SonTek 3.0-Mhz RiverCAT ADP Demonstration on the

Transcription

1 SonTek/YSI, Inc Nancy Ridge Drive Suite A San Diego, CA USA SonTek 3.0-Mhz RiverCAT ADP Demonstration on the Orange River River near Aliwal North, Free State, South Africa Results and observations of the SonTek/YSI, Inc. 3.0-Mhz RiverCAT ADP demonstration on the Orange River near Aliwal North, Free State, South Africa, January 29 th - 30 th, 2003 By: John V. Sloat, Principal Hydrologist, SonTek/YSI, Inc.

2 Instrument Configuration, Deployment Description, Results, and Conclusions A 3-beam 3.0-Mhz RiverCAT ADP was deployed in three ways (towed from cableway, attached to inflatable boat and towed, attached to an inflatable boat and motored across) to collect discharge data on the Orange River Aliwal North, Free State, South Africa, January 29 th - 30 th, 2003 Equipment Configuration of the 3.0-Mhz SonTek RiverCAT A 3.0-Mhz SonTek RiverCAT was deployed to measure discharge for the purposes of both demonstration and training in the Orange River near the town of Aliwal North, Free State, South Africa. The measurement was located near a section of river where a bank-operated cableway has been installed. This is located approximately 1-km upstream of a weir and Water Affairs stream gauging station number D1H003. Traditional style measurements implementing a mechanical current meter to measure velocity and depth and using a panel method to compute discharge has been used routinely at this site to provide data used to develop a discharge rating for the weir. At the time of measurement, the river level was at a record low and no discharge measurements were available to input into the rating curve. By definition, the RiverCAT is a unique configuration of a RiverSurveyor ADP. SonTek/YSI, Inc. manufactures three RiverSurveyor ADP configurations (Standard, Splashproof, and RiverCAT) that are used to measure discharge in large and small rivers. At the request of the readers of this summary report, SonTek/YSI, Inc. will be pleased to provide additional information for each of the three (3) RiverSurveyor configurations that have been specially developed for the RiverSurveyor class of ADP instruments. For this demonstration SonTek/YSI, Inc. selected the RiverCAT configuration primarily because of the convenience for demonstration and the flexibility it provides to conduct a variety of deployment configurations. 2

3 The RiverCAT configuration is composed of a transducer housing, a watertight electronics compartment, and 2 sealed-aluminum hulls. This configuration is shown below in Figure 1. Alternative to the aluminum hulls used for this demonstration are options for slightly larger fiberglass hulls and a single-man inflatable Kayak. Each of the alternative configurations is shown in Figure 2. The transducer housing is specially designed to be small (10-cm diameter) and lightweight (3.5 kilograms). Inside the transducer housing is located the compass, tilt/roll sensor, and receiver electronics for each transducer. The transducer housing is connected to the main watertight housing via a high-frequency communications cable. The watertight electronics box contains the 24-volt power-supply, an internal RS-232 radio modem for communication to a computer located on the shore, a communications antenna, and electronics boards for the acoustic Doppler profiler (ADP). Aluminum braces are also used to firmly attach each aluminum hull to the watertight electronics box and to the transducer housing. The total weight of the RiverCAT is approximately 12.5 kilograms. data are transmitted (via the RS-232 radio modem) continuously from the RiverCAT to a laptop computer located on the shore within sight of the measurement vessel. The laptop computer is also connected to a RS-232 radio-modem (base-station) and communicates with the RiverCAT continuously. Using the laptop computer and the basestation radio-modem located on shore the RiverCAT is programmed and discharge data collected using the SonTek RiverSurveyor windows software. Unfortunately, during this study, an internal component of the radio-modem failed and thus measurements were made by directly connecting the computer to the RiverCAT via an RS-232 DB9 cable. Transducer- Housing Watertight Electronics Compartment Tether-line Aluminum-Hulls Figure 1: 3.0-Mhz SonTek RiverCAT configuration used in demonstration on the Orange River. 3

