Search USGS⌕ Search

Geology topics

Donald A. Singer

Publications and source records attributed to Donald A. Singer.

At least 19 recordsLinked to original sources
USGS2021

Mineral deposit discovery order and three-part quantitative assessments

Larger oil pools tending to be discovered earlier in an exploration play suggests the same pattern might exist for mineral deposits and could be used in predicting sizes of undiscovered deposits in mineral assessments. The volume of individual petroleum pools is highly correlated with surface projection area of pools in basins. The gradual additions to individual oil pool reserves over time adds to the appearance of larger pools being discovered earlier. Comparisons of surface projected areas of mineral deposits to their tonnages showed significant positive relationships in all 10 deposit types analyzed, suggesting that larger deposits should be discovered earlier than small deposits. Analysis of deposits consistent with three-part mineral assessments identified 9 combinations of mineral deposit types in large regions each containing multiple geological permissive tracts showing negative and 1 positive relationships of deposit size with discovery date significant at the 1% level. Twenty other tests of regions containing multiple permissive settings had either negative or positive relationships, none significantly different from those that might occur by chance. The large regions are mostly based on political boundaries. These results suggest mineral deposit discovery order is not the same as observed in oil pool exploration. The widely employed three-part quantitative mineral resource assessments are an obvious choice to benefit from patterns of declining deposit sizes with order of discovery. The 30 tests of relationships of discovery dates to deposit sizes demonstrated here were performed with deposits consistent with those in three-part assessments, but the large areas are not consistent with permissive tracts used in these assessments because they also contain substantial non-permissive geology. In 100 permissive tracts assessed with three-part assessments of multiple deposit types located throughout the world, the median number of known well-explored deposits is 1 and 90 percent of tracts report less than 9 deposits. The number of well-explored deposits in three-part assessed tracts tends to be quite small, limiting any ability to recognize a discovery order versus size relationship. In a three-part assessment of undiscovered porphyry copper deposits of South America, only 7 of 26 delineated tracts contained more than 2 known deposits and only 1 had a significant negative relationship between tonnage of known deposits and year of discovery (p = 0.04). Most predicted undiscovered deposits in this tract were expected to be under extensive unexplored post-mineralization cover, meaning the general grade and tonnage model should be applied because the discovery order process starts over. Projection of deposit sizes based on discovery order would provide a biased estimate of the undiscovered deposit sizes in this case. Thus, although a discovery order versus size relationship could exist in three-part mineral assessments, only rarely might the pattern be useful to predict sizes of undiscovered deposits.

Ore Geology Reviews↗
USGS2012

Podiform chromite deposits--database and grade and tonnage models

Chromite ((Mg, Fe ++ )(Cr, Al, Fe +++ ) 2 O 4 ) is the only source for the metallic element chromium, which is used in the metallurgical, chemical, and refractory industries. Podiform chromite deposits are small magmatic chromite bodies formed in the ultramafic section of an ophiolite complex in the oceanic crust. These deposits have been found in midoceanic ridge, off-ridge, and suprasubduction tectonic settings. Most podiform chromite deposits are found in dunite or peridotite near the contact of the cumulate and tectonite zones in ophiolites. We have identified 1,124 individual podiform chromite deposits, based on a 100-meter spatial rule, and have compiled them in a database. Of these, 619 deposits have been used to create three new grade and tonnage models for podiform chromite deposits. The major podiform chromite model has a median tonnage of 11,000 metric tons and a mean grade of 45 percent Cr 2 O 3 . The minor podiform chromite model has a median tonnage of 100 metric tons and a mean grade of 43 percent Cr 2 O 3 . The banded podiform chromite model has a median tonnage of 650 metric tons and a mean grade of 42 percent Cr 2 O 3 . Observed frequency distributions are also given for grades of rhodium, iridium, ruthenium, palladium, and platinum. In resource assessment applications, both major and minor podiform chromite models may be used for any ophiolite complex regardless of its tectonic setting or ophiolite zone. Expected sizes of undiscovered podiform chromite deposits, with respect to degree of deformation or ore-forming process, may determine which model is appropriate. The banded podiform chromite model may be applicable for ophiolites in both suprasubduction and midoceanic ridge settings.

