\ifx\pdfoutput\undefined \documentclass[12pt]{article} \usepackage[dvips]{graphicx} \else \documentclass[pdftex,12pt]{article} \usepackage[pdftex]{graphicx} \pdfinfo{ /Title (Float Placement and Anchor Deployment for MOOS Mooring) /Author (Andrew Hamilton) } \usepackage[pdftex,letterpaper,plainpages=false,breaklinks=true,colorlinks=true,urlcolor=black,citecolor=black,linkcolor=black,pdfpagemode=None,pdfstartview=FitV]{hyperref} \pdfcompresslevel=9 \DeclareGraphicsExtensions{.pdf,.png,.jpg,.bmp} %Redefine order so pdf is used if avail. \fi \usepackage{amsmath} \newif\ifpdf \ifx\pdfoutput\undefined \pdffalse \else \pdfoutput=1 \pdftrue \fi %Page formatting for ONR specifications \setlength{\topmargin}{0in} \setlength{\topskip}{0in} \setlength{\oddsidemargin}{0in} \setlength{\evensidemargin}{0in} \setlength{\headheight}{0in} \setlength{\headsep}{0in} \setlength{\textheight}{9in} \setlength{\textwidth}{6.5in} %\setlength{\columnsep}{.15in} %\setlength{\parindent}{0.5in} \title{MOOS Vertical Profiler} %\author{\Large A. Hamilton} \date{} \begin{document} \maketitle The vertical profiling features of the MOOS system are currently poorly defined and decisions need to be made regarding the best path to follow in developing or purchasing the equipment required to enable the profiling capability. This document captures the functional requirements for the MOOS profiler from a variety of other MOOS documentation and examines some of the system design implications of the functional requirements Several commercially available vertical profilers are available and it is the aim of this report to establish the suitability of these for use in the MOOS system. The key differences between the capabilities of previously deployed vertical profilers and the MOOS requirements is the need for extended deployments with near real-time telemetry. The telemetry needs are currently being addressed by some profiler manufacturers, but the power limitations of batteries remain a harder problem to solve. After outlining the MOOS vertical profiler functional requirements, the suitability of the various MOOS mooring configurations for hosting a vertical profiler is discussed. A power budget analysis is performed for a generic profiler and specific MOOS architecture requirements are discussed. Several available profiler solutions are then compared against the MOOS functional requirements for both the upper water column profiler and the benthic profiler. Finally, a concluding section presents recomendations on a reasonable path for MBARI to follow in developing the vertical profiling capabilities envisioned by MOOS. %Commercially available equipment by the name of vertical profiler is typically designed for relatively short deployments in which the device operates nearly autonomously, collecting and storing data for analysis upon recovery of the device. Within the MOOS architecture, there is the additional requirement than the equipment be capable of near real-time communication with the rest of the system. Also, it is desirable that the profiler be able to easily accept new instruments in a standardized manner. Because the MOOS ``Remote Instrument Node'' device (RIN) shares these requirements, a better name for the profiler in the MOOS architecture may be ``Vertically Mobile Instrument Node'' (VerMIN?). Adopting the point of view that ideally the MOOS vertical profiler would have all the capabilities of a RIN, and additionally be able to crawl up and down the riser cable, provides a starting point for specifying the profilers requirements. Determining what subset of the RIN's capabilities the profiler is capable of, in accordance with size and power limitations, is the object of specifying the MOOS vertically mobile instrument node. \section{Functional Requirements} The MOOS system includes a vertical profiling functionality in both the upper water column and the benthic region. The requirements are different and specified seperatly. \subsection{Upper Water Column Profiler} The MOOS system functional requirements documents makes several references to vertical profiling capabilities in the upper water column. The functional requirements presented here include those as well as other requirements that are relevant. \begin{itemize} \item The profiler shall be deployable