Messina dissertation defense-4_27_16
- 1. Terrestrial sediment dynamics in a small, tropical, fringing-reef embayment by Alex Messina SDSU/UCSB Joint-Doctoral Program in Geography photo: Messina N Pago Pago Harbor Pacific Ocean South ReefNorth Reef Stream Outlet Faga’alu, American Samoa
- 2. Motivation and Research questions Chapter 1: Where is sediment coming from? and What to do about it? Chapter 2: How does water circulate over the reef? Chapter 3: Where is sediment accumulating on the reef? Sediment accumulation in Faga’alu, Jan 2012 video: Messina Sediment harming coral in Faga’alu 1. Watershed inputs 2. Hydrodynamics 3. Sediment Accumulation RIDGE to REEF
- 3. Chapter 1: Where is sediment coming from? Sediment from Natural Sources and Human Sources Human sources: • Quarry • Storm drains • Roads Natural sediment from forest QuarryRoad runoff Storm drains
- 4. Subwatersheds isolate sediment sources: Natural, quarry, village 2 PT’s (Pressure Transducers) 2 Turbidimeters 1 Autosampler 1 Grad student Sediment yield measured at three locations using: QUARRY 10km
- 5. Measurements: • Water discharge (Q) (L/sec) • Suspended Sediment Concentration (SSC) (mg/L) Depth with pressure transducer (PT) Flow measurements relate depth to water discharge (Q, volume/time) Depth SSYEV = Q x SSC 1. Measure SSC in water samples collected by Autosampler and grab 2. Model SSC from Turbidity data Autosampler Retrieving samples Turbidimeter in stream Grad student
- 6. Measuring sediment and discharge during storms Timelapse videos! Filtering and weighing sediment in laboratory Auto-sampler Measuring Q with flow meter
- 7. Detecting changes in fluvial sediment Q-SSC problematic due to scatter 1. Discharge-Concentration relationship 2. Changes in annual yields 3. Event-wise analysis UPSTREAM DOWNSTREAM CONCENTRATION DISCHARGE (Q) FOREST QUARRY VILLAGE
- 8. Detecting changes in fluvial sediment Sequential downstream sources are confused Q-SSC problematic due to scatter 1. Discharge-Concentration relationship 2. Changes in annual yields 3. Event-wise analysis UPSTREAM DOWNSTREAM CONCENTRATION DISCHARGE (Q) FOREST QUARRY VILLAGE FOREST QUARRY VILLAGE FOREST QUARRY VILLAGE Non-storm Storm
- 9. Continuous Turbidity to… Continuous SSC Q (from depth and rating curve) Integrated over storm to get total SSY = Q x SSC KEY METRIC: Total SSY from storm event KEY METRIC: Total SSY from storm event TimeStorm Start Storm End Storm Event
- 10. SSYEV vs. “Storm Metrics” (precipitation and discharge) How to compare sediment yield from different sources and events? (1)SSYEV(tons/km2) Maximum event discharge (Q) (m3/sec/km2) Example of a “Storm Event” Maximum Event Q Total SSYEV 102 101 100 10-1 10-2 10-3 142 Storm Events measured
- 11. • Compare total and % contributions from sources • KEY METRIC: Disturbance Ratio (DR): DR = SSY / SSYFOREST DR = 1 is no disturbance How to compare sediment yield from different sources and events? (2) SSYEV can be used to make a budget of sources Results from 8 storms Precip SSYEV (tons) mm Upper Lower_Quarry Lower_Village Total Min 12 0.06 0.08 0.3 0.7 Max 86 9.6 8.2 5.3 23.1 Total 299 13.4 16.4 16.0 45.7 % 29 36 35 100 % Area 50 16 34 100 DR 1.0 4.1 1.8 1.7 From 42 storms (UPPER and LOWER only): • Human-disturbed subwatershed contributed ~87% of SSYEV to the Bay • Human-disturbed areas have increased SSY ~3.9x above natural yields to the Bay
- 12. How to compare sediment yield from different sources and events? (2) SSYEV can be used to make a budget of sources Results from 8 storms Precip SSYEV (tons) mm Upper Lower_Quarry Lower_Village Total Min 12 0.06 0.08 0.3 0.7 Max 86 9.6 8.2 5.3 23.1 Total 299 13.4 16.4 16.0 45.7 % 29 36 35 100 % Area 50 16 34 100 DR 1.0 4.1 1.8 1.7 SSY from forested and disturbed areas Upper Lower_Quarry Lower_Village Total Area disturbed (%) 0.4 6.5 11.7 5.2 Forested areas (tons) 13.3 3.7 7.8 25.0 Disturbed areas (tons) 0.1 12.7 8.2 20.7 % from disturbed areas 1 77 51 45 DR for disturbed areas 3 49 8 15 • Quarry makes up small area but high SSYEV • High DR at quarry due to constant disturbance • Compare total and % contributions from sources • KEY METRIC: Disturbance Ratio (DR): DR = SSY / SSYFOREST DR = 1 is no disturbance From 42 storms (UPPER and LOWER only): • Human-disturbed subwatershed contributed ~87% of SSYEV to the Bay • Human-disturbed areas have increased SSY ~3.9x above natural yields to the Bay