Scientific Investigations Report↗
USGS2011

Probabilistic estimates of number of undiscovered deposits and their total tonnages in permissive tracts using deposit densities

Empirical evidence indicates that processes affecting number and quantity of resources in geologic settings are very general across deposit types. Sizes of permissive tracts that geologically could contain the deposits are excellent predictors of numbers of deposits. In addition, total ore tonnage of mineral deposits of a particular type in a tract is proportional to the type’s median tonnage in a tract. Regressions using size of permissive tracts and median tonnage allow estimation of number of deposits and of total tonnage of mineralization. These powerful estimators, based on 10 different deposit types from 109 permissive worldwide control tracts, generalize across deposit types. Estimates of number of deposits and of total tonnage of mineral deposits are made by regressing permissive area, and mean (in logs) tons in deposits of the type, against number of deposits and total tonnage of deposits in the tract for the 50th percentile estimates. The regression equations ( R 2 = 0.91 and 0.95) can be used for all deposit types just by inserting logarithmic values of permissive area in square kilometers, and mean tons in deposits in millions of metric tons. The regression equations provide estimates at the 50th percentile, and other equations are provided for 90% confidence limits for lower estimates and 10% confidence limits for upper estimates of number of deposits and total tonnage. Equations for these percentile estimates along with expected value estimates are presented here along with comparisons with independent expert estimates. Also provided are the equations for correcting for the known well-explored deposits in a tract. These deposit-density models require internally consistent grade and tonnage models and delineations for arriving at unbiased estimates.

Natural Resources Research↗
USGS2011

Effects of intrusions on grades and contents of gold and other metals in volcanogenic massive sulfide deposits

The reason some VMS deposits contain more gold or other metals than others might be due to the influence of intrusions. A new approach examining this possibility is based on examining the information about many VMS deposits to test statistically if those with associated intrusions have significantly different grades or amounts of metals. A set of 632 VMS deposits with reported grades, tonnages, and information about the observed presence or absence of subvolcanic or plutonic intrusive bodies emplaced at or after VMS mineralization is statistically analyzed. Deposits with syn-mineralization or post-mineralization intrusions nearby have higher tonnages than deposits without reported intrusions, but the differences are not statistically significant. When both kinds of intrusions are reported, VMS deposit sizes are significantly higher than in the deposits without any intrusions. Gold, silver, zinc, lead, and copper average grades are not significantly different in the VMS deposits with nearby intrusions compared to deposits without regardless of relative age of intrusive. Only zinc and copper contents are significantly higher in VMS deposits with both kinds of intrusive reported. These differences in overall metal content are due to significantly larger deposit sizes of VMS deposits where both intrusive kinds are observed and reported, rather than any difference in metal grades.

Ore Geology Reviews↗
USGS2011

A lognormal distribution of metal resources

For national or global resource estimation of frequencies of metals, a lognormal distribution has commonly been recommended but not adequately tested. Tests of frequencies of Cu, Zn, Pb, Ag, and Au contents of 1 984 well-explored mineral deposits display a poor fit to the lognormal distribution. When the same metals plus Mo, Co, Nb2O3, and REE2O3 are grouped into 19 geologically defined deposit types, only eight of the 73 tests fail to be fit by lognormal distribution, and most of those failures are in two deposit types suggesting a problem with those types. Estimates of the mean and standard deviation of each of the metals in each of the deposit types are provided for modeling.