in conjuction with at least a subset of the mooring configurations that comprise MOOS. \item The profiler shall operate between a depth of 500 meters and the surface. \item The profiler shall be capable of performing 6 full profiles per day. \item The profiler shall be capable of near real-time communications with the shore. \item The profiler shall support an instrument payload the weighs less than ? lbs in water. \item The profiler shall support an instrument payload that draws no more than 10 watts during profiling actions. \item The profiler shall be able to participate in the MOOS power and data architecture. \item The profiler shall be able to accept new instruments in the same manner as a RIN. \end{itemize} \subsubsection{Environmental Requirements} \begin{itemize} \item The profiler shall be operational whenever the sea conditions are within the ``90\% design condition'' for the MOOS mooring system. For the Outer Monterrey Bay site, these are: \begin{enumerate} \item Surface Currents $<$ 50 cm/s \item Significant wave height $<$ 5 meters \item Average wind speed $<$ 20 kts \end{enumerate} \end{itemize} \subsubsection{Operational/Maintenance Requirements} \begin{itemize} \item The profiler shall not need servicing more often that the MOOS Moorings themselves. i.e. one year between servicing. \end{itemize} \subsubsection{Derived Requirements} The profiling speed requirement is flexible and can be adjusted to maximize efficiency in the face of tradeoffs between instrument energy usage, propulsion energy usage, and duty cycle requirements for battery recharging or communications. A profiling speed of 15 cm/s is a rate that will allow the 6 required profiles per day to be done to 500m with a 50\% duty cycle. There is room in this requirement to optimize the speed for efficiency and the 50\% duty cycle requirement is retained below to investigate the trade-offs of designing for a shallower depth. \subsection{Benthic Vertical Profiler} The vertical profiling requirements are similar to the upper water column, however the sea state conditions are replaced by a specification of a maximum allowable benthic current. \begin{itemize} \item The profiler shall be deployable in conjuction with at least a subset of the mooring configurations that comprise MOOS. \item The profiler shall operate between the bottom and a heigh of 300 meters. \item The profiler shall be capable of performing 6 full profiles per day. \item The profiler shall be capable of near real-time communications with the shore. \item The profiler shall support an instrument payload the weighs less than ? lbs in water. \item The profiler shall support an instrument payload that draws no more than 10 watts during profiling actions. \item The profiler shall be able to participate in the MOOS power and data architecture. \item The profiler shall be able to accept new instruments in a relatively simple manner. \end{itemize} \subsubsection{Environmental Requirements} The profiler should be operational when benthic currents are less than the 90\% design criteria of ? cm/s. \subsubsection{Operational/Maintenance Requirements} \begin{itemize} \item The profiler shall not need servicing more often that the MOOS Moorings themselves. i.e. one year between servicing. \end{itemize} \subsubsection{Derived Requirements} The profiling speed requirement is flexible and can be adjusted to maximize efficiency in the face of tradeoffs between instrument energy usage, propulsion energy usage, and duty cycle requirements for battery recharging or communications. \section{MOOS Mooring Configurations} The MOOS mooring system as originally concieved is in fact a family of moorings, each suited to different applications. To date, most of the work has been done on the mooring configuration that delivers power and data to the seafloor via an EOM cable. This design requires the presence of a ``snubber'' element that is a large diameter hose descending approximately 16 meters from the surface before the smaller EOM cable takes over. Traversing this snubber element with an upper water column vertical profiler seems very difficult and greatly complicates design of the profiler. Furthermore, recent tests have brought to question the behavior of that EOM cable in rough seas and there are concerns about placing