Search Documents

By P. Nelson, K. L. Gunsallus, A. R. Ingraffea, J. F. Beech

INTRODUCTION Fracture toughness, KIC, is an intrinsic material property and is a measure of the energy required to create new surface area in a material. Fracture toughness measurements can be made for a wide range of rock types (Ouchterlony, 1980). As a measure of energy of comminution, fracture toughness might be used for more sensitive predictions of Tunnel Boring Machine (TRM) performance than are possible with other index measures in current use. This paper describes part of an ongoing, comprehensive, analytical/experimental program aimed at evaluation of KIC in such a role. The program includes efforts to: 1. Provide a firm analytical basis for a test specimen geometry: the core-derived short-rod. 2. Develop simple, inexpensive testing equipment and techniques. 3. Evaluate the accuracy of the above by testing rock and other materials and comparing the measured fracture toughness with those obtained by other methods. 4. Measure fracture toughness of a substantial number of different rocks from three ongoing TBM projects. 5. Compare fracture toughness results with other strength measures. 6. Compare all test results with observed TBM performance in the three projects. Presented here are summaries of all but the last of these efforts. Comparative evaluation of KIC as a more senstive rock property for TBM performance prediction will be the subject of the oral presentation and a subsequent paper.

Jan 1, 1982

By B. Petkof

Olivine has been an accepted industrial mineral for the last 50 years. Its initial acceptance was based on the recognition of its refractory properties by European research workers in the middle to late 1920's. Initial U.S. olivine production began in the early 1930's in North Carolina and continues there today. Olivine also is produced in the State of Washington and in other countries.

Jan 1, 1982

By John J. Mulligan

Coal is found in every major geographic region throughout Alaska and in two offshore areas. Since about 1852 over 100 small mines have opened and closed. Last years production was almost entirely from one mine in central Alaska. Clearly the utilization of Alaska's abundant coal resources presents political, economic and technical problems that are far from solution. Location and Physical Features Figure one illustrates the geographical and geological similarity between Alaska and western United States. The Arctic region of Alaska generally resembles Wyoming and the Dakotas. The Brooks Range is an extension of the Rocky Mountains. The Central region of Alaska between the Brooks Range and the Alaska Range bears a general resemblance to the basin and range country of western United States. The Southcentral and Southeastern regions has a similar resemblance to the Sierra Nevadas and the west coast of the United States.

Jan 1, 1982

By R. W. M. Lai

The study of flotation kinetics was first reported by Garcia-Zuniga1 in 1935. Since then, literature on the subject has grown with the time. 2-19 Although there is an abundance of flotation kinet¬ics models available in the literature to describe flo¬tation systems, they are mostly of academic interest and are seldom applicable for use in the industrial environment. Several reasons contribute to the wide gap in the industrial -application of the academic models. These include: 1) the lack of generality in the model, e.g., the model is limited to certain size ranges; 14 2) the academic models involve complicated mathematics; 5,6,10,11 3) confusion about the order of the flo¬tation rate, e.g., several authors proposed a first order rate model, 1,15,16 whereas the others propagated a second order rate model: 2,17,18,194) the use of an unrealistic rigid value, such as the use of a maximum recovery of 100%, 5,8,19 which is unrealistic in industrial operations; and 5) the use of graphs that confuse the reader rather than interpret information. 2,3,10 This paper intends to introduce a simple and easy to understand empirical flotation kinetics model along with a way of presenting the model in an informative graphical form, which should stimulate the interest of other practical engineers to become concerned with practical application within the mineral industry.

Jan 1, 1982

By John W. Stevenson

At the No. 4 Mine of Jim Walter Resources in Tuscaloosa County in the North Central part of the State of Alabama, the Mary Lee and Blue Creek coal beds lay 2000 feet below the surface. The quantity of methane gas stored within the coal bed reservoir at this depth and released during development mining necessitates that each coal producing section be ventilated using large air quantities approaching 100,000 cfm. In order to meet coal contractural commitments in the early stages of mine development by installing the necessary number of continuous mining machine units and to develop the mine to projected air shaft sites, high pressure, large volume mine ventilating fans were required. This presentation discusses the installation start-up and changeover procedures when two size GAF 31.5-28-1 TLT-Babcock 3500 Horsepower fans with adjustable pitch in flight blades were put on line while phasing out two Jeffrey Mining Machine Company 8HUA 117 Fans. A graph will be presented to depict the most economic diameter shaft size under 2000 feet of cover with different air quantity requirements.

Jan 1, 1982

By S. K. Kawatra

The use of inferential particle sizing for a closed wet grinding circuit control has been practiced for several years. This paper presents a review of some inferential sizing equations developed for pilot and industrial plants using regression analysis. An analytical development is then presented showing how the general inferential sizing model arises from the nature of the classifier model. The relation between the general sizing model and the individual equations is then discussed.