Diqiu Kexue - Zhongguo Dizhi Daxue Xuebao/Earth Sc↗
USGS2009

Volcanogenic Massive Sulfide Deposits of the World - Database and Grade and Tonnage Models

Grade and tonnage models are useful in quantitative mineral-resource assessments. The models and database presented in this report are an update of earlier publications about volcanogenic massive sulfide (VMS) deposits. These VMS deposits include what were formerly classified as kuroko, Cyprus, and Besshi deposits. The update was necessary because of new information about some deposits, changes in information in some deposits, such as grades, tonnages, or ages, revised locations of some deposits, and reclassification of subtypes. In this report we have added new VMS deposits and removed a few incorrectly classified deposits. This global compilation of VMS deposits contains 1,090 deposits; however, it was not our intent to include every known deposit in the world. The data was recently used for mineral-deposit density models (Mosier and others, 2007; Singer, 2008). In this paper, 867 deposits were used to construct revised grade and tonnage models. Our new models are based on a reclassification of deposits based on host lithologies: Felsic, Bimodal-Mafic, and Mafic volcanogenic massive sulfide deposits. Mineral-deposit models are important in exploration planning and quantitative resource assessments for two reasons: (1) grades and tonnages among deposit types vary significantly, and (2) deposits of different types occur in distinct geologic settings that can be identified from geologic maps. Mineral-deposit models combine the diverse geoscience information on geology, mineral occurrences, geophysics, and geochemistry used in resource assessments and mineral exploration. Globally based deposit models allow recognition of important features and demonstrate how common different features are. Well-designed deposit models allow geologists to deduce possible mineral-deposit types in a given geologic environment and economists to determine the possible economic viability of these resources. Thus, mineral-deposit models play a central role in presenting geoscience information in a useful form to policy makers. The foundation of mineral-deposit models is information about known deposits. The purpose of this publication is to present the latest geologic information and newly developed grade and tonnage models for VMS deposits in digital form. This publication contains computer files with information on VMS deposits from around the world. It also presents new grade and tonnage models for three subtypes of VMS deposits and a text file allowing locations of all deposits to be plotted in geographic information system (GIS) programs. The data are presented in FileMaker Pro and text files to make the information available to a wider audience. The value of this information and any derived analyses depends critically on the consistent manner of data gathering. For this reason, we first discuss the rules used in this compilation. Next, we provide new grade and tonnage models and analysis of the information in the file. Finally, the fields of the data file are explained. Appendix A gives the summary statistics for the new grade-tonnage models and Appendix B displays the country codes used in the database.

Open-File Report↗
USGS2009

Carbonatites of the world, explored deposits of Nb and REE— Database and grade and tonnage models

This report is based on published tonnage and grade data on 58 Nb- and rare-earth-element (REE)-bearing carbonatite deposits that are mostly well explored and are partially mined or contain resources of these elements. The deposits represent only a part of the known 527 carbonatites around the world, but they are characterized by reliable quantitative data on ore tonnages and grades of niobium and REE. Grade and tonnage models are an important component of mineral resource assessments. Carbonatites present one of the main natural sources of niobium and rare-earth elements, the economic importance of which grows consistently. A purpose of this report is to update earlier publications. New information about known deposits, as well as data on new deposits published during the last decade, are incorporated in the present paper. The compiled database (appendix 1; linked to right) contains 60 explored Nb- and REE-bearing carbonatite deposits - resources of 55 of these deposits are taken from publications. In the present updated grade-tonnage model we have added 24 deposits comparing with the previous model of Singer (1998). Resources of most deposits are residuum ores in the upper part of carbonatite bodies. Mineral-deposit models are important in exploration planning and quantitative resource assessments for two reasons: (1) grades and tonnages among deposit types vary significantly, and (2) deposits of different types are present in distinct geologic settings that can be identified from geologic maps. Mineral-deposit models combine the diverse geoscience information on geology, mineral occurrences, geophysics, and geochemistry used in resource assessments and mineral exploration. Globally based deposit models allow recognition of important features and demonstrate how common different features are. Well-designed deposit models allow geologists to deduce possible mineral-deposit types in a given geologic environment, and the grade and tonnage models allow economists to estimate the possible economic viability of these resources. Thus, mineral-deposit models play a central role in presenting geoscience information in a useful form to policy makers. The foundation of mineral-deposit models is information about known deposits. This publication presents the latest geologic information and newly developed grade and tonnage models for Nb- and REE-carbonatite deposits in digital form. The publication contains computer files with information on deposits from around the world. It also contains a text file allowing locations of all deposits to be plotted in geographic information system (GIS) programs. The data are presented in FileMaker Pro as well as in .xls and text files to make the information available to a broadly based audience. The value of this information and any derived analyses depends critically on the consistent manner of data gathering. For this reason, we first discuss the rules used in this compilation. Next, the fields of the database are explained. Finally, we provide new grade and tonnage models and analysis of the information in the file.