- 13. Conclusions from Chapter 1: Where is anthropogenic sediment coming from? Quarry! • Quarry covered ~1% of watershed, but contributed ~36% of SSYEV • Mitigate sediment discharge from quarry Methodological contributions: -Automated storm identification -Quantify change with event-wise SSY -Disturbance Ratio Messina, A., Biggs, T. (2016) “Contributions of human activities to suspended sediment yield during storm events from a small, steep, tropical watershed.” Journal of Hydrology, in press Retention ponds installed Oct 2014
- 14. Chapter Two: How is water circulating over the reef? Water circulation controls sediment dynamics Energetic hydrodynamic forcing compared with other reefs: -Variable winds -Variable waves -> High spatial variability in current velocity and direction How do currents vary spatially over the reef? How do currents vary under calm conditions, high winds, and high waves? WIND/WAVES
- 15. Chapter Two: How is water circulating over the reef? Water circulation controls sediment dynamics Energetic hydrodynamic forcing compared with other reefs: -Variable winds -Variable waves -> High spatial variability in current velocity and direction How do currents vary spatially over the reef? How do currents vary under calm conditions, high winds, and high waves? WIND/WAVES Exposed to big waves!
- 16. Wave height recorder Building drifters 3 acoustic current profilers 5 GPS-recording drifters Deployed via paddleboard EULERIANLAGRANGIAN METHODS Two ways to observe flow: • Eulerian: flow past fixed point • Lagrangian: follow water parcel
- 17. Chapter Two: How is water circulating over the reef? • Lagrangian = spatial coverage Lagrangian drifters GPS-tracked drifters, to determine spatial patterns related to wind and wave forcing
- 18. Chapter Two: How is water circulating over the reef? • Eulerian = temporal coverage Eulerian current meters Current meters at fixed points to determine temporal patterns related to wind and wave forcing
- 19. Unprecedented spatial coverage: 30 deployments of 5 drifters Wide range of forcing conditions -> “end members” Gridded drifter observations: 100m x 100m Divided into three periods, isolating forcing conditions: -Tide (Calm) -Strong onshore winds -Large waves 100 m 100m
- 20. TIDES (CALM) STRONG WINDS LARGE WAVES Spatial patterns: 1. Faster speeds, consistent directions over southern reef (crest) 2. Slower flow, variable direction over northern reef and channel Forcing patterns: 1. Tides (calm): Slow speeds, variable directions 2. Strong Winds: Slow speeds, toward stream outlet 3. Large Waves: Fastest speeds, most uniform directions; clockwise flushing pattern DRIFTERS: Mean flow speed and direction Slow, variable direction Slow, onshore direction Fast, clockwise circulation
- 21. TIDES (CALM) STRONG WINDS LARGE WAVES Spatial/Forcing patterns: •Similar to Drifters, but no spatial variation over the reef, clockwise pattern •Contextualize drifter measurements, and show flow decreases with tide stage Comparing Eulerian/Lagrangian: 1. Speeds faster for drifters (50-650%): • Point – Area • Surface – Water column • Stokes’ drift • Sampling/Analytical error 2. Implications ADCPs: Mean flow speed and direction Fastest, esp. on southern reefSlow, less variable directionsSlowest, most variable directions
- 22. Water residence time Spatial patterns • Lowest over southern reef (crest) • Highest over northern reef and near stream outlet Forcing patterns • Lowest during large waves • Highest during calm and strong onshore winds Implications: • Stream discharge deflected over northern reef • Potential for sediment impacts highest over northern reef, under calm or onshore wind
- 23. Conclusions from Chapter 2: How is water circulating over the reef? • Wave-breaking on southern reef crest strong control on circulation • Highly heterogeneous currents over short spatial scales • Stream discharge likely deflected over northern reef and channel • Lagrangian velocities were faster than Eulerian; can overestimate flow Methodological contributions: -Combined Lagrangian/Eulerian approach -Spatial coverage of drifters over reef flat -Spatially distributed residence time -End member forcing Messina, A., Storlazzi, C., Cheriton, O., Biggs, T. (in review) “Eulerian and Lagrangian measurements of water flow and residence time in a fringing reef flat-lined embayment: Faga’alu Bay, American Samoa.” Future work: real-time tracking
