Hydrodynamic and Sediment Fluxes through the Ria Formosa Inlets André Miguel Duarte Pacheco

Chapter I. Introduction Chapter II. Assessment of Tidal Inlet Evolution and Stability using Sediment Budget Computations and Hydraulic Parameter Analysis Pacheco, A., Vila-Concejo, A., Ferreira, Ó., Dias, J.A., Marine Geology 247, Chapter III. A Statistical Evaluation of Models for Extrapolating Current Velocities from Boat-mounted ADCP Profiles Pacheco, A., Faísca, L., Almeida, L., Ferreira, Ó., Williams, J.J., Dias, A. (under review). Continental Shelf Research Chapter IV. Hydrodynamics and Equilibrium of a Multiple Inlet System Pacheco, A., Ferreira, Ó., Williams, J.J., Garel, E., Dias, J.A., Marine Geology 274, Chapter V. Application of Sediment Transport Models to a Multiple Inlet System Pacheco, A., Williams, J.J., Ferreira, Ó., Garel, E., Reynolds, S., (under review). Estuarine and Coastal Shelf Science Chapter VI. Long-term Morphological Impacts of the Opening of a New Inlet in a Multiple Inlet System Pacheco, A., Ferreira, Ó., Williams, J.J. (under review). Earth Surface Processes and Landforms Chapter VII. General Conclusions Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 2 THESIS OUTLINE

Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 3 CHAPTER I. Introduction Scientific interest in the Ria Formosa The Ria Formosa in the context of International Research on Tidal Inlets The system has historically responded to disturbances with significant changes in its morphology, always maintaining multiple inlets open; Several interventions occurred recently (human occupation on the barriers, channel dredging for navigation proposes, inlet openings and stabilisation), which makes Ria Formosa an important test case to evaluate the impacts of human interventions; On the contrary of single inlet systems, multiple inlets are hydrodynamically connected, where nonlinear processes an important role on their stability, which needs to be better understood; Data on hydrodynamic variables, tidal prism and sediment fluxes, collected simultaneously on multiple inlet systems is scarce; scientific knowledge of these processes is considered to be the key to understand inlet interactions; This thesis presents an in-depth study of Ria Formosa Inlets; Results are expected to show how dynamic equilibrium can be achieved, with potential applications for the formulation of ICM’s policies of these systems

CHAPTER II. Assessment of Tidal Inlet Evolution and Stability using Sediment Budget Computations and Hydraulic Parameter Analysis APPROACH Couple of sediment budget (best estimates and uncertainty) with inlet hydraulic parameters Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 4 The approach developed is applied to Faro-Olhão Inlet AB

CHAPTER II. Assessment of Tidal Inlet Evolution and Stability using Sediment Budget Computations and Hydraulic Parameter Analysis A series of digital maps was produced based on multi-year data acquired from charts, surveys and aerial photos; Sediment volumes and fluxes were determined for 6 coastal cells delineated on the basis of the morphological features of the inlet; Cell volumes, fluxes and uncertainties were calculated for three periods ( ; ; ); Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 5 Inlet Stability Parameters Configuration of five cross-section profiles along the inlet channel (P1-5) and 10 equi-spaced cross-section profiles normal to the ebb and flood flow lines (A cs 1 – Ac s 10) Average cross-sectional area (A avg ) of inlet channel area (P1- P5) and minimum cross-sectional area (A c ) of Faro-Olhão inlet measured using the CEM method, Inlet hydraulic parameters (Ac RH Mdepth) for 13 surveys between

MAIN CONCLUSIONS from the sediment budget computations : capture the LST in order to build both deltas; channel scouring began; : both deltas accumulate sediments at the same rate; major changes occurred on the barriers; scouring activity intensifies at inlet channel; : stabilisation of both coastlines; drastic intensification of scouring at the inlet gorge from inlet parameters, generally used to infer equilibrium The inlet reached equilibrium ( ); SB revealed that only now the adjacent coast is adjusting to the inlet presence; Inlet dynamics can be inferred by analysing one budget cell (SAND SHARING SYSTEM) (Dean, 1988; Elias and van der Speck, 2008) Evaluate inlet parameters helps; but to infer inlet equilibrium, SB must be considered Other external controls must be included on Ac/P formulations (e.g. stratigraphy) Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 6 CHAPTER II. Assessment of Tidal Inlet Evolution and Stability using Sediment Budget Computations and Hydraulic Parameter Analysis

Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 7 CHAPTER III. A statistical evaluation of models for extrapolating current velocities from boat-mounted ADCP profiles SUBJECT Presents a statistical evaluation of two theoretical fitting methods – logarithmic and power law – commonly used to extrapolate velocities to unmeasured areas of a vertical profile, such as near-bottom velocities MOTIVATION Increase use of boat-mounted ADCPs and the inexistence of any study on evaluating model performance when using this equipment To compute cross-sectional discharge, mean velocities and bed-friction velocities, custom ADCP software makes use of theoretical models to reconstruct the velocity profile in the unmeasured areas, based on models calibrated and tested for fixed current meters (i.e. using time-average profiles) PROBLEM OF USING BOAT-MOUNTED ADCPs Instantaneous velocity profiles are incapable of smoothing out fluctuations induced by micro- and macro-turbulence and the potential effects of channel irregularities; HOWEVER Boat-mounted ADCPs can successfully estimate discharges WHY? Spatial averaging thought the transect; Measure depth average velocity and cross-section velocities using raw and each model

Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 8 CHAPTER III. A statistical evaluation of models for extrapolating current velocities from boat-mounted ADCP profiles Statistical analysis R 2 between raw U cs and predictions by the two theoretical models Two bifactorial ANCOVA tests: 1)Model performance was assessed by the correlation (i.e. outcome variable) between real and model U cs ; tide (2 groups) and inlet (6 groups) as independent variables; 2) Again tide and inlet as independent variables, but now the difference in R 2 between PL and LL as outcome variable. ANCOVA determines differences between groups while controlling the effect of one or more continuous predictor variables (COVARIATES). Covariate is controlled allowing the effect of each independent variables has on the outcome to be analysed. TWO COVARIATES used separately: (i) U cs raw data (expressing the velocity and direction of the water flow); (ii) U cs raw modulus (expressing the velocity of water flow alone); Assumption of homogeneity of slopes was checked.

Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 9 CHAPTER III. A statistical evaluation of models for extrapolating current velocities from boat-mounted ADCP profiles FIRST TEST R 2 values are not affected by raw U cs but by its modulus; Tidal signal (flood/ebb) is negligible; >> U cs decrease the performance of both models; when this effect is excluded, differences between inlets are highlighted – poorer fit of both models in smaller A c SECOND TEST Better overall performance of LL method; particular at flood tide; Significance only found for U cs raw, indicating that both velocity and direction of the flow contribute to the LL model’s superiority Fit of both models is negatively correlated with velocity Better performance of both models at inlets with larger A c Both models' performance decrease during ebb Not particular related to velocity but to channel depth LL model fits the raw U cs with the greatest statistical significance

MOTIVATION Multiple inlet systems: existence of residual circulation – direct control on the net transport of material Understand the interactions between the inlets How those interactions can contribute to the persistence of multiple inlet systems CHAPTER IV. Hydrodynamics and equilibrium of multiple-inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 10 SUBJECT Detailed measurements of hydrodynamic variables over complete spring and neap tidal cycles in a multiple inlet system (Ria Formosa, Portugal) BACKGROUND Multiple inlet systems are unstable (van de Kreeke, 1990) Evidence show that some multiple inlet systems can be stable Degree of equilibrium depends on the degree of connectivity between basins van de Kreeke et al. (2008) Importance of the connectivity between the inlets on the equilibrium of multiple inlet systems needs to be better understood

CHAPTER IV. Hydrodynamics and equilibrium of multiple-inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 11 Since the 14th Century to present System maintained 4 to 7 inlets Opening of Faro-Olhão Inlet Capture of large P from Armona Inlet FO reaching equilibrium Armona narrowing Flood/ebb durations - ST and NT Non-linear tidal distortion varies considerably Salles (2001) and Salles et al. (2005) What are the present circulation patterns? Can they help to understand the medium to long-term stability of the system?

