OpenMI-Life Steering Committee meeting minutes



Technical Assistance

(Implementation Support) and

Management Consultancy

ADCP – How to specify an ADCP System and what components are required

1. Introduction

An Acoustic Doppler Current Profiler (ADCP) system offers users a new approach to measuring discharge in rivers in India using high-frequency sound waves (ultrasound). Under HP-I a small number of ADCP systems were procured by CWC and have been applied fairly successfully over the past 8-10 years. However, the technology behind ADCPs has moved on significantly over the past 5 years or so, particularly as far as the electronics and software are concerned and such equipment is now very widely used in the USA, China and Europe.

ADCP technology can be applied to large or small rivers, but is recommended for use under HP-II on large wide rivers in India, where sediment loads are often high. For small rivers, TAMC suggests that conventional current meter gauging by either wading or from boats offers the most appropriate means of measuring discharge.

This note does not attempt to describe the principals behind the ADCP method of river discharge gauging, but is intended to help IAs under HP-II to understand what components are required for an operational system and how to produce a suitable specification of a system under World Bank Procurement rules. However, useful reviews of how ADCPs work are available in a number of US Geological Survey (USGS) publications many of which can be accessed through their website: and a good reference report is available at: . However, some of the reports on instrument performance which exist on the USGS website refer to older versions of instrument or software and technical improvements in both may now invalidate any accuracy or comparison data quoted.

2. ADCP components

An ADCP system consists of a series of components that together make up an operational direct river discharge measuring package. A full system for use in India should consist of:

i. An ADCP instrument unit containing all of the required electronics, batteries, compass and tilt sensors etc;

ii. A small (1 – 1.2m long) flotation platform for deployment of the ADCP unit (normally a plastic or fibreglass trimaran or boat);

iii. A data transmission system to transmit data from the ADCP to a receiver on the river bank. This may be a Bluetooth system (cheap but has a limited range of 100m or perhaps up to 200m) or a radio modem system (more expensive but with a range of several kms);

iv. A robust laptop PC to receive and process the data on the riverbank;

v. A Differential Global Positioning System (DGPS) instrument to monitor the position of the ADCP on the river (needed in situations where the river bed is unstable and where bed material is moving, which is probably the normal situation in Indian rivers during the monsoon);

vi. Necessary connection cables and spare batteries etc plus manuals.

For Indian use, where ADCPs are probably going to be deployed on wide (>150 -200m) rivers with high sediment loads and mobile beds, it is suggested that:

i. A radio telemetry system is required rather than the Bluetooth option;

ii. A 600 kHz transducer system might be better able to cope with high sediment loads than a 1200 kHz or 1.0 MHz system (see comments on transducer frequencies in following section);

iii. Because it is likely that rivers will have moving beds during periods of high flows, systems should be equipped with a DGPS function.

3. Comments on available commercial ADCPs

ADCPs were first developed only some 20-25 years ago, but development has been rapid and the technique is now widely used as stated in Sect 1 above. Original development was primarily by a company called Teledyne RDI in California in the USA ( or ), for marine use but the company worked closely with the USGS in adapting the technology for use in river flow gauging. However, a second US company, Sontek/YSI (), has also been producing a similar product for the past 15 years or so, and their product is widely used in the USA and China for example. The term ‘ADCP’ is in fact copyrighted by RDI, and Sontek refer to their equipment as ADPs (Acoustic Doppler Profilers), however an RDI ‘ADCP’ is in exactly the same type of streamflow measuring instrument as a Sontek ‘ADP’.

Until recently RDI produced two instruments, a small one (StreamPro) for small rivers (up to 2.5-3m deep) and a larger version (Rio Grande, or ‘large river’ in Spanish) for rivers up to 30-100m deep. However, the Rio Grande is being replaced at the moment by a newer version called a RiverRay, which is supposed to be much easier to use and which requires less specialised knowledge by users. All RDI instruments use a 4 transducer array set at a 20o beam angle, with river depth being measured using this transducer array.

Sontek also have two instruments, an S5 for smaller streams (up to 5 m deep), and a larger M9 for large streams up to 40 to 50 m deep. The M9 is interesting as it contains two separate transducer ranges, one of 3.0 MHz for ‘small to medium depth rivers’ with limited sediment, and another of 1.0 MHz for large rivers with a higher sediment load, and is thus effectively two systems in one unit. The S5 has a similar array of 4 velocity measuring transducers as the RDI setup (although with a 25o degree beam angle) but includes a fifth vertical transducer to measure depth directly. The M9 has nine transducers, two sets of 4 with different frequencies for different modes plus a single vertical depth measuring transducer. This approach differs from the current RDI range where two Rio Grande options are available, one with a 1200 kHz frequency for medium rivers and one of 600 kHz for large, sediment laden rivers. There is no real advantage in Sontek over RDI in my opinion, as the 3.0 MHz option within the M9 would be of little use in deep, sediment laden rivers where it is to be used in India; the 3.0 MHz frequency is not suitable for rivers with a high sediment load.

Recently the German company, Ott, has produced a Qliner ADCP, for use in small to medium rivers (maximum 20m depth) (). This instrument does not yet have much of a ‘track record’ and has only been available for two years, and it is suggested that it is not ideally suited to Indian conditions and hence should not be considered for HP-II.

ADCP – Summary of differences between different models

|Frequency |Instrument |Good Points |Bad Points |

|High - >2 MHz |RDI StreamPro 2000kHz |Small bins (but high noise error)|Most susceptible to moving bed |

|(> 2000 kHz) |Sontek RiverSurveyor M9 3.0 MHz mode |Better accuracy for same bin size|Not good for high sediment load |

| | | |rivers |

|Medium – 1-2 MHz |RDI Rio Grande 1200 kHz |Versatile and suitable for most |Not suitable for very shallow |

|(1000 – 2000 kHz) |Sontek RiverSurveyor M9 1.0 MHz mode |conditions (can be used on small |rivers (< 1m) |

| | |and larger rivers) |May be problems in floods & high |

| | | |sediment loads |

|Low - < 1 MHz |RDI Rio Grande 600 kHz |Robust measurements |Big bins (low noise error) |

|(< 1000 kHz) |Sontek RiverSurveyor M9 1.0 MHz mode |More ‘energy’ to get through |Suitable for large rivers (>3m |

| | |sediment laden rivers (600 kHz |depth) and high sediment loads |

| | |probably better than 1.0 MHz) | |

|Note: 1 MHz = 1000 kHz | | | |

Tests were carried out on a number of different RDI and Sontek instruments in a current meter test tank some years ago in the USA (see ). However, these tests were carried out more than 7 or 8 years ago and results may not be applicable to current equipment sold by these manufacturers. Nevertheless, the paper demonstrated that the 1200 kHz RDI instrument was most accurate over a range of velocities but followed closely by the 600 kHz RDI. A Sontek 1500 kHz ADCP was somewhat less accurate overall, and their 500 kHz instrument showed considerable scatter in velocity measurement. A recent comparison of ADCP products by the national hydrometric authority in the UK (the Environment Agency) found that the RDI Rio Grande was superior to the Sontek RiverSurveyor for a number of reasons. However, Sontek products have a good reputation worldwide and have a proven track record. For HP-II, either a Rio Grande 600 kHz model (or the new RiverRay model), or a Sontek M9 RiverSurveyor would be suitable.

Frank Farquharson

6th November 2009

-----------------------

HP II

Indian Hydrology Project

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download