Data interpretation is difficult under the simplest of conditions and requires hydrologic studies that are carefully designed. Good research presupposes carefully collected data, as well as data that were obtained at times when the information content was highest. To accomplish the careful design of hydrologic studies and to obtain quality data, the U.S. Geological Survey works through small organizational units. The principles of a well-designed study are: 1) Use the scientific method; 2) define the problem and the approach; 3) state the objectives of the data collection; 4) determine what and how to sample; 5) decide where to sample; 6) begin report preparation early in the study; 7) do the work with little or no rework; and 8) complete the report on schedule.

Two water-quality synoptic studies were made on the Colorado River in the Grand Canyon, Arizona. Field measurements and the collection of water samples for laboratory analysis were made at 10 mainstem and 6 tributary sites every 6 hours for a 48-hour period on November 5-6, 1990, and again on June 18-20, 1991. Field measurements included discharge, alkalinity, water temperature, light penetration, pH, specific conductance, and dissolved oxygen. Water samples were collected for the laboratory analysis of major and minor ions (calcium, magnesium, sodium, potassium, strontium, chloride, sulfate, silica as SiO 2), trace elements (aluminum, arsenic, boron, barium, beryllium, cadmium, cobalt, chromium, copper, iron, lead, lithium, manganese, molybdenum, nickel, selenium, thallium, uranium, vanadium and zinc), and nutrients (phosphate, nitrate, ammonium, nitrite, total dissolved nitrogen, total dissolved phosphorus and dissolved organic carbon). Biological measurements included drift (benthic invertebrates and detrital material), and benthic invertebrates from the river bottom. Laboratory Several laboratory instrumental methods were used for the analysis of samples collected during the synoptic studies. Techniques were selected based on their optimal use for specific measurement criteria, primarily accuracy, precision, sensitivity, and selectivity. These techniques have been developed over a period of several years for the specific application to environmental water analysis. Alkalinity was determined using an automated Gran titration method (Kramer, 1982). Titrations were performed on a Radiometer auto-titrator using the supernatant portions of the sample after letting the sample settle for at least 24 hours and titrated with a 0.1 N standard sulfuric acid solution. Major cations were determined using inductively coupled plasma-atomic emission spectrometric (ICP-AES) analysis. Specific details of the method and techniques are described in Garbarino and Taylor (1979). The elements calcium, iron, magnesium, silica (as SiO 2), sodium, and strontium were determined simultaneously on a sample aliquot that had been filtered (0.4-µm) and acidified (HNO 3) in the field, using ultraclean procedures. Potassium was determined on the same aliquot by an air/acetylene flame atomic absorption spectrometric technique (Fishman and Friedman, 1985).

Four different bottom‐placed artificial substrates were compared with the Ponar grab for collecting benthic invertebrates. Artificial substrate samples of organisms were larger and more diverse than those of the grab. Barbeque Basket samplers caught the most taxa and individuals and Beak Trays caught the least. Chironomids and crustaceans were dominant in artificial substrate samples. Exposure habitat (left or right bank) determined taxa availability, whereas sampler design determined suitability for colonization by the taxa. Diversity for Beak Tray samples was lower than that for other artificial substrates but higher than for Ponar samples. The Barbeque Basket, Bull Basket, and Multiple Plate samples were taxonomically similar. Ponar samples were different, and Beak Trays were of intermediate similarity. As qualitative samplers, Barbeque Baskets were 63 percent efficient, followed by Bull Baskets (55 percent), Multiple Plates (48 percent), Beak Trays (38 percent), and Ponar Grabs (6–10 percent). Bull Baskets required the least and Beak Trays the most replicates to be within a preselected percentage error of the mean at the 95 percent probability level for numbers of taxa and individuals, and for diversity. Under conditions of the study, Bull Baskets ranked highest, followed by Barbeque Baskets and Multiple Plates, in selected performance criteria. Differences between grab and artificial substrate samples are explainable in terms of major riverine habitats and characteristics of the collection methods.

No abstract