Relatively stringent accuracy and precision requirements for well casing surveys and associated water-level measurements are generally difficult to achieve. Field experience and research indicate that the achieved accuracy is commonly less than expected. Realistic accuracy expectations require that the investigator know what is achievable for a site and how to convey the proper information to the surveyor. The point-difference accuracy of vertical surveys is calculated, based on the measured inaccuracy of the vertical survey divided by the square root of the length of the survey, and therefore by the size of the study area and number of wells surveyed. The accuracy of water-level measurements and contour maps is determined primarily by the equipment and procedures used for measuring the depth to water in the well casings. When setting accuracy requirements for water-level measurements and contour maps, the investigator must consider the end use of the data. Composite error from the two primary sources of error determines how much precision and accuracy are reasonably achievable.

Cognizant scientist. Prepared test plan and designed experiment. Supervised performance of the test. Prepared analytical service request. Interpreted data and reportexi results. Hot cell technician. Performed test. Statistical analysis of data. Managed chemical and radiechemicai analytical work. Technical reviewer. Task Leader. 3.0 Experimental " %nmle Descri~tion. The sample used in this test was labeled as C104-GL. The C-104 HLW sample was composite as described in test plan BNFL-29953-031, C-104 San@e Co@osititzg.Figure 3.1 summarizes the compositing and sub-sampling scheme. The C-104 sample was received from Hanford's 222-S Laboratory on March 3, 1999. This material was received in 14 glass jars. Figure 3.1 lists the sample numbers along with the mass of material recovered from each jar. The material in the jars was transferred to a winless steel mixing vessel equipped with a motorized iinpeller. Before being used, all components of the mixing vessel were rinsed with methanol and then dried at 102°C for 12 h. Materials in the vessel were mixed for 1 h and 20 min before collecting sub-samples. The materials were actively mixed while sub-samples were collected through a 1.9-cm (.75-in.) ball valve located on the bottom of the vessel. The hot-cell temperature during the mixing process was 34"C. The first three sub-samples (C-104 COMP & B, and GL) were collected and allowed to setde. After approximately 10 days, the volume of settled solids in these three samples was measured to determine the effectiveness of the sub-sampling technique at collecting samples with representative solids/liquid ratios. The three sub-samples contained 88.9, 89.2, and 89.9 volO/osettled solids indicating that the sampling technique provided representative sub-samples. C-104"AsR~ived" samples Sample # Weight g Sample# Weight g c-'Aa Reoeivad" Analvticd Samdes G104(%mp A 168.9g C-104 Comp B 170.3 g C-104 Corn E 125.2 c-lcomposi:s'mogen WIi Solublitv vs. Temperature and El

No abstract