Portland’s Cryptosporidium Detections: Investigation and Outcome

ANN LEVY YONE AKAGI SADIE SILKIE

AWWA-PNWS APRIL 27, 2018 Cryptosporidium • Chlorine-resistant pathogenic protozoan

• Sources include human sewage, livestock, wildlife.

• C. hominis and C. parvum cause the vast majority of cases in humans

• PWB’s LT2 Variance granted in 2012; revoked 2017 • Serves approximately 970,000 people • 19 wholesale water districts • 101 million gallons per day average (80 – 160 mgd)

3 Historical Cryptosporidium Monitoring (Raw Water Intake)

2012–2016 2010 No Crypto, Variance No Crypto, Variance Study (2x/week, 26,159L) (4x/week, 10,271L)

2011 2017 1 oocyst 43 oocysts/11,510L 1986–2002 2003–2009 (weekly) (4-5x/week) 88 oocysts/21,700L, inconsistent No Crypto, monthly samples frequency & methods (4,337L)

1990 1995 2000 2005 2010 2015 2020

4 Treatment Variance Conditions Intake Monitoring: Watershed Protection: ◦ Observation Monitoring ◦ Maintain Protections ◦ Two 50-liter samples per week ◦ Monitor Trespass ◦ If Crypto is detected, then Demonstration ◦ Contain Human Waste Monitoring ◦ Annual Watershed Report: ◦ Notifications, Press release ◦ Sampling: ◦ Increased Monitoring For 1 year ◦ Tributary Monitoring ◦ 250 L per week ◦ Scat Monitoring ◦ Variance revoked if >0.075 oocysts / ◦ Inspections: 1000 L ◦ Security, Diversion Pool Fence, Wildlife, Landslides, Sanitary Facilities Types of Cryptosporidium Sampling at PWB • Raw Water Intake • 50-L samples (or 5 x 10-L) • 2-4 x/week • 100-200L/week North Fork • Major Tributaries • 10-50-L samples Reservoir Bull Run River • Monthly 1 • 6-8 rain events/year Reservoir Fir Creek 2 • Wildlife Scat Bull Run • Over 150 samples/year South Fork Lake Types of Watershed Inspections ❑ Watershed Security Patrols ❑ Diversion Pool Fence Inspections ❑ Sanitary Facility Inspections ❑ Wildlife Surveys ❑ Landslide Surveys Timeline of 2017 Crypto Detections and Follow-Up

1/8: 2/13 – 3/15: 11/21 &12/13: Demonstration Bull Run turned off; 9/18-9/19: Extensions of Variance Monitoring Begins 100% groundwater First Flush Revocation Deadlines (5 X 50-L per 5/19: OHA 8/1: OHA rain event week) Revocation Order; Amended 2/1: 3/8: PWB letter to OHA Switch to 4x50-L Revocation 10/9: PWB Clinician re: variance status per week Order; proposed alert extended compliance deadlines schedule and interim measures

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Average Monthly Concentrations (oocysts/L) 0.006 0.010 0.002 0 0 0 0 0 0.002 0.008 0.008 0.008

6 oocysts 11 oocysts 2 oocysts 2 oocysts 8 oocysts 7 oocysts 7 oocysts Cryptosporidium, Turbidity and Main Stem Flow, 2017

3.5 9,000 Crypto, Intake Turbidity, Intake Discharge, Main Stem Bull Run River

8,000 3

7,000 2.5

Turbidity (NTU) Turbidity 6,000 / /

2 5,000 (count/50 L) (count/50

1.5 4,000 Maximum Daily Discharge (cfs) Discharge Daily Maximum

3,000 1

Cryptosporidium Cryptosporidium 2,000

0.5 1,000

0 0 Cryptosporidium, Giardia, Fecal Coliform, and Turbidity, 2017 30 4 Crypto Giardia Turbidity FC

3.5 25 Turbidity Turbidity 3 20 2.5

(count/50 L)/ (count/50 15

2 (NTU)

Giardia 1.5 10 & &

1 mL) (cfu/100 coliform Fecal 5 0.5

Cryptosporidium 0 0 Cryptosporidium Source / Location Investigations (Winter – 2017) Possible Source / Location Investigation Conclusions / Comments

Lab Effect Reviewed relevant lab QC data, etc. Reasonably ruled out

Apparatus / Sample Line Biofilm Flushed line, cleaned / replaced tubing Reasonably ruled out

Diversion Pool Multiple site inspections; camera inspection of intake Unlikely

Reservoirs (obvious signs) Shoreline inspections; camera inspection of North Tower Unlikely

Human Sewage Inspected Bear Creek House facilities Reasonably ruled out Domesticated Animals Checked with Watershed Rangers for recent incidents Reasonably ruled out Likely – wildlife source(s) + Upstream Watershed Expanded monitoring rain event(s) ? ?