4 Figure 2: Picture showing alternative configurations of the RiverCAT: Fiberglass Hulls (left) and a single-man inflatable kayak (right). RiverCAT Deployment Description: The 3.0-Mhz RiverCAT was deployed using three mounting and crossing techniques: 1. Aluminum hulls towed next to an inflatable boat: In this instance, the computer and operator was located in the inflatable boat and the RiverCAT mounted to the aluminum hulls was towed next to the craft. The inflatable boat was then towed across the river using a large mechanical bank-operated cableway located at the site. 2. Fix-mounted over the bow of the inflatable boat and towed by cableway: In this case, and thanks to the ingenuity and abilities of the Water Affairs hydrologis ts, the RiverCAT was detached from the aluminum hulls and mounted over the bow (front) of the inflatable boat. The boat was then towed across the river by the mechanical bank-operated cableway. 3. Mounted over the bow of the inflatable boat and motored: This deployment was similar to deployment number 2 with the difference being the mean by which the RiverCAT was transected across the river. In this case, the inflatable boat was driven across the river using a motor. This deployment was done also to enable the measurement parties to conduct measurements at different locations in the river other than at the cableway. For each measurement, the RiverCAT was positioned alongside a riverbank and held nearly stationary within about 5 6 meters of the rivers edge. Data collection was started (using the RiverSurveyor software) and the RiverCAT was allowed to collect between 5 10 seconds of data before starting its course to the opposite riverbank. The primary reason for this is that data collected at the start of the measurement is used to estimate the flow from the measurement starting point to the actual rivers edge. In a general sense, the more stationary the vessel is the better the true starting distance from the bank can be measured resulting in more representative data estimates of the unmeasured flow. In addition, by collecting 5 10 seconds worth of data a better estimate can be made of the unmeasured velocity section based on measured velocity data near that point. Once the data collection was started, the RiverCAT was steadily transected across the river to the opposite riverbank. During the measurement, the RiverCAT crossing speed was typically at or below the actual speed of the water. In addition, during each measurement, the pulling (towing) speed as well as the course (direction) of the RiverCAT was held as steady as possible in order to obtain the most accurate measurement possible. Depending on the deployment used during the demonstration, there were several times when the boat speed and 4

5 course varied more than an acceptable level, however, having this data provides a good example of the significance of a relatively steady boat speed and course relative to the repeatability and accuracy of the discharge data. In the following sections, a comparison between techniques is provided and shows the improvements as boat course, pitch, and roll is held steady while crossing varying boat speeds. Once the RiverCAT reached the opposite riverbank it was allowed to become stationary and collect data for 5 10 seconds. This is for similar reasons as described during the start of each measurement. Typical measurement durations ranged from 3 4 minutes to about minutes depending upon the method used to cross the river. In practice, measurement durations between 8 10 minutes are typical for a river of similar geometry. The maximum depth was nearly 2 meters with the average depth being closer to about 1.5-meters. This is considered relatively shallow water for this type of instrument and as such slower, steadier boat speeds will have a tendency to have a more significant impact on the accuracy and repeatability as opposed to a river of similar width but twice as deep. 5

6 Figure 3: RiverCAT deployment on the Orange River: Fixed-mounted over bow of inflatable vessel. Figure 4: Mechanical bank-operated cableway motor and assembly with Water Affairs operator. 6