- 24. Chapter 3: WHERE is sediment accumulating? and WHEN? What processes control sediment accumulation, in space and time? gross and net? How sediment input and hydrodynamics interact? Monthly? Seasonal? Are accumulation rates above harmful levels? High waves > Low water residence time > prevent deposition & remove deposited sediment High SSY from watershed and/or Low wave-driven circulation Hypotheses High sediment accumulation when:
- 25. Sampling Gross and Net accumulation -10 quasi-monthly, for 1 year - gross -> in TRAPS - net -> on PODS “sediment trap” “sediment pod” Methods: Sediment Collection & Analysis Analysis Grain size and Composition -fine/coarse fractions separated -rinsed of salts -analyzed for composition: 0rganic, Carbonate, Terrigenous Sieving/Filtering apparatus Organisms/Gravel removedSediment collection on SCUBA Rinse and Oven-dry
- 26. Interaction of Waves and SSY High SSY from watershed and/or Low wave-driven circulation Hypotheses Increased sediment accumulation from: Hypothetical phasing of Waves and SSY Removal Deposition
- 27. SSY (tons): Measured/Modelled-Qmax model (Ch1) Waves (mean height, m): Model-NOAA WaveWatch3 Daily mean wave height, and total SSY over deployment period (dashed lines) Interaction of Waves and SSY High SSY from watershed and/or Low wave-driven circulation Hypotheses Increased sediment accumulation from: SSYEV(tons/km2) Maximum event discharge (Q) (m3/sec/km2) Qmax – SSYEV model (Ch 1) **Sediment mitigation decreased SSY, so two models calibrated
- 28. Time-Lapse photography Moultrie GameSpy I-35 (8MP, 15 min interval) Sediment plume following large rain 2/21/14 – Calm conditions 15:45 North Reef South ReefStream 16:15 Sediment plume deflected over North reef and Channel 17:00
- 29. Spatial patterns of sediment accumulationBENTHICSEDIMENT
- 30. TRAPS(GROSS) • Higher accumulation on north reef and near channel • Composition reflected surrounding benthic sediment Spatial patterns of sediment accumulationBENTHICSEDIMENT *Note: different chart scales
- 31. TRAPS(GROSS)PODS(NET) • Higher accumulation on north reef and near channel • Composition reflected surrounding benthic sediment • Higher accumulation in traps vs. on pods Spatial patterns of sediment accumulationBENTHICSEDIMENT *Note: different chart scales
- 32. TRAPS(GROSS)PODS(NET) • Higher accumulation on north reef and near channel • Composition reflected surrounding benthic sediment • Higher accumulation in traps vs. on pods Spatial patterns of sediment accumulationBENTHICSEDIMENT *Note: different chart scales
- 33. PODS TRAPS Seasonal SSY and Wave patterns: • Highest SSY in July (dry season) due to one large storm • Waves were larger in dry season (May- Oct), smaller in wet season (Nov-Mar)NORTHERNSOUTHERN PODS MEANACCUMULATION
- 34. PODS TRAPS Temporal patterns – PODS: • Accumulation on Pods did not correlate with SSY or Waves • Much higher accumulation (esp. terrig) on northern reef • Higher terrigenous accumulation after large SSY event Seasonal SSY and Wave patterns: • Highest SSY in July (dry season) due to one large storm • Waves were larger in dry season (May- Oct), smaller in wet season (Nov-Mar)NORTHERNSOUTHERN PODS MEANACCUMULATION
- 35. Temporal patterns – TRAPS: • Carbonate accumulation in Traps correlated with Waves • Similar to Pods, much higher on northern reef • Similar composition as on Pods • Highest accumulation due to large wave events, esp. southern reef TRAPS Seasonal SSY and Wave patterns: • Highest SSY in July (dry season) due to one large storm • Waves were larger in dry season (May- Oct), smaller in wet season (Nov-Mar) MEANACCUMULATION NORTHERNSOUTHERN Large Waves
- 36. Temporal patterns at sites: TRAPS Exceeded coral health thresholds in some cases, mostly on northern reef Carbonate accumulation correlated with Waves on reef crest (1C, 2C, 3C) and reef crest (1B, 3B) Accumulation low where surrounding availability is low (2B) Terrigenous accumulation correlated with SSY only near stream (2A)
- 37. Controls on sediment accumulation NORTHERNSOUTHERNCENTRAL Accumulation in TRAPS vs. SSY, Waves Sediment accumulation correlated with Waves Suggests waves resuspend and transport carbonate sediment over the reef SSY only near stream (2A) SEDIMENTACCUMULATION