? Watershed Surveillance Sampling: Feb-Apr 2017 • Reservoirs • Key Stations • Minor Streams • Biofilm • Scat Reservoir 1 Monitoring: February – April 2017

Main Stem

North Fork

Rodent trapping

Fir Creek Tributary and Reservoir Detections in 2017 Tributaries Reservoirs 1 out of 99 samples positive (1%) 6 out of 61 samples positive (10%)

• Oct. 20, 2017 at South Fork • Feb. 21 at Reservoir 1 and Reservoir 2 • Nov. 6 at Reservoir 1 and Reservoir 2 • Dec. 13 at Reservoir 1 Cryptosporidium Genotyping ▪ EPA Method 1623 only identifies oocysts by microscopy; does not determine viability, infectivity, or species/genotype ▪ Genotyping may provide information on: ▪ Public health significance / potential for human infection ▪ Possible wildlife source(s)

▪ C. hominis and C. parvum cause most human infections (~95%)

▪ Many types carried by wildlife are considered low risk Genotyping – PWB scat and water samples Scat Samples ▪ Analyzed for Cryptosporidium presence by PCR ▪ PCR DNA products are sequenced

Water Samples ▪ Genotyping is attempted on all Crypto-positive water samples ▪ Positive Method 1623 slides are scraped, DNA extracted, and PCR attempted ▪ DNA sequences compared with: ◦ GenBank reference DNA sequences, & ◦ PWB wildlife scat data from Bull Run ▪ Less than 50% success rate (Jan – Mar 2017 = 22%; Sep – Dec 2017 = 0%) Genotypes found in Bull Run Wildlife Scat 2013-2018 Species/Genotype Bull Run Host(s) Total # # in WY 2017 PNW17a Deer mouse (15), Bobcat (1), Deer (1) 17 14 PNW17b Deer mouse (7), (2), Bobcat (1) 10 8 Novel (no match 99.5% or better) Bobcat (3), Deer (1), Deer Mouse (3), Skunk (1) 8 1 Generic (not classified) Deer (3), Bobcat (2), Coyote (1), Deer mouse (1) 7 2 C. andersoni Bobcat, Shrew 5 1 Deer genotype Deer 3 1 C. parvum Deer Mouse, Deer, Elk 3 0 PNW17c Bobcat, Coyote, Deer mouse 3 3 PNW15a Mountain beaver 3 0 C. canis or Coyote 2 0 Bear genotype Bear 2 2 C. ubiquitum Mt. Beaver, Bobcat 2 0 PNW17d Deer mouse 2 2 C. felis Bobcat 1 0 Skunk genotype Snowshoe Hare 1 0 Mink genotype Bobcat 1 0 Genotyping Results: 2017 Water Samples Water DNA Sequencing Results Similar Bull Run Scat References in Scientific Sample Samples Literature

• Novel wildlife type, Jan. 3 None None (or rat) • Closest to Rat Genotype I Cluster PNW17a • Match each other 100% • 13 Deer Mouse and • Belding’s ground Feb. 1 • Unclassified rodent 1 Bobcat from 2017 squirrel & genotype, possibly • 1 Deer sample from • Deer Mouse Feb. 21 Squirrel Genotype IV 2016 • Wild rodent • 2 Rodent from 2014 Cluster PNW17b • 6 Deer Mouse and • Close (99.4%) to Deer Feb 14 2 Coyote from 2017 Deer Mouse Mouse Genotype IV • 1 Bobcat from 2016 • 1 2014 Deer Mouse Cryptosporidium Detections in 2018

3 4,000 3 30

3,500 2.5 2.5 25

3,000

2 (NTU) Turbidity 2 20

Turbidity (NTU) Turbidity 2,500 / /

1.5 2,000 1.5 15 mL) /100

cfu

(count/50 L)/ (count/50 (count/50 L) (count/50 1,500

1 Giardia 1 10 & &

1,000 (cfs) Discharge Daily Maximum Fecal coliform ( coliform Fecal 0.5 0.5 5

Cryptosporidium 500 Cryptosporidium 0 0 0 0

Crypto, Intake Turbidity, Intake Discharge, Main Stem Bull Run River Crypto Giardia Turbidity FC Regulatory Outcome

2017 Crypto concentration >0.075 oocysts/1000 L annual limit for the variance OHA revoked the variance as of December 18, 2017

Bilateral Compliance Agreement ◦ Treatment plan – Filtration by 2027 ◦ Interim control measures and sampling schedule Filtration Decisions Conventional Direct Membrane Slow Sand 10 year schedule • Capacity – April 2018 • Location – May 2018 • Type of filtration – June 2018 Interim Measures

• Continuing with same watershed protections and monitoring • Sampling for Crypto at the intake (100 - 200L/week) • Continuing to gather genotyping data • Monthly, quarterly and annual reports • Public notices, required language • Outreach plan for vulnerable populations

• Public Health Surveillance Implementation Project • Working with EPA, CDC and local health officials Title and End Photos by Roman Johnston Thank You