a profiler or a profiler docking station on the cable. For these two reasons, the simpler configuration of the MOOS mooring in which the EOM cable only extends to a depth of 500 meters is the best configuration for support of a vertical profiler. The use of low modulus mooring cable materials below the 500 meter mark allows the mooring to be deployed without the snubber elements, such that a profiler will be able to travel to a location much nearer the ocean surface, depending on wave conditions. Because the performance of the upper water column profiler greatly depends upon the behavior of the mooring riser cable, the original MOOS test mooring is taken as representative case for the ensuing analysis of profiler requirements. Figure \ref{fig:config} shows the configuration of that MOOS test mooring in a variety of surface currents, as predicted by the WHOI-Cable dynamics model. For the bentic profiler, concerns about the durability of the EOM cable make placing a vertical profiler on the lower section of the cable undesirable. Because the MOOS system enables the delivery of power and data to a remote location on the seafloor, there is no reason to not install a dedicated subsurface float for benthic profiling operations. \begin{figure} \begin{center} \includegraphics[height = 0.35\textheight]{Figures/config} \end{center} \caption{Configuration of the MOOS test mooring cable in various currents.} \label{fig:config} \end{figure} \section{Energy Budget} The energy budget for the vertical profiler is critical in determining a design that will meet the functional requirements, and will dictate the implementation of significant features such as a re-charging ability. Energy will need to be spent on three items, energy for propulsion to motivate the profiler up and down the cable, energy to power science instruments on board, and power for a profiler controller to interface and communicate with the rest of the MOOS system. \subsection{Propulsion Energy} The energy spent on propulsion is dependant on three main factors, the speed of profiling, the drag forces acting on the profiler (hydrodynamic and frictional), and the current the profiler is operating in. Figure \ref{fig:energy} shows the results of an analysis to estimate the energy required per profile, appendix \ref{app:analysis} contains the details of the analysis. \begin{figure} \begin{center} \includegraphics[height = 0.35\textheight]{Figures/energy-400m} \end{center} \caption{Energy spent per 100 meter round trip profile around a point 400 meters deep on the MOOS Test Mooring cable.} \label{fig:energy} \end{figure} Figure \ref{fig:energy} shows the quadratic dependence of energy spent on both the current speed and profiling speed and assumes a cylindrical profiler 1 meter long and .5 meters in diameter. Table \ref{tab:propulsion_energy} presents estimates of the total energy requirements over the long term for a variety of profiler depth ranges and speeds. In these estimates, an average current speed of 20 cm/s is assumed and the cable angle of the MOOS test mooringat at a 400m depth is used. More refined results could be estimated by integerating the energy usage over the depth of the profile but for a first approximation, the averageing technique is used. The last row in the table is given for comparison with actual results reported for the McLane Moored Profiler \cite{}. That report indicates that the MacLane profiler is performing about 10\% better than these estimates, reflecting the low-drag shape of the MacLane profiler and the fact that in those tests, the currents were lower and the cable was nearly vertical. \begin{table} \label{tab:propulsion_energy} \begin{center} \begin{tabular}{|c|c|c|c|c|c|c|} \hline Profile Depth & Speed & Energy/Day & Energy/Month & Energy/Year & Average Power \\ (meters) & (cm/s) & (Whr) & (Whr) & (kWhr) & (Watts) \\ \hline\hline 200 & 5.6 & 1.0 & 30 & 0.4 & 0.04 \\ \hline 300 & 8.3 & 1.8 & 55 & 0.7 & 0.08 \\ \hline 400 & 11.1 & 2.9 & 85 & 1.1 & 0.12 \\ \hline 500 & 13.9 & 6.75 & 200 & 2.5 & 0.28 \\ \hline 500 & 20.0 & 0.4 & 360 & 4.4 & 0.50 \\ \hline \end{tabular} \caption{Propulsion energy needs for a variety of profile depths/speeds (100 \% efficiency).