OVERTOPPING FLOW PROPERTIES CHARACTERIZATION IN LABORATORY AND PROTOTYPE THROUGH THE COMBINATION OF NON INTRUSIVE INSTRUMENTAL TECHNIQUES

Overtopping events may cause different failure modes depending on overtopped flow characteristics. Most of the studies about overtopping hazard analysis link the damages caused by the overtopping event to its mean overtopping discharge (q), which provides no information about overtopped flow characteristics or its spatial distribution. In this paper it is presented a non intrusive measurement system based on video imagery techniques and optical level sensors (OLS), which aim is to obtain overtopped highly aerated flows´ principal characteristics: velocity, volume, and density, in order to deep in the knowledge of this phenomenon, and minimizing the damages that it may cause to port´s infrastructures and exploitation


INTRODUCTION AND MOTIVATION
In the last 10 years, port´s infrastructures have shown a clear trend to grow seawards, looking for greater drafts at deeper waters with the aim of giving service to a fleet that continuously increases its dimensions.Within this context, defense structures have a remarkable role due the economic impact that they have in the total cost of the projects, and the technical challenge that they represent for its construction.
The high cost of a meter length of sea defense makes necessary to exploit the calm waters placed landwards in order to profit the inversion of its construction.As a consequence of it, some port´s activities placed adjacent to seawalls are exposed to severe maritime climate action, and therefore overtopping events are more likely to occur affecting vulnerable areas where important economic activities take place.
Overtopping may cause different failure modes depending on flow characteristics; it is known that green water is associated to higher damages than white water or spray (Eurotop 2007).In order to cope with overtopping hazards next three actions should be always to be taken into account (Allsop 2005):  Move human activities. Accept occasional hazards. Increase defense standards.
Traditionally coastal engineering has focused on increasing the protection standards.These acts derive on important material costs which in some cases can make the project economically unfeasible.
Taking the previous paragraphs into account, the authors of this paper would like to highlight the importance of the first two responses to overtopping hazards on port´s exploitation and design.The first one, "move human activities", could be expressed as changing the activity to a less vulnerable one against overtopping.This is to say, port´s activity management.In the case of the second one, "accept occasional hazards", it is necessary to quantify the consequences of overtopping hazard in order to decide whether are assumable or not, that is to say quantify the risk and determine if is there a need to take preventive or corrective actions.(Figure 1).
The combination of these two first logical responses is risk management.This concept aims to help into the making decision process bounding the risk assumed and optimizing the exploitation of port´s activities and structural design.A first approach to risk management is PORMAT, a general methodology to introduce overtopping risk management in port´s exploitation and design (Alises A. et al 2012) which is applied with a software in matlab® to facilitate its application by final users.NIOVE PROMES aims to be suitable for both lab tests and prototype overtopping measurement.At this stage of the project, works are focused on the development of the methodology and the design of the instrumental setting.Therefore different physical models have been designed in order validate the proposed methodology against different types of overtopping events.
Main objectives of this paper are summarize as follows:  Develop of the methodology which should be applied in order to obtain overtopping flow properties characteristics (velocity, density and volume). Design the instrumental setting. Synchronize the instrumental system. Design a 2-D physical model.

OVERTOPPING MEASUREMENT TECHNIQUES
Overtopping has been widely studied in the last years.Extensive research about overtopping has been performed both in small and large scale lab tests such as VOWS, OPTICREST.CLASH project is probably the main reference on overtopping study, within this project, important advances have been done following different lines: obtaining new methods for mean overtopping discharge assessment (Verhaeghe H 2004); including new model and prototype´s experimental data on available formulae in order to make it more accurate; evaluating overtopping hazards (Bouma et al. 2004); and Organizing all the information from experimental test realized both in CLASH PROJECT and in previous research works in a single data base (Verhaeghe H et al 2003).
Traditionally overtopping has been studied using the mean overtopping discharge as the principal key response.This variable gives aggregated information of overtopping´s frequency and volume during a sea state.Therefore, given a mean overtopping discharge value, is not possible to determine whether overtopping events are small and frequent, or rare but with a high volume associated instead.As a consequence mean overtopping discharge may be valid to analyze overtopping events from a qualitative point of view, but is not sufficient by itself to characterize it.
With the aim acquiring deeper knowledge of overtopping events, several instrumental techniques have been developed and applied in the last decade to overtopping analysis, both in lab and prototype tests (Schüttrumpf H. et al 2002;Pullen T. et al 2003;Troch P. et al 2004;De Rouck J. et al 2004;Briganti R. et al 2005;Pullen T. et al 2009).The use of these techniques has allowed both obtaining additional overtopping key responses, and modifying previous existing ones in order to obtain a better characterization of overtopping events (see "modified mean overtopping discharge (q new )" Campos, A et al 2012).
Video imagery techniques have been already successfully applied for studying flow structure interaction (Molina, R et al. 2008).The development of a new non linear approach to solve the geometrical problem (Molina, R 2006) has allowed to obtain direct measurements of processes with a high spatial distribution.In addition, statistical image analysis and other video imagery techniques such as time-stack, allow to obtain statistical information that cannot see the naked eye.All this techniques have been integrated within ZEUS project which is being developed by UGR-HRL since 2005.
Despite the potential that video imagery techniques have shown when dealing with high spatial and temporal distributed phenomena such as overtopping events, it is not capable by itself, to obtain information about the intrinsic characteristics of the overtopped flow.Therefore this technique needs to be combined with an Optical Level Sensor (OLS) system which can be placed inside the mass of airwater mixture, in order to achieve information from the inside of the overtopped volume and measure its intrinsic properties.
OLS (Figure 2) is a digital output instrument which indicates whether it is humid (1), or dry (0).Its operation is based on the total internal reflection principle.This type of sensor is being applied in different branches of the engineering such as automotive or medical industries.The typical technical specifications of OLS are following: The overtopping tests performed at the HRL´s wave flume have been over instrumented with the aim to compare and correlate the results obtained by the proposed combination (OLS and video imagery techniques), against conventional overtopping measurement systems.In order to do so the six different measurement systems employed and the wave paddle movements have been recorded synchronously.
The sample frequency of all instruments has been 1000 Hz except high speed cameras which have recorded at 100 fps due to computer´s capacity.In order to ensure the synchronization between high speed video cameras and the rest of the measurement systems, these first ones have worked on external trigger mode and trigger signal has been also recorded at 1000Hz sample rate.