(A) tidal prism; (B) mean cross-section velocities; (C) residual discharge; and (D) mean residual velocity 90% of P ANC 6% FO 62% ARM 28% CHAPTER IV. Hydrodynamics and equilibrium of multiple-inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 12

Faro-Olhão and Armona inlets are always interconnected ST - Faro-Olhão Inlet – although mean ebb velocities are higher and ebb duration shorter, maximum velocities are landward oriented (sediment import into the system) NT – inlets act more independently; residual circulation is lower; an inner circulation operates between the two main channels As suggested by van de Kreeke, the morphology of the inner channels connecting the inlets exert an important role by controlling the interactions between inlets There seems to be a contradiction of the bay to fill or self-maintain through flushing Strong residual flow and change of Prism Together with the magnitude and direction of maximum velocity factors determining the flow and transport dominance Seems to be the key factor contributing to the short to medium-term stability (Van de Kreeke and Robaczewska, 1993; Salles et al., 2005) CHAPTER IV. Hydrodynamics and equilibrium of multiple-inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 13

SUBJECT Quantification of ST in tidal inlets MOTIVATION Fundamental requirement to support both conceptual and numerical modelling of tidal inlet function and evolution High frequency flow measurements of turbulence Samples of surficial sediments Water levels Vertical current velocity profiles Time-average shear stress (  ) Drag Coefficient (Cd) Apparent bed roughness (za) Empirical relation between Ks and Ka CHAPTER V. Application of sediment transport models on a multiple inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 14 Estimation of critical flow conditions (  cr ) BED LOAD TRANSPORT (Yalin, 1964; van Rijn 1984a; Madsen 1991; Nielsen, 1992) SUSPENDED TRANSPORT U DA values from ADCP and sediment properties C(z) was determine by 3 methods (power lar, rouse profile, C-W method) Ca (Smith and McLean, 1977; van Rijn 1984b; Zysernan and Fredsoe, 1994) Use of ADV data

1.21E Standard deviation -1.44E Mean -1.54E C-W (Soulsby, 1997) -1.33E C-Power (Soulsby, 1997) -1.33E C-Rouse (Soulsby, 1997) m 3 year -1 *Net (kgm -1 s - 1 ) Ebb (kgm -1 s - 1 ) Flood (kgm -1 s - 1 ) SUSPENDED TRANSPORT 6.39E Standard deviation -1.94E Mean -1.81E Nielsen (1992) -1.51E Madsen (1991) -2.88E Van Rijn (1984) -1.57E Yalin (1964) m 3 year -1 *Net (kgm -1 s - 1 ) Ebb (kgm -1 s - 1 ) Flood (kgm -1 s - 1 ) BED LOAD TRANSPORT Estimates of mass (kgm -1 s -1 ) and volumetric (m 3 year -1 ) transport rates using different models for bedload and suspended transport for a ADCP tidal cycle performed at Ancão tidal inlet For bedload, porosity (ε =0.4) is considered, i.e., volume of settled-bed material The cross-sectional width is w=110 m, integrated transect value derived from the ADCP measurements CHAPTER V. Application of sediment transport models on a multiple inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 14 Bedload transport methods derive by a maximum factor of 2 van Rijn method produces the higher estimates Mean value of bedload is 3 times higher at ebb than at flood The suspended sediment estimates are all very similar Sediment transport mainly occurs as bedload Despite the presence of waves, the bulk of suspended sediments is accomplish by turbulent diffusion processes related to the strong tidal currents Changes on hydraulic bed roughness between flood and ebb tides due to the increase of current speed, enhanced sediment transport offshore

CHAPTER V. Application of sediment transport models on a multiple inlet systems Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 14 ESTABLISHMENT OF A USEFUL CONCEPTUAL ST MODEL Approach accurately reproduce the magnitude and direction of ST Ancão Inlet net annual seaward sediment flux is estimated to be 100,000m 3 /year ~ to the annual LST rate measured at the updrift margin during fair-weather conditions Faro-Olhão Inlet is a net importer of sediment to the system (1/3 of the annual sediment entry in the system by comparing pre and pos dredging surveys) The other inlets seem capable of flushing sediments during fair weather conditions (EVIDENCE OF MULTIPLE INLET PERSISTENCY) Armona ST balance  0 Armona exports sand during ST and imports during NT IMPORTANCE ON PROPERLY QUANTIFY EBB TIDAL VOLUMES FROM PREVIOUS HYDRAULIC CONFIGURATIONS METHODOLOGY PROVED TO BE AN USEFUL TOOL TO COMPUTE MEDIUM TO LONG-TERM SB’s IF coupled with long term quantification of ebb/flood deltas volumes

SUBJECT Evaluate the morphological response of the coastline to the opening and stabilisation of an inlet on a multiple inlet system MOTIVATION Relate those changes to the inlet A c and P redistributions until a new equilibrium is reached Understand the importance of the ebb tidal deltas on SB’s as “valves” for the coastal sedimentary supply FOCUS FO inlet became the main inlet of the system, ARM lost efficiency EVALUATE SYSTEM EQUILIBRIUM AND PREDICT IT FUTURE EVOLUTION METHODS Use of a range of morphological and hydrodynamic indicators (evolution of A c and P, length and area of barrier islands) CHAPTER VI. Long-term morphological impacts of the opening of a new inlet on a multiple inlet system Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 16