Open-File Report↗
USGS2009

Sediment-hosted zinc-lead deposits of the world— Database and grade and tonnage models

This report provides information on sediment-hosted zinc-lead mineral deposits based on the geologic settings that are observed on regional geologic maps. The foundation of mineral-deposit models is information about known deposits. The purpose of this publication is to make this kind of information available in digital form for sediment-hosted zinc-lead deposits. Mineral-deposit models are important in exploration planning and quantitative resource assessments: Grades and tonnages among deposit types are significantly different, and many types occur in different geologic settings that can be identified from geologic maps. Mineral-deposit models are the keystone in combining the diverse geoscience information on geology, mineral occurrences, geophysics, and geochemistry used in resource assessments and mineral exploration. Too few thoroughly explored mineral deposits are available in most local areas for reliable identification of the important geoscience variables, or for robust estimation of undiscovered deposits - thus, we need mineral-deposit models. Globally based deposit models allow recognition of important features because the global models demonstrate how common different features are. Well-designed and -constructed deposit models allow geologists to know from observed geologic environments the possible mineral-deposit types that might exist, and allow economists to determine the possible economic viability of these resources in the region. Thus, mineral-deposit models play the central role in transforming geoscience information to a form useful to policy makers. This publication contains a computer file of information on sediment-hosted zinc-lead deposits from around the world. It also presents new grade and tonnage models for nine types of these deposits and a file allowing locations of all deposits to be plotted in Google Earth. The data are presented in FileMaker Pro, Excel and text files to make the information available to as many as possible. The value of this information and any derived analyses depends critically on the consistent manner of data gathering. For this reason, we first discuss the rules applied in this compilation. Next, the fields of the data file are considered. Finally, we provide new grade and tonnage models that are, for the most part, based on a classification of deposits using observable geologic units from regional-scaled maps.

Open-File Report↗
USGS2008

Porphyry copper deposits of the world: Database and grade and tonnage models, 2008

This report is an update of earlier publications about porphyry copper deposits (Singer, Berger, and Moring, 2002; Singer, D.A., Berger, V.I., and Moring, B.C., 2005). The update was necessary because of new information about substantial increases in resources in some deposits and because we revised locations of some deposits so that they are consistent with images in GoogleEarth. In this report we have added new porphyry copper deposits and removed a few incorrectly classed deposits. In addition, some errors have been corrected and a number of deposits have had some information, such as grades, tonnages, locations, or ages revised. Colleagues have helped identify places where improvements were needed. Mineral deposit models are important in exploration planning and quantitative resource assessments for a number of reasons including: (1) grades and tonnages among deposit types are significantly different, and (2) many types occur in different geologic settings that can be identified from geologic maps. Mineral deposit models are the keystone in combining the diverse geoscience information on geology, mineral occurrences, geophysics, and geochemistry used in resource assessments and mineral exploration. Too few thoroughly explored mineral deposits are available in most local areas for reliable identification of the important geoscience variables or for robust estimation of undiscovered deposits?thus we need mineral-deposit models. Globally based deposit models allow recognition of important features because the global models demonstrate how common different features are. Well-designed and -constructed deposit models allow geologists to know from observed geologic environments the possible mineral deposit types that might exist, and allow economists to determine the possible economic viability of these resources in the region. Thus, mineral deposit models play the central role in transforming geoscience information to a form useful to policy makers. The foundation of mineral deposit models is information about known deposits. The purpose of this publication is to make this kind of information available in digital form for porphyry copper deposits. The consistently defined deposits in this file provide the foundation for grade and tonnage models included here and for mineral deposit density models (Singer and others, 2005: Singer, 2008).