- 38. Conclusions from Chapter 3: WHERE is sediment accumulating? • Northern reef and near Channel • Due to circulation patterns and SSY from stream WHEN is sediment accumulating? •High waves transport benthic sediment •SSY is important, but complex and short time scale Methodological contributions: -Combined sediment traps and pods: gross and net -Related accumulation to measured SSY -Sampled across gradients in distance from stream outlet and hydrodynamic energy Messina, A., Storlazzi, C., Biggs, T. (2016) “Watershed and oceanic controls on spatial and temporal patterns of sediment accumulation in a fringing reef flat embayment: Faga’alu, American Samoa.” in preparation
- 39. Conclusions photo: Messina For Faga’alu • SSY significantly increased by human disturbance (mostly the quarry) But now it’s fixed! • Waves cause heterogeneous currents, protect southern reef but stress northern reef • Sediment accumulation strongly influenced by surrounding benthic sediment, moved by wave-driven flow • Terrigenous accumulation correlated with SSY only near stream, impacted northern reef over longer timescales • Reef recovery is anticipated but uncertain timescale and flushing of deposited sediment • Daily sediment accumulation patterns and impacts on coral health are still unknown
- 40. Conclusions photo: Messina For Fringing Reefs • Rare to have all three R2R components, baseline for management assessment • This study provides example of relatively simple Ridge-to-Reef study to inform coral management • SSY in steep, tropical islands is sensitive to human disturbance • Waves and currents can significantly alter LBSP impacts • Time scales of sediment transport, deposition, and reworking are uncertain, so watershed restoration may take a long time to observe • Coral is under threat from global stressors, but we can save coral from terrigenous sediment stress!
- 41. Meagan CurtisJameson Newtson Trent Biggs Curt StorlazziDr. Mike Favazza QUESTIONS? Fa’afetai tele lava (big thanks) to all who helped in the field! Thanks to Mayor Uso and Faga’alu Village Rocco Tinitali Mr. Jeffrey Roger “Young” Greg McCormick
- 43. Vegetation overgrowth 2012-2014 Surfaces covered in gravel 2013 Groundwater diversion So WHAT to do about it?
- 46. SSYEV(tons/km2) Maximum event discharge (Q) (m3/sec/km2) Pre-mitigation: 2012-2014 Post-mitigation: 2014- 2016 (ongoing) Future research: Did sediment mitigation work? Maximum event discharge (Q) (m3/sec/km2) SSYEV(tons/km2) 102 101 100 10-1 10-2 10-3
- 47. Pre-mitigation: 2012-2014 Post-mitigation: 2014- 2016 (ongoing) Did sediment mitigation work?
- 48. Turbidity-SSC Rating Curves: A unique regression for each location/instrument Equation to convert continuously recorded Turbidity (NTU) to SSC (mg/L)
- 49. SSY = Q x SSC PT pressure – Barometric Pressure = Water pressure Water pressure x Density of water = Stream stage Q = a x Stream Stage + b SSC = a x Turbidity + b
- 50. Rain Gauge – Tipping Bucket Rain falls into the ‘buckets’ and tips when full, filling the other side. The logger counts the tips
- 51. Turbidimeter Installations Always in the water Free of debris Not get buried Easy to access to maintain
- 52. PT in metal stilling well OBS in plastic housing Solar panel Datalogger/battery Barologger Fale (fall-ay) at FOREST location
- 53. Barologger (top) PT (in water) in PVC stilling well Autosampler on upstream side of bridge Solar panel Datalogger/battery OBS500 Tubing and water level sensor Autosampler in box
- 54. Field equipment can malfunction due to all sorts of issues! Storm damage! Turbidimeter destroyed by large storm Autosampler completely disappeared! Ants colonized this rain gauge! Rain gauge clogged with debris!
- 55. Vandalism
- 56. Pressure Transducer • HOBO and Solinst Level-logger • They do the same thing • Measure pressure from air and water Turbidimeter Water Level Air Pressure Water Pressure • Measure how cloudy the water is • Shine light into water, measure reflection
- 58. Comparing Eulerian/Lagrangian • ADCPs show flow into bay only, drifters show change in flow trajectories • Drifters show faster flows • Surface vs depth-integrated • Spatial variation in grid cell • Stokes’ drift Spatial patterns: 1. Faster flow, unidirectional over southern reef (crest) 2. Slower flow, variable direction over northern reef, and near stream outlet Forcing patterns: 1. Tides (calm): Slow flows, variable directions 2. Strong Winds: Slow flows, toward stream outlet 3. Large Waves: Fastest flows, most uniform directions; clockwise flushing pattern