TARAZI DSHISER LLCHILDERS REBOLDT CAJournal of Water Pollution Control Federation, Vol 42, No 5, Part I, P 708-732, May 1970 10 Fig, 11 Tab, 6 Ref2010-02-152010-02-151970 Mayhttp://hdl.handle.net/1969.3/23239VOL-732A SIGNIFICANT DIFFERENCE EXISTS BETWEEN THE DATA OBTAINED BY GR SAMPLING AND BY FLOW-WEIGHTED COMPOSITE SAMPLING. THE SIGNIFICANCE OF THESE DIFFERENCES WAS MAINLY DEPENDENT ON THE FOLLOWING FACTORS: (1) THE OVERALL TIME PERIOD FOR WHICH THE WASTE WATER IS TO CHARACTERIZE, I.E., WHETHER DAILY OR OTHER TIME INCREMENTS OF POLLUTION LOADING ARE TO BE DETERMINED, (2) THE CONFIDENCE LEVELS DESIRED, (3) THE VARI SPECIFIC WASTE WATER STREAM, AND (4) WHETHER CONCENTRATION LEVELS OR MATERIAL TRANSPORT RATES ARE DESIRED. A COMPARISON IS MADE BETWEEN GR INDUSTRIES ON THE HOUSTON SHIP CHANNEL. FROM PAST EXPERIENCE THAT THE NUMBER OF GR CONFIDENCE LIMITS TO A FLOW-WEIGHTED COMPOSITE IS EXCESSIVE FOR MOST OF THE WASTE WATER ENTERING THE SHIP CHANNEL. TWO WASTE WATER STREAMS WERE UTILIZED IN THIS DEMONSTRATION. DAILY COMPOSITE AND GR THE FIRST INCLUDED PH, TOTAL RESIDUE, FILTERED COD, TOD, TOTAL FILTERED CARBON, AND INORGANIC FILTERED CARBON. THE SECOND STREAM INCLUDED ONLY TOTAL, FILTER 'T' TEST IS USED TO STATISTICALLY ANALYZE THE COLLECTED DATA AND DEVELOP CONFIDENCE LIMITS. (HANCUFF-TEXAS)*'T' TESTS*MONITORING*SAMPLING*SAMPLING TECHNIQUES*STATISTICSANALYTICAL TECHNIQUESAUTOMATIONEvaluationGRAB SAMPLESHoustonHouston Ship ChannelINDUSTRIESPOLLUTIONPROBABILITYSTATISTICAL METHODSW 3040 Wastewater treatment processesTEXWASTE WATERWASTE WATER CHARACTERISTICSWaste water treatmentWATERJournal