Open-File Report↗
USGS2008

Quantitative mineral resource assessment of copper, molybdenum, gold, and silver in undiscovered porphyry copper deposits in the Andes Mountains of South America

Quantitative information on the general locations and amounts of undiscovered porphyry copper resources of the world is important to exploration managers, land-use and environmental planners, economists, and policy makers. This publication contains the results of probabilistic estimates of the amounts of copper (Cu), molybdenum (Mo), gold (Au), and silver (Ag) in undiscovered porphyry copper deposits in the Andes Mountains of South America. The methodology used to make these estimates is called the 'Three-Part Form'. It was developed to explicitly express estimates of undiscovered resources and associated uncertainty in a form that allows economic analysis and is useful to decisionmakers. The three-part form of assessment includes: (1) delineation of tracts of land where the geology is permissive for porphyry copper deposits to form; (2) selection of grade and tonnage models appropriate for estimating grades and tonnages of the undiscovered porphyry copper deposits in each tract; and (3) estimation of the number of undiscovered porphyry copper deposits in each tract consistent with the grade and tonnage model. A Monte Carlo simulation computer program (EMINERS) was used to combine the probability distributions of the estimated number of undiscovered deposits, the grades, and the tonnages of the selected model to obtain the probability distributions for undiscovered metals in each tract. These distributions of grades and tonnages then can be used to conduct economic evaluations of undiscovered resources in a format usable by decisionmakers. Economic evaluations are not part of this report. The results of this assessment are presented in two principal parts. The first part identifies 26 regional tracts of land where the geology is permissive for the occurrence of undiscovered porphyry copper deposits of Phanerozoic age to a depth of 1 km below the Earth's surface. These tracts are believed to contain most of South America's undiscovered resources of copper. The second part presents probabilistic estimates of the amounts of copper, molybdenum, gold, and silver in undiscovered porphyry copper deposits in each tract. The study also provides tables showing the location, tract number, and age (if available) of discovered deposits and prospects. For each of the 26 permissive tracts delineated in this study, summary information is provided on: (1) the rationale for delineating the tract; (2) the rationale for choosing the mineral deposit model used to assess the tract; (3) discovered deposits and prospects; (4) exploration history; and (5) the distribution of undiscovered deposits in the tract. The scale used to evaluate geologic information and draw tracts is 1:1,000,000.

Open-File Report↗
USGS2007

Mineral deposit densities for estimating mineral resources

Estimates of numbers of mineral deposits are fundamental to assessing undiscovered mineral resources. Just as frequencies of grades and tonnages of well-explored deposits can be used to represent the grades and tonnages of undiscovered deposits, the density of deposits (deposits/area) in well-explored control areas can serve to represent the number of deposits. Empirical evidence presented here indicates that the processes affecting the number and quantity of resources in geological settings are very general across many types of mineral deposits. For podiform chromite, porphyry copper, and volcanogenic massive sulfide deposit types, the size of tract that geologically could contain the deposits is an excellent predictor of the total number of deposits. The number of mineral deposits is also proportional to the type’s size. The total amount of mineralized rock is also proportional to size of the permissive area and the median deposit type’s size. Regressions using these variables provide a means to estimate the density of deposits and the total amount of mineralization. These powerful estimators are based on analysis of ten different types of mineral deposits (Climax Mo, Cuban Mn, Cyprus massive sulfide, Franciscan Mn, kuroko massive sulfide, low-sulfide quartz-Au vein, placer Au, podiform Cr, porphyry Cu, and W vein) from 108 permissive control tracts around the world therefore generalizing across deposit types. Despite the diverse and complex geological settings of deposit types studied here, the relationships observed indicate universal controls on the accumulation and preservation of mineral resources that operate across all scales. The strength of the relationships ( R 2 =0.91 for density and 0.95 for mineralized rock) argues for their broad use. Deposit densities can now be used to provide a guideline for expert judgment or used directly for estimating the number of most kinds of mineral deposits.

