Two methodologies for assessing boron in quaternary salar and lacustrine settings

Abstract

Many of the attempts by the U.S. Geological Survey and other groups to assess the mineral resource potential of selected areas have produced qualitative results, i.e. one area is more favorable or has a higher potential for mineralization. A relatively small subset of assessments has produced quantitative estimates; many of these were based on subjective quantitative estimation of number of deposits Very few assessments, especially quantitative assessments, have seriously addressed the potential of industrial minerals. Quantitative assessment of industrial minerals by the U.S. Geological Survey is hindered by a relative lack of expertise in industrial minerals, lack of data for industrial minerals, and lack of confidence in subjective quantitative assessment methodologies. This study developed two potential methodologies, a statistical approach and a process-based approach, to quantitatively assess borate in playas and minimize these concerns. Selection of a Quaternary surficial deposit type where many of the criteria could be measured from existing geologic maps or inferred from current conditions and processes was one of the steps taken to minimize concerns about the techniques. Qualitative and quantitative data were collected on 22 playa-lacustrine basins containing one or more borate deposits and 21 basins in the same areas that have no known borate deposit(s). Statistical techniques were successfully employed using dominantly lithologic data to differentiate mineralized vs. non-mineralized basins (stepwise discriminant analysis) and to estimate boron endowment of mineralized basins (stepwise regression analysis). Boron estimation for the 22 mineralized basins was also attempted from a simple process-based model. Precipitation/evaporation estimates and the amount of Quaternary volcanics in each basin were used as a basis for estimating total boron concentrations in the playas. Selection of a Quaternary deposit type was made in part to minimize unrecognized processes and maximize the ability to measure variables. However, results using this methodology were highly variable and reflect the uncertainties of geologic process, climate, and time estimations, especially factors related to loss of boron from the mineralizing system. The methodologies developed here offer two different perspectives for quantitatively estimating B endowment that can be used to establish whether a basin is likely to contain boron mineralization, calculate a minimal endowment level and the probability of that level of endowment, and determine a maximum level of endowment given no boron losses to the mineralizing system. These results can be used in isolation or to constrain a subjective assessment.