Technical Studies


Improving Sediment Monitoring Technique and Establishing Pilot Monitoring Stations (2/2)

The rivers and creeks in Taiwan have characteristics of small basin, high flow velocity, and rapid change between base flow and storm flow, most transportation of suspended sediment occurs in a few torrential rain incidents. While suspended-sediment concentration (SSC) can be observed by manual sampling and lab testing, but it is difficult to predict the right sampling timing and mobilize field crew during storms. The SSC in river has high spatial and temporal variability, but durable, efficient and automated techniques for high SSC monitoring are yet to be discovered. To overcome the aforementioned problems, a new method based on time domain reflectometry (TDR) was proposed and a pilot SSC automatic monitoring station at Nei-Wan was established to evaluate TDR performance and compare with other commercial instruments. The objectives of this study are to continue developing improved TDR method and set up new monitoring stations to collect more data for comparison and further analysis.

To develop improved TDR method, both the hardware and software were reexamined. The TDR probe was better designed with the assistance of 3D EMF modeling. Based on the waveform simulations, a new travel time analysis method was proposed such that it is less affected by electrical conductivity and can be easily automated. The sensitivity and resolution of the TDR method were theoretically derived based on dielectric mixing model. Their influencing factors including probe length, data acquisition parameters, material properties, and temperature were identified. The calibration method and temperature compensation method were developed accordingly. These new developments results in a significant improvement of measurement accuracy. These developments improve the SSC accuracy drastically and the measurement is insensitive to electrical conductivity and particle size. The effect of cable length can also be removed through the improved and simplified calibration. The measurement range of the TDR method is theoretically unlimited, but the signal attenuation in high electrical conductivity environment may prevent effective measurements. In a normal fluvial environment, the measurement range is greater than 200,000 ppm. Unlike other techniques having a probe with built-in sensor, the TDR probe is simply a waveguide which can be easily made to fit different environments. The submerged TDR probe is connected to a TDR pulser above the water. The TDR probe does not have any electronic device, so it is rugged and can be economically replaced. Multiple TDR probes can be connected to a TDR pulser through a multiplexer and automated, hence increasing the temporal and spatial sampling. The invented TDR SSC waveguide can be combined with other TDR sensing waveguide, such as water level, souring, soil moisture content, and rainfall, to form an integrated TDR monitoring system for hydrology.

Three new automatic TDR SSC monitoring stations were established at Chi-Chi weir, Kao-Ping weir, and Shin-Gung reservoir this year. The Nei Wan station was upgraded to incorporate the new developments. All stations were fully automated for storm process monitoring. The monitoring data and site visits provided valuable experiences and helped discover several practical problems, including necessity of on-site calibration for probe parameters, modification direction for probe design and installation, importance of grounding for temperature sensors and data acquisition system, and countermeasures for noise in the TDR waveform. The Chi-Chi station was suggested to be the main research site. An automatic pump sampler was added and the optical turbidity meter and acoustic attenuation meter were moved to the Chi-Chi station from Nei Wan. The monitoring data shows that the TDR and acoustic methods gave higher SSC measurements than optical backscatter and pump sampler. The discrepancy requires further investigations. In addition to the development and field testing, some suggestions were gave regarding the strategy for suspended-sediment sampling and standard procedure for setting up a SSC monitoring station.

Future studies suggested include maintenance and improvement of established monitoring stations, further field testing and evaluation, and modification of TDR probe design, signal processing, and field installation. Development of a TDR data acquisition system that allows on-site data processing is also suggested so that real-time SSC warning and feedback to the pump sampler can be implemented.<!--EndFragment--> �l lmо�pan>