Mathematical Geosciences↗
USGS2007

Descriptive and Grade-Tonnage Models and Database for Iron Oxide Cu-Au Deposits

Iron oxide Cu-Au deposits are veins and breccia-hosted bodies of hematite and/or magnetite with disseminated Cu + Au ? Ag ? Pd ? Pt ? Ni ? U ? LREE minerals formed in sedimentary or volcano-sedimentary basins intruded by igneous rocks. Deposits are associated with broad redox boundaries and feature sodic alteration of source rocks and potassic alteration of host rocks.

Open-File Report↗
USGS2007

Short Course Introduction to Quantitative Mineral Resource Assessments

This is an abbreviated text supplementing the content of three sets of slides used in a short course that has been presented by the author at several workshops. The slides should be viewed in the order of (1) Introduction and models, (2) Delineation and estimation, and (3) Combining estimates and summary. References cited in the slides are listed at the end of this text. The purpose of the three-part form of mineral resource assessments discussed in the accompanying slides is to make unbiased quantitative assessments in a format needed in decision-support systems so that consequences of alternative courses of action can be examined. The three-part form of mineral resource assessments was developed to assist policy makers evaluate the consequences of alternative courses of action with respect to land use and mineral-resource development. The audience for three-part assessments is a governmental or industrial policy maker, a manager of exploration, a planner of regional development, or similar decision-maker. Some of the tools and models presented here will be useful for selection of exploration sites, but that is a side benefit, not the goal. To provide unbiased information, we recommend the three-part form of mineral resource assessments where general locations of undiscovered deposits are delineated from a deposit type's geologic setting, frequency distributions of tonnages and grades of well-explored deposits serve as models of grades and tonnages of undiscovered deposits, and number of undiscovered deposits are estimated probabilistically by type. The internally consistent descriptive, grade and tonnage, deposit density, and economic models used in the design of the three-part form of assessments reduce the chances of biased estimates of the undiscovered resources. What and why quantitative resource assessments: The kind of assessment recommended here is founded in decision analysis in order to provide a framework for making decisions concerning mineral resources under conditions of uncertainty. What this means is that we start with the question of what kinds of questions is the decision maker trying to resolve and what forms of information would aid in resolving these questions. Some applications of mineral resource assessments: To plan and guide exploration programs, to assist in land use planning, to plan the location of infrastructure, to estimate mineral endowment, and to identify deposits that present special environmental challenges. Why not just rank prospects / areas? Need for financial analysis, need for comparison with other land uses, need for comparison with distant tracts of land, need to know how uncertain the estimates are, need for consideration of economic and environmental consequences of possible development. Our goal is to provide unbiased information useful to decision-makers.

Open-File Report↗
USGS2007

Volcanogenic Massive Sulfide Deposit Density

A mineral-deposit density model for volcanogenic massive sulfide deposits was constructed from 38 well-explored control areas from around the world. Control areas contain at least one exposed volcanogenic massive sulfide deposit. The control areas used in this study contain 150 kuroko, 14 Urals, and 25 Cyprus massive sulfide subtypes of volcanogenic massive sulfide deposits. For each control area, extent of permissive rock, number of exposed volcanogenic massive sulfide deposits, map scale, deposit age, and deposit density were determined. The frequency distribution of deposit densities in these 38 control areas provides probabilistic estimates of the number of deposits for tracts that are permissive for volcanogenic massive sulfide deposits-90 percent of the control areas have densities of 100 or more deposits per 100,000 square kilometers, 50 percent of the control areas have densities of 700 or more deposits per 100,000 square kilometers, and 10 percent of the control areas have densities of 3,700 or more deposits per 100,000 square kilometers. Both map scale and the size of the control area are shown to be predictors of deposit density. Probabilistic estimates of the number of volcanogenic massive sulfide deposits can be made by conditioning the estimates on sizes of permissive area. The model constructed for this study provides a powerful tool for estimating the number of undiscovered volcanogenic massive sulfide deposits when conducting resource assessments. The value of these deposit densities is due to the consistency of these models with the grade and tonnage and the descriptive models. Mineral-deposit density models combined with grade and tonnage models allow reasonable estimates of the number, size, and grades of volcanogenic massive sulfide deposits to be made.