} \end{center} \end{table} \subsection{Instrumentation Energy} The energy consumption of the instrumentation on the profiler will be variable in the MOOS system because the system is to be designed to accomodate a range of instrumentation. In order to generate an estimate of the energy requirements, table \ref{tab:instrument_energy} shows the energy usage of a number of common instruments. In these estimates, it is assumed that the profiler is moving at 15 cm/s such that an entire 500m upcast takes approximately 1 hour. Hence the instruments are active 6 hours per day. Clearly, speeding up the profile reduces the energy spent on instrumentation and once instruments are selected and the exact ``hotel load'' of the profiler is established, an optimal speed exists at which the propulsion power is one-half of the hotel load. \begin{table} \label{tab:instrument_energy} \begin{center} \begin{tabular}{|c|c|c|c|c|c|} \hline Instrument & Energy/Day & Energy/Mo & Energy/Yr & Avg. Power \\ & (Whr) & (Whr) & (kWhr) & (Watts) \\\hline\hline CTD - SBE49 & 21 & 630 & 7.7 & 0.9 \\ \hline CTD - FSI Excell uCTD & 1.7 & 50.4 & 0.61 & 0.07 \\ \hline Nitrate-ISUS & 45 & 1350 & 16.4 & 1.9 \\ \hline CO2-ISUS & 45 & 1350 & 16.4 & 1.9 \\ \hline UpWelling Irr-Hydrorad 2 & 9.9 & 298 & 3.6 & 0.4 \\ \hline Clorophyll Flour.-HobiLabs HS2 & 26.5 & 794 & 9.7 & 1.1 \\ \hline Current/Zooplank. Dens.- RDI ADCP& 33.0 & 990 & 12.0 & 1.4 \\ \hline Current - FSI 2D-ACP & 2.0 & 61.2 & 0.74 & 0.09 \\ \hline BioLum - Bathyphotometer & 36.4 & 1093 & 13.3 & 1.5 \\ \hline PhytoPlank. Eff. - FRR Flouro. & 54.1 & 1622.0 & 19.7 & 2.25 \\ \hline MOOS Mooring Controller & 24.0 & 720.0 & 8.8 & 1.0 \\ \hline Seabird Inductive Modem & 4.8 & 144.0 & 1.75 & 0.2 \\ \hline \end{tabular} \caption{Energy consumption of various instruments likely to be hosted by the profiler ($V = .15 m/s$.} \end{center} \end{table} \subsection{Controller and Communication Energy} The requirement that the profiler participate in the MOOS architecture dictates the capabilities of the controller that manages the profilers actions. The MOOS Mooring Controller that is currently being developed at MBARI appears to be a good candidate for this role, allowing the vertical profiler to appear to the system as any other BIN. This choice would leverage the work already done in integrating the controller into the MOOS architecture for the controllers on the buoy itself and on the BINS. Because this controller is still under development, power requirements are not yet settled, although it appears that an average of 1.0W is a conservative estimate. Excessive power usage and the large size of this controller are the downsides of using this design on a profiler itself. However, the advantages of having the software infrastructure that interfaces with the MOOS system may outweigh these drawbacks. Communication between the profiler and the MOOS System is a critical element of the development to meet the near real-time data transfer functional requirement. For cable mounted profilers, inductive modem technology is attractive from a simplicity and energy efficiency standpoint. However, the data rates of the commercially available modems are limited to 4800 baud (50 Mb/Day). The last row of table \ref{tab:instrument_energy} shows the estimated daily, monthly, and yearly energy needs for the controller and the inductive modem system. \subsection{Impact of Energy Considerations} \section{Types of Profilers} A number of commericially available profilers exist and can be cateragorized by the propulsion method. No products exist that meet the MOOS needs exactly, the requirements of long deployment and near real-time communications being the specific requirements that are generally not met. However, it is worth examining the available options in an attempt to estimate where the shortcomings are and to see if one of the commericial options might serve as a base platform upon which the MOOS profiler could be created by adding the needed telemetry and power recharging capabilities. \subsection{Crawlers} The Mclane Moored Profiler, produced by McLane Research Labs is a profiler which attaches to a cable and propels itself up and down via a direct cable drive system. This system is best suited to use with a subsurface float that is not affected by wave effects. Engineers at McLane Research report that efforts to use this