7 Figure 5: Cableway towing inflatable boat during RiverCAT measurement on the Orange River near Aliwal North, Free State, South Africa, January 29 th 30 th, RiverCAT Measurement Results: measurements were made for the purposes of both demonstration of the technology and training over a two-day period. As previously mentioned, three methods were used to collect discharge data using the RiverCAT. In addition, following brief instructions, several members of the Water Affairs collected data independent of other parties in an effort to maximize the training effort on the second day. results presented in this report have not been edited and appear just as they were collected in the field. In practice, once the data collection is complete, the hydrologist will post-process (using the RiverSurveyor software) the data to ensure its reliability and to quality check the results. In some instances, parameters such as edge distances, velocity reference (bottom-track or Differential GPS), start/end points may be edited based on field results which can slightly change the final discharge value. Measurement Session One (Afternoon, January 29 th ) Measurements made on January 29 th, 2003 were made in the evening. Conditions during the measurement were windy and there was a moderate chop (small waves). In addition, the RiverCAT was towed alongside the inflatable boat resulting in an increased pitch/roll during the measurement. The discharge results were acceptable, however, they can be shown to be more variable from transect to transect when compared to measurements made on the following days. This increased variance is most assuredly on account of the increased pitching and rolling of the RiverCAT when towed next to the boat. As mentioned, the steadier the RiverCAT can be transected across the river, the better the replication and accuracy can be achieved. The average discharge was 88.3 m 3 /s with a standard deviation of 3.6 m 3 /s and a coefficient of variation (standard deviation/ average discharge) of 4.0 %. Table 1 shows the measurement start times and results of data collection during this period. 7

8 Table 1: Afternoon Measurement January 29th, cableway, motor boat using rivercat hulls towed next to motor boat. Start Time End Time # Profiles Top Middle Bottom Total 29/01/ :21:12 29/01/ :26: /01/ :15:01 29/01/ :18: /01/ /01/ Standard Deviation Coefficient of Variation 4.03% Average Measurement Session Two (Morning, January 30 th ) Measurements made on January 30 th, 2003 were made with the RiverCAT fixed-mounted to the inflatable boat (as shown in figure 4). Measurements made in the morning were made by using the boat s motor to cross the river and measurement sections were made at the cableway sight and a previous cableway site about 500-meters downstream. In this case, the water-surface was relatively calm and measurements can be seen to be relatively repeatable. In some cases, the boat speed exceeded an acceptable speed, however, results were still within an acceptable margin of error. The variation between measurements is noticeably smaller than measurements made the day previous which is a result of fixed-mounting the RiverCAT directly to the boat AND because the water-surface conditions were much calmer. The average discharge was 92.6 m 3 /s with a standard deviation of 2.24 m 3 /s and a coefficient of variation (standard deviation/ average discharge) of 2.4 %. Table 2 shows the measurement start times and results of data collection during this period. It should also be noted that the water level had risen from the previous day approximately 10-cm. which can be seen as an increase of about 4.5 m 3 /s from measurement made the previous day. The 10-cm rise was both visually observed and measured using an Argonaut-SL flow meter temporarily installed at the site on January 29 th. Lastly, the RiverCAT completed six discharge measurements within 1-hour 15-minutes time. Table 2: Morning Measurements, January 30th, fixed-mounted, motor, 2 different cross-sections, Jackie and John measuring. Start Time End Time # Profiles Top Middle Bottom Total 30/01/ :08:40 30/01/ :20: /01/ :22:35 30/01/ :26: /01/ :28:55 30/01/ :32: /01/ :33:39 30/01/ :38: /01/ :41:20 30/01/ :46: /01/ :13:24 30/01/ :17: Standard Deviation (cms) 2.24 Coefficient of Variation 2.42% Average (cms)