Scientific Investigations Report↗
USGS2005

Porphyry copper deposits of the world: database, map, and grade and tonnage models

Mineral deposit models are important in exploration planning and quantitative resource assessments for two reasons: (1) grades and tonnages among deposit types are significantly different, and (2) many types occur in different geologic settings that can be identified from geologic maps. Mineral deposit models are the keystone in combining the diverse geoscience information on geology, mineral occurrences, geophysics, and geochemistry used in resource assessments and mineral exploration. Too few thoroughly explored mineral deposits are available in most local areas for reliable identification of the important geoscience variables or for robust estimation of undiscovered deposits-thus we need mineral-deposit models. Globally based deposit models allow recognition of important features because the global models demonstrate how common different features are. Well-designed and -constructed deposit models allow geologists to know from observed geologic environments the possible mineral deposit types that might exist, and allow economists to determine the possible economic viability of these resources in the region. Thus, mineral deposit models play the central role in transforming geoscience information to a form useful to policy makers. The foundation of mineral deposit models is information about known deposits-the purpose of this publication is to make this kind of information available in digital form for porphyry copper deposits. This report is an update of an earlier publication about porphyry copper deposits. In this report we have added 84 new porphyry copper deposits and removed 12 deposits. In addition, some errors have been corrected and a number of deposits have had some information, such as grades, tonnages, locations, or ages revised. This publication contains a computer file of information on porphyry copper deposits from around the world. It also presents new grade and tonnage models for porphyry copper deposits and for three subtypes of porphyry copper deposits and a map showing the location of all deposits. The value of this information and any derived analyses depends critically on the consistent manner of data gathering. For this reason, we first discuss the rules used in this compilation. Next, the fields of the data file are considered. Finally, we provide new grade and tonnage models.

Open-File Report↗
USGS2003

Use of a probabilistic neural network to reduce costs of selecting construction rock

Rocks used as construction aggregate in temperate climates deteriorate to differing degrees because of repeated freezing and thawing. The magnitude of the deterioration depends on the rock's properties. Aggregate, including crushed carbonate rock, is required to have minimum geotechnical qualities before it can be used in asphalt and concrete. In order to reduce chances of premature and expensive repairs, extensive freeze-thaw tests are conducted on potential construction rocks. These tests typically involve 300 freeze-thaw cycles and can take four to five months to complete. Less time consuming tests that (1) predict durability as well as the extended freeze-thaw test or that (2) reduce the number of rocks subject to the extended test, could save considerable amounts of money. Here we use a probabilistic neural network to try and predict durability as determined by the freeze-thaw test using four rock properties measured on 843 limestone samples from the Kansas Department of Transportation. Modified freeze-thaw tests and less time consuming specific gravity (dry), specific gravity (saturated), and modified absorption tests were conducted on each sample. Durability factors of 95 or more as determined from the extensive freeze-thaw tests are viewed as acceptable—rocks with values below 95 are rejected. If only the modified freeze-thaw test is used to predict which rocks are acceptable, about 45% are misclassified. When 421 randomly selected samples and all four standardized and scaled variables were used to train aprobabilistic neural network, the rate of misclassification of 422 independent validation samples dropped to 28%. The network was trained so that each class (group) and each variable had its own coefficient (sigma). In an attempt to reduce errors further, an additional class was added to the training data to predict durability values greater than 84 and less than 98, resulting in only 11% of the samples misclassified. About 43% of the test data was classed by the neural net into the middle group—these rocks should be subject to full freeze-thaw tests. Thus, use of the probabilistic neural network would meanthat the extended test would only need be applied to 43% of the samples, and 11% of the rocks classed as acceptable would fail early.

Natural Resources Research↗