1) FO A c versus ARM W R 2 =0.76 2) CUL A versus ARM WR 2 =0.91 Exponential behaviour from mid-1980’s to present Coincident with increase scouring in the submarine gorge of FO ARM lost flow dominance to FO Between 1970’s-1990’s Ratio between Faro-Olhão inlet A c / Armona Inlet W and between Culatra island A / Armona Inlet W YearFOARM x10 6 m 3 46x10 6 m x10 6 m 3 58x10 6 m x10 6 m 3 49x10 6 m x10 6 m 3 49x10 6 m x10 6 m 3 46x10 6 m x10 6 m 3 24x10 6 m 3 ARM P ~ c.25% of original volume A c FO on 1948 ~ c.25% of A c in 2006 CHAPTER VI. Long-term morphological impacts of the opening of a new inlet on a multiple inlet system Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 17

Opening of FO greatly reduced the flow through ARM (~ 75%) Ebb-tidal current loss over the ebb shoal Pushed the shoal landward (EBB DELTA COLAPSE) SUPPLY of SEDIMENT TO THE GROWTH OF CUL & ARM NARROWING Process peaked up between 1980s and 2000 Coincident with the intensify scouring at FO FO AND ARM INLET EVOLUTION SHOW A CLEAR MORPHOLOGICAL RELATION P EXCHANGE AND DOMINANCE FO A c stabilised; ARM W is constant; CUL dimensions stabilised FO retains LST; CUL processes are cross-shore (net change 0) NO CONTRIBUTION TO ARM EBB TIDAL DELTA ARM BYPASSES SAND DOWNDRIFT AND IS EBB DOMINATED FURTHER RESEARCH IS NEEDED TO QUANTIFY EBB DELTAS STORAGE AND HOW THAT SEDIMENT IS MOBILISED DURING INCREASE WAVE ACTIVITY Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 18 CHAPTER VI. Long-term morphological impacts of the opening of a new inlet on a multiple inlet system

Develop inlet history based on SBs and link it with inlet parameters (1) interpret sediment pathways; (2) revealed accretion/erosion tendencies of different coastal cells; (3) uncertainty computations avoided over-interpretation and highlights how the SBs can be better constrained in the future Evaluating the impacts of opening of tidal inlets on multiple inlet systems (1) understand P readjustments and it importance on shaping the adjacent coastlines; (2) understand the role that ebb-tidal delta have as “valves” of sediment supply; Methods to determine P and ST were presented (1) infer the stability of the system; (2) understand the circulation patterns and evaluate their influence on ST patterns; (3) highlight inlet interconnections and hypothesise it role on multiple inlet stability; (4) both conceptual and numerical modelling of tidal inlet function and evolution; (5) planning of coastal actions; Evaluate the performance of theoretical methods on U extrapolations (1) define the best method to estimate bed-friction velocities used on ST formulae; (2) improve the use of boat-mounted ADCPs on P and ST estimates; CHAPTER VII. General conclusions Hydrodynamics and Sediment Fluxes through Ria Formosa Inlets 19 FARO-OLHÃO OPENING Provides an important informative case on multiple inlet dynamics Coast is reaching dynamic equilibrium; FO and ARM are always interconnected, an interconnection that is variable through the lunar cycle and seems to contribute to inlet stability; Brouwer, 2006; Brouwer et al., 2008; van de Kreeke et al., 2008 RECCOMENDATIONS FOR FURTHER WORK Formulae used to evaluate inlet equilibrium must be reviewed Annual variations of LST should be enclosed on various models of coastal morphology The role of residual circulation in enhancing stability must be better understood Evaluate the importance of the long-term equilibrium of sediment storage in the ebb tidal delta have on the equilibrium of multiple inlet systems Causes and conditions that determine equilibrium (van de Kreeke et al., 2008), mainly length and friction of inlet channels, are not clear and must be better studied

Supervised by Óscar Ferreira (Universidade do Algarve, Portugal) Jon Williams (University of Plymouth, UK) Financial support for the fieldwork given by the project IDEM-Inlet Dynamics Evolution and Management at the Ria Formosa (POCI/MAR/56533/2004) Historical data provided by Parque Natural da Ria Formosa and Instituto Portuário dos Transportes Marítimos THANK YOU