1. Monitoring in the Ahupua‘a
Michael Tomlinson
Department of Oceanography

2. Mokupuni (large islands) of Hawaiʻi
(Aliʻi Nui or Head Chief)
Kauaʻi
Niʻihau
Kahoʻolawe
Lānaʻi
Molokaʻi
Maui
Hawaiʻi
Oʻahu

3. Moku of Oʻahu
(Aliʻi ʻai moku)

4. Ahupuaʻa of the Kona Moku
(Konohiki)
Waikīkī

5. “Typical”
Ahupua‘a
Konohiki Oversees
• Water
• Forestry (timber,
clothing)
• Agriculture (kalo loʻi,
breadfruit, etc.)
• Onshore/Nearshore
Fisheries (including
aquaculture)
• Offshore Fisheries

6. Another
depiction of
an ahupuaʻa
(Kamehameha
Schools, 1993)

7. Waikīkī ahupuaʻa then (~1865) . . .
Mānoa Valley from Waikīkī, Painting by Enoch Wood Perry, 1865

8. . . . and now!

9.  Characterize water quality (WQ) in watershed
 Study temporal and spatial variations in WQ
 Determine effect of NPS pollutants on WQ
 Quantify natural and NPS contributions during base-
flow and storm conditions
 Evaluate potential effects of NPS pollutants on
nearshore biota
 Determine trace element and total suspended
sediment (TSS) loads to coastal ocean
UHM Watershed Study Objectives

10. UHM Quarterly (Q) & Continuous
Monitoring (CM) Stations

11. Tomlinson & Pygmy Flowmeter

12. Quarterly Manual Sampling

14. CM Station WK (upper watershed)

15. CM Station KHS (lower watershed)

16. Extreme Event Monitoring – Storms
The good,
the bad, &
the really UGLY!

19. Turbidity as a Surrogate for
Suspended Sediment

20. DGT Time-Integrating Sampler Study

21. Discrete Sampling Program
 Manual quarterly sampling,
usually base flow (4 years)
 Automated storm sampling
(4 years)
 Streamflow & T, C, pH, DO
& turbidity at 5-minute
intervals (4 years)
 Estuarine grab sampling &
water quality measured in
situ concurrently with DGTs
(7 months)

22. DGT Study Design
 Compare 7 months of
DGT results with stream
data from discrete base-
& storm-flow samples
collected over 4 years
 Compare DGT results
with data from weekly
discrete samples
collected concurrently
with DGT retrievals over 7
months
Estuary
Lower Watershed
Upper Watershed

23. Components of a DGT Sampler
 ABS plastic outer sleeve & piston
 0.45-µm, polysulfone membrane filter
 Polyacrylamide hydrogel (~95% water)
 Layer of Chelex-100® resin in hydrogel

24. DGT Deployment Schemes
Estuary
Streams

25. DGT-Grab Comparison - Stream
Expected DGT concentrations (dissolved → colloids, i.e.,
~0.02 µm) to be lower than discrete samples (0.2-µm filters)

26. DGT vs. 0.2-µm Filter

27. DGT vs. Grab Sample Copper Mystery
(So, what happened here? Manual sampling missed
something? Diel cycle in Cu?)

29. Pacific Islands Ocean Observing System

30. Water Quality Component Locations

31. HiOOS Water Quality Sensors

32. Storm
Effects
(March
2009)

33. AUV – Δ Salinity at 2 - 4 m
Note fresher
water near shore
as a result of the
13MAR09 storm

34. Mar-09 Storm–Long Lasting Effects

35. 11-March-2011 Japan Tsunami
Hawaiian
Islands

36. Japan Tsunami Water Quality Effects

37. Japan Tsunami Water Quality Effects

38. Comparing Storm & Tsunami Turbidity

39. All Clear?

40. All of this within the Waikīkī ahupuaʻa

41. Mahalo! Questions?
Michael Tomlinson
UHM Oceanography, Flagstaff, AZ 86004
928-266-2236, mtomlins@hawaii.edu

42. For attending the 2014 AIPG & AHS National Conference!

43. Continuous Monitoring Challenges
 Many samples, disparate intervals
 Cellular transmission
 Biofouling
 Calibration & biological long-term drift
 Data review and quality control

44. Many Samples, Disparate Intervals
Component
Interval
(min)* №/Yr
NWS Precipitation 15 35,040
USGS Streamflow 15 35,040
NOS Tides 6 87,600
NOS Meteorology (wind, T, P) 6 87,600
HiOOS NS (P, T, S, chl, turb) 4 131,400
HiOOS WQBs (T, S, DO, chl, turb) 20 26,280
HiOOS KNO (waves, currents, scatter, T) 20 26,280
HiOOS AUV (bathy, T, S, chl, scatter, curr) ~0.001 ~57,500/hr
Event Sampling (varies) varies varies
* Statistical analysis may require uniform interval using GRAN, Aquarius®, etc.)

45. Cb = bulk solution concentration
δ = DBL (diffusive boundary layer) thickness
Δg = diffusive gel thickness (ideally ≥10 × δ)
How
the
DGT
Works

46. DGT Assumptions & Requirements
 Diffusive boundary layer thickness δ
(unknown) not significant relative to length
of DGT diffusion path Δg
 Diffusion coefficients of the aquo ions
represent most of the species present
 Biofouling is not interfering with diffusion
process
 Ionic strength >1 mM (~60 µS/cm)
 pH must be >5 and <10

47. Discrete Sample
Processing
Step 1
Filtration
(0.2 µm)
Step 2
Acidification
(quartz distilled
HNO3)
Step 3
FIA (8-HOQ
resin)
Step 4
ICP-MS
analysis

48. DGT ProcessingStep 1 - DGT disassembly
Step 2 - Removal of resin gel
Step 3 - Resin gel leaching (24 hr)
Step 4 -
ICP-MS
analysis
of DGT
leachate

49. Calculating Mean Concentration
where:
Cw = mean metal concentration in water
M = mass diffused into DGT
Δg = diffusive hydrogel thickness +
membrane filter thickness
DT = diffusion coefficient at any temperature
t = deployment (exposure) time
A = area of DGT window

50. WQB Sensor Information
Sensor Res/Prec Accuracy Cost
SBE 16plus CTD $23,000
Temperature 0.0001 °C 0.005 °C
Conductivity 0.00005 S/m 0.0005 S/m
SBE43 & 63 DO – 2%
WET Labs FLNTU
Chlorophyll 0.01 µg/L –
Turbidity 0.01 NTU –
ISUS NO3 Sensor ±0.5 µM ±2 µM or 10% $34,000
STOR-X Telemetry $14,000
C6 Multisensor Platform* various various $17,000
* Equipped with chlorophyll, CDOM, OB/FWA, turbidity, phycoerythrin, &
crude oil sensors; battery pack; and mechanical wiper (wish list).