Conventional instrumental setting
Overtopping events have being measured by the overtopping reception tank.Overtopped water has being measured both in volume, by a pressure transducer placed at the bottom of the tank, and in weight by the usage of a load cell.In order to obtain best quality records, 0.05m thick of open cell polymeric porous media (Cabrerizo 2010) was placed at the bottom of each of the nine overtopping reception tank´s cell, which acted as a damper minimizing the oscillations of the free surface within the reception tank, and therefore minimizing the noise at the pressure transducer´s signal due to the water column oscillation.
Wave impacts have been measured with 9 pressure transducers placed on the front wall of the physical model.The records obtained will provide information about wave breaking conditions and will allow to correlate them with the type of overtopping event.
A total of 8 resistive gauges have been installed along the wave flume (see Figure 3) in order to record the free surface during the tests.The distance between the 4 resistive gauges of each group, a first group placed next to the model and the other next to the paddle, has being determined following Mansard´s criteria to separate incident and reflected waves (Mansard 1981).
Finally, all tests realized have been recorded with an IP camera with a general view of the model to control the experiments performed and recreate them if needed.

NIOVE PROMES instrumental setting
The proposed instrumental system consists of a total of 48 OLS and 2 high speed video cameras.OLS have been placed over the crest of the physical model in 4 groups of 12 OLS, which are housed within hydrodynamic shaped beams in order to affect as less as possible the overtopped flow.These groups have been installed in a staggered form, two of them next to the front wall of the model, and the other two 0.15 m behind.Attending to the vertical distribution, the instrumental density is not constant along the beams in which are placed the OLS; the first 0.15 m of each beam is instrumented with 6 OLS separated each 0.025 m, and the other six dist 0.05 m from each other.The aim of this vertical distribution is focusing the measurements on those areas in which the boundary between water and air-water mixture is expected.
Two high speed video cameras have been placed in a zenithal and lateral view.These cameras have been controlled using ZEUS, an open source program that integrates camera management, acquisition tasks and post-processing techniques, which is being developed by the HRL.
Finally in order to obtain best quality video post-processing results, light conditions have been controlled, and model´s crest and flume´s wall painted in mild gray to make the background´s color neither help nor harm any of the three RGB channels of video imagery acquisition.In the introduction and motivation of the present paper, it has been pointed out the necessity of including risk management in port´s exploitation and design.It is known that overtopping consequences are not strictly associated either to mean flow, nor to the accumulated overtopped volume.
Last step of the presented methodology is therefore to obtain overtopping kinetic and potential energy, which are function of the velocity and density of the overtopped flow, and can be expressed mathematically as (Ryu et al (2007) Where α is the promenade in time density factor and PE and KE are the time-averaged potential and kinetic energy respectively.The terms tgs and tge indicates the duration of the overtopping event and, and U and W are the mean velocities in the x and z directions.

FURTHER RESEARCH
In the present paper NIOVE PROMES project has been presented.The description of its instrumental setting and the methodology applied to obtain overtopping flow properties (velocity, density and volume) has been done.And finally, conventional instrumental setting and lab tests in which NIOVE PROMES will be applied, have been described.
The HRL will keep working on the measurement of overtopping events with the aim of getting deeper knowledge of this phenomenon, and how to minimize the hazards that it may cause to port´s infrastructures and exploitation.
NIOVE PROMES further research may be summarized in the following steps: 1. Apply the presented methodology to different sea defense´s typologies, and compare its results against conventional instrumentation.2. Use a set of measurements from all of the instrumental techniques applied for numerical model calibration.3. Adapt the laboratory instrumental setting to prototype, and obtain measurements of real overtopping events.4. Determine overtopping vulnerable areas, and provide necessary data to calculate the risk associated to overtopping events. Figur Figur to ob withProm P The over velo over Figur posit dry)

Table 1 . Wave conditions and principal structure parameters established for the tests realized at HRL´s wave flume
):