9 Measurement Session Three (Afternoon, January 30 th ) Measurements made in the afternoon were made by several different measurement parties and also included measurement made solely by Water Affairs personnel. In contrast to measurements made earlier in the day these measurements were all made at the location of the cableway and were made by towing the boat first with the cable, then motoring across the river without the assistance of the cable. The variation of these measurement were even further reduced when compared to measurements made earlier in the day. Most likely this was due to the fact that only one measurement section was used during data collection. The average discharge was 92.8 m 3 /s with a standard deviation of 1.52 m 3 /s and a coefficient of variation (standard deviation/ average discharge) of 1.64 %. The discharge results are nearly an exact match of data collected in the morning training session. For this measurement session, the RiverCAT made 5 measurements within about 1-hour 20-minutes time. In reality, considering that the measurements on average will take between 8 12 minutes, it is reasonable to state that 5 8 complete discharge measurements could be made in about 1-hours time at this site. The benefit of making replicate measurement is the observer has the ability to not only make a good estimate of the total discharge, but also evaluate the variability of the measurement and make an reasonable assessment if the variance is acceptable or not. In the case when it may appear to be too high, this may lead the observer to choose another measurement location or operate the boat differently across the river. Table 3: Afternoon Measurements, January 30th, cableway, mixed Water-Affairs measurement parties Start Time End Time # Profiles Top Middle General Results Bottom Figure 6 shows the RiverSurveyor software screen as was viewed during each discharge measurement. Although, this figure represents only a single measurement, it does represent the general characteristics of the river during each day of data collection. The figure shows the maximum depth near 2-meters, a minimum depth of about 0.8-meters, a cross-section width of about 151-meters, and velocities averaging around cm/s for most of the cross-section. Figure 7 shows the discharge calculation screen that is used in post-processing by the observer. The discharge calculation screen allows the observer to adjust start and end locations, determine what type of bank slope to use, and to determine the true orientation of the river (i.e. what is considered the downstream direction). In addition, it also provides quality parameters that at times are useful to recognize a potential reason for high-variance or to assure the observer that data was collected properly. These quality parameters can be viewed on the bottom right section of the discharge calculation screen. Total 30/01/ :47:19 30/01/ :03: /01/ :28:10 30/01/ :33: /01/ :34:35 30/01/ :40: /01/ :49:50 30/01/ :55: /01/ :55:55 30/01/ :02: Standard Deviation (cms) 1.52 Coefficient of variation 1.64% Average (cms)

10 Figure 6: Real-Time RiverSurveyor Data for Orange River Measurement 10

11 Figure 7: RiverSurveyor Summary Sheet for Orange River Measurement 11

12 Report Summary and Conclusions: A SonTek 3.0-Mhz RiverCAT was demonstrated on the Orange River near Aliwal North, Free State, South Africa. During the demonstration participants were able to witness first-hand the measurement RiverCAT discharge measurement process, the ease-of-use programming for data-collection, the setup of equipment, view first-hand RiverCAT realtimne data collected with the RiverSurveyor software, and analyses discharge results. The measurement conditions were most acceptable for demonstration and training. The discharge reported from three separate measurement sessions was 88.5, 92.6, and 92.8 m 3 /s, respectively. This first measurement session occurred in the evening of January 29 th, 2003 and the following measurements were made in the morning and again in the evening of January 30 th, There was also observed a 10-cm. rise in water level between the first and second measurement, which most likely accounts for the increase in flow on the second and third measurement sessions by the RiverCAT. Most notable of the results is the replication that was achieved considering the objectives of the demonstration and training. In this case, data represents measurement made: at different times; at 2 3 different locations; measured by different parties, including several conducted solely by Water Affairs personnel; and using several towing techniques. The maximum variance of the measurement was about 4% and the minimum was only 1.5%, which in the opinion of the author of this report, suggests several things, namely: 1. The site, for the given conditions, chosen for measurement is well suited for the technology, 2. The measurements although made inconsistently, replicated well suggesting that even though conditions are not perfect, the robustness of the technology and the ability of the personnel collecting data was well within the criteria necessary to collect accurate, repeatable, and meaningful data, 3. The results of future measurements at this site will improve even further once a formal operational procedure is outlined, and finally, 4. measurements using the RiverCAT made at other sites similar to the Orange River measurement location can be expected to achieve a similar level of performance. Myself, and SonTek/YSI, Inc. would personally like to thank all the participants for attending this demonstration of the SonTek RiverCAT. A very special thanks goes out specifically to the members of the Water-Affairs that participated in the demonstration as well as organized and prepared equipment enabling a successful few days in the field. In all sincerity, I have worked with numerous water agencies around the world in my career as a hydrologist and rarely have had the opportunity to work beside such an experienced, dedicated, and attentive staff. It truly made my first field experience in South Africa a genuine pleasure. If there are any questions about the information and data contained in this report please feel free to contact me directly by voice: (858) or via john@sontek.com. At your request I will also be glad to provide additional specifications and recommendations specific to your required discharge measurement applications. Best Regards, John V. Sloat, Principal Hydrologist SonTek/YSI, Inc. 12