Jan 1, 1982

By Brian E. McGee

The Federal Coal Management Program represents the culmination of several years of effort by the Department of the Interior (001) to reformulate the manner in which coal owned by the United States is leased and developed. These efforts began in 1970 with a study by the Bureau of Land Management (BLM), which indicated that unless drastic changes were made in. the federal coal leasing procedures, federal coal lands would be leased almost for the asking while production of coal from those lands had been decreasing. Consequently, informal action was taken in 1971, which resulted in no further leasing being authorized until February, 1973. At that time, a formal moratorium on most coal leasing was established, with exceptions granted only for leases needed to satisfy short-term needs such as maintaining existing mines or supplying reserves for new contracts in the immediate future. This leasing policy embodied a long term objective of developing an entirely new planning system for federal coal leasing, as well as to schedule the preparation of a programmatic impact statement for this new system. As a result of the 1973 moratorium and the ensuing application of short-term criteria, only 10 federal coal leases (covering approximately 30,177 acres) were issued from February, 1973, through January l, 1976.

Jan 1, 1982

By Jacque Turgeon, Gontran Foy

"The Mineral Research Centre of the Quebec Department of Energy and Resources has worked extensively on the development of a dry process to extract pure mica flakes from a phlogopite ore. The liberation of mica from the gangue is effected basically by impact crushing, and the ground ore is screened into several fractions, each having a fairly narrow range of grain size. Each sized fraction is then fed to an air separator where heavy grains are separated from light flaky grains. Heavy fractions are returned to hammer mills where gangue minerals are ground finer and blocky mica particles are exfoliated. A second stage of screening is performed and each fraction is sent to an air separator. This sequence is repeated until mica products are virtually free of gangue. These different fractions are then treated in a cage mill in which mica flakes undergo a more orderly exfoliation along cleavage planes. The ground mica is again submitted to a final screening stage, after which each fraction constitutes a final commercial product.This process permits the recovery of about 70% of the mica contained in the ore. The over-all process will be described along with technical details of each unit operation. Recent work on a wet grinding method to produce very fine mica will also be discussed.IntroductionRecords of production of mica in Quebec date back to 1886. Production has been more or less continuous since that time, with peak activity during the war years. Production was interrupted in 1966, but started again in 1972 when Marietta Resources and Laviolette Mining and Metallurgical Corporation started their joint operation in Canton Suzor, northwestern Quebec. Mica deposits of more or less importance exist in different parts of Quebec, of which three have attracted attention in recent years: the Canton Suzor deposit, the Canton Lamy deposit and the Canton Chasseur deposit. The Suzor orebody was first identified by Carl Faesler in 1936, who described the rock as ""brown biotite in flakes half an inch in diameter, with interstitial augite, apatite and feldspar""."

Jan 1, 1982

By D. S. Harries, R. I. L. Guthrie

"The use and development of scientific disciplines involved in heat transfer technology, related in general to steel masses and in particular to batch annealing, is demonstrated.The measurable benefits of cooperation are tabled; on industry's side, this includes a 10 to 20 per cent decrease in annealing time, and a corresponding higher productivity potential. The returned benefits to the academic engineer include the development and refinement of new computer algorithms related to heat transfer mechanisms.IntroductionThe object of this discussion is not really to describe an improvement in the batch annealing operation, but to use the project as an example of joint work between industry and the academic field. Both parties have derived a number of substantial benefits. These benefits will be listed, of course, but the batch anneal project will be used just as a typical example of cooperation. Why it happened will be explained, what actually happened will be described and finally some thoughts about areas in the field of heat transfer, where cooperation will continue in the future, will be highlighted. The paper will close with a message.Industry OverviewAt the outset, perhaps it should be understood that, in general, the Research Department at Dofasco believes that the universities and their staff are not destined for the problem-solving role in industry. Its opinion is that the university staff is best suited to the gathering and dissemination of knowledge. There are other more professional problem solvers for industry, including the steel industries' own individual investigation groups. Dofasco, therefore, looks to its academics for consultation. It benefits by bouncing ideas off them, and through rubbing shoulders with them in such a way that their knowledge can be tapped, and thus establish more effective scientific routes. In addition, the younger and brighter engineer s can broaden their scope through the intellectual contact with experts. The university professors can, it is believed, benefit through closer contacts with result-orientated research. They can also develop an appreciation of the fiscal control necessary for industrial investigations, without government subsidies, grants, etc., and with massive overheads."

Jan 1, 1982

By K. O. Groves

Although chlorine oxidation has been found to be effective in destroying the CN ion, an excess of chlorine is needed to drive the reaction to completion. This often results in the formation of residual chloramines, which are toxic to aquatic life and often regulated by various states. The interhalogen, bromine chloride (BrCl), has been found to be just as effective but without the formation of persistent, residual chloramines. The hydrolysis and subsequent disproportionation of the highly-polar BrCl molecule essentially yields a solution of hypobromous acid (HOBr) and nonoxidizing hydrