds, reeds not developed. Slow sludge mineralisation
• Not critical minimium resting period
• Unsuitable sand on top of 2nd stage

9

Lessons learnt 1: Sludge layer mineralisation

Time in

• peration

Sludge accumulation (cm) Accumulation Rate (cm/year) Dry matter (%) OM (% of DM) IR mm.min-1 December 2015- January 2016 RB1

6 months 2-5 cm 8.1 76 <0.52

RB3

3 months 2-3 cm 6.5 74 <0.39

March 2016 RB1

9 months 3-5 cm 4-7 cm 12.8 67.4 <0.35

RB3

6 months 3-5 cm (estimation) 11.5 64.5 <0.34

Literature Arias 2013 Site A France (Molle et al. 2006) Site B France (Molle et al. 2006) (Boutin and Liénard 2003) Site C France Molle et al. 2006 Site D France

1 year 8 years 14 years 1 year 4-7 cm 22.5cm 1 cm 0.4-0.8 cm year 1.6 cm/year 1 cm/year 30.2 (mixed sludge)21.8 37.6 (mixed sludge) 49.2 ≈0.6 (winter) >3 >1.8 >21

Some images from August/September 2015

10

Lessons learnt 2: Hydraulic load distribution

• More than 48 h of continuous ponding
• Recommended water free surface 8.5 h per day
• Higher sensitivity to periodicity than intensity of hydraulic overloads
• Adaption to local conditions needed

10 20 30 40 50 60 0-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1 1-1.2 1.2-1.4 1.4-1.6 1.6-1.8 1.8-2 >2

% time Hydraulic load (m.d-1 ) Hulland Ward Site A France 2011

Design value 0.37m/day Insufficient reoxygenation affecting nitrification Retardation sludge mineralisation

10 20 30 40 50 60 0-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1 1-1.2 1.2-1.4 1.4-1.6 1.6-1.8 1.8-2 >2

% time Hydraulic load (m.d-1 ) Hulland Ward Site A France 2011

1.2 m2·pe 1.7 m2·pe

What do we need to achieve?

Re-oxygenation of filter body -> nitrification of absorbed ammonium Sludge mineralisation -> IR

12

Lessons Learnt 3: Sand specification

Brix and Arias, 2005, The use of vertical flow constructed wetlands for on-site treatment of domestic wastewater: New Danish guidelines. Ecological Engineering, 25: 491-500

13

5 10 15 20 25 50 100 150 200 250 300 350 .Ammonia concentration in final effluent (mg/L) Day operation (3 year) 9 a.m. 9 p.m.

Lessons Learnt 4: It needs time

• Feeding/resting cycle impacts ammonia profile
• Critical resting time

14

Lessons Learnt 4: It needs time

Sequence point* Bed 1 Bed 2 Bed 3

Beginning of the feeding period End of the feeding period Beginning of the resting day Middle of the resting period End of the resting period

2nd day 1st day 2nd day 7th day 1st day 7th day 7th day 7th day 7th day 1st day 7th day 14th day 14th day 14th day

Lessons learnt from 1st UK system

15

• 1. Careful control of hydraulic loading until system is mature
• 2. Protect minimum resting period (re-oxygenation and sludge layer mineralisation)
• 3. Careful sizing of first stage (Design on hydraulic profile)
• 4. Careful media selection
• 5. 2nd stage aeration is an effective mitigation. But still working towards fully passive
• 6. Minimal intervention after commissioning
• 7. Allow 1-2 years for commissioning. We are working to accelerate maturation

Demonstrated effective treatment. Very promising technology for future UK implementation

Would you install this scheme in other sites? Yes, but with lessons learnt

16

Current work: sludge layer

Vision: control sludge layer mineralisation and characteristics (IR) by modifying

• perational procedures (rest/feed cycles)

17

Current work: sludge layer

1) Sludge layer properties: TS, VS, height…

18

Current work: sludge layer

2) Preferential flow pathways

Cracks

form in result of sludge dewatering during resting period

Voids formed by plants

Occur around stems, in root zone and formed around plants debris/

Voids formed by animals

Occur in result of animal (e.g. worms) activity.

Can we accelerate crack formation?

Conclusions

20

• 1. Very promising technology for UK implementation, allowing passive or

very low energy treatment with minimum intervention and deslugding

• 2. Careful control of hydraulic loading, specially until system is mature
• 3. Currently working to engineer sludge layer mineralisation in order to

accelerate maturation and maintain mineralisation

Would you install this scheme in other sites? Yes, but with lessons learnt

Any questions?

Dan Cunliffe, Alex Cook, Tomas Jordan, Ricky Shepherd, Ivan Blanco, Andy Richards, Pete Vale, Richard Smith, Bruce Jefferson, Gaby Dotro, Fred Coulon, Ledicia Pereira, Antonio Labella de Tomás, Olha Khomenko