profiler below a surface bouy have failed as a result of the cable motion. The cable motion draws the profiler up and down, increasing drag and reducing the depth resolution of the science data. As a result, this solution is inappropriate for use as the MOOS upper water column profiler, but may be a good candidate for the benthic profiler. This is especially true because this profiler is designed and tested for use in full ocean depths (6000m). \subsection{Bouyancy Driven} Bouyancy driven profilers such as the MBARI vertical profiler (MVP) are potentially very efficient in terms of propulsion energy and have the advantage of being decoupled from the cable motions. For these reasons, the MVP is a decent candidate for development into the MOOS profiler. For this profiler to meet the MOOS requirements, several shortcomings in the design need to be overcome. First the amount of available bouyancy change needs to be increased. Figure \ref{fig:hold_angles} shows the results of an analysis undertaken to determine the minimum bouyancy change required in order for the profiler to descend against a current on the MOOS test mooring, the colored lines indicate the maximum cable angle that the profiler can descend on for various currents, overlayed in black is the predicted angle of the MOOS test mooring. The MVP's (+/- .5lb bouyancy) ability to descend in a 30 cm/s current is marginal. Increasing the bouyancy range of the profiler by a factor of four or more is required to have an adequate factor of safety. Beyond this change, the depth rating of the MVP would need to be increased to 500 meters and the telemetry and power transfer needs addressed. \begin{figure} \begin{center} \includegraphics[height = 0.35\textheight]{Figures/hold_angles} \end{center} \caption{Ability of a bouyancy driven profiler to descend on an angled cable in a current.} \label{fig:hold_angles} \end{figure} \subsection{Wave-Powered} Brook Ocean Technology is producing a wave-powered profiler known as the Seahorse Profiler. This profiler utilizes a one-way rachet clamp to harness the motion of the mooring cable due to ocean waves. This operation is such that the profiler ratchets down the cable and floats freely up the cable and is thus decoupled from the cable motion during the measurement phase. These profilers are typically ballasted 6 lbs bouyant and have an upward profiling speed of .5 meters per second. This high speed reduces the instrumentation power needs significantly because the instruments can spend much more time in low power modes. Table \ref{tab:instrument_energy-.5} shows a recalculation of the total intstrument energy needs based on this higher profiling speeds. The potential risks associated with this profiler are concerns about the durability of the rachet clamp over the long term, potentially damaging effects on the mooring riser cable due to the rachet clamp, and the behavior in very calm seas. The manufacture reports some downward motion of the profiler in sea states that are much milder than would be expected for even a short time in the eastern pacific. \begin{table} \label{tab:instrument_energy-.5} \begin{center} \begin{tabular}{|c|c|c|c|c|c|} \hline Instrument & Energy/Day & Energy/Mo & Energy/Yr & Avg. Power \\ & (Whr) & (Whr) & (kWhr) & (Watts) \\\hline\hline CTD - SBE49 & 21 & 630 & 7.7 & 0.9 \\ \hline CTD - FSI Excell uCTD & 1.7 & 50.4 & 0.61 & 0.07 \\ \hline Nitrate-ISUS & 45 & 1350 & 16.4 & 1.9 \\ \hline CO2-ISUS & 45 & 1350 & 16.4 & 1.9 \\ \hline UpWelling Irr-Hydrorad 2 & 9.9 & 298 & 3.6 & 0.4 \\ \hline Clorophyll Flour.-HobiLabs HS2 & 26.5 & 794 & 9.7 & 1.1 \\ \hline Current/Zooplank. Dens.- RDI ADCP& 33.0 & 990 & 12.0 & 1.4 \\ \hline Current - FSI 2D-ACP & 2.0 & 61.2 & 0.74 & 0.09 \\ \hline BioLum - Bathyphotometer & 36.4 & 1093 & 13.3 & 1.5 \\ \hline PhytoPlank. Eff. - FRR Flouro. & 54.1 & 1622.0 & 19.7 & 2.25 \\ \hline MOOS Mooring Controller & 24.0 & 720.0 & 8.8 & 1.0 \\ \hline Seabird Inductive Modem & 4.8 & 144.0 & 1.75 & 0.2 \\ \hline \end{tabular} \caption{Energy consumption of various instruments likely to be hosted by the profiler ($V = .5 m/s$).} \end{center} \end{table} \subsection{Other Options} \section{Conclusions for MOOS Program} \section{Risk Items} \begin{appendix} Forces acting on profiler \end{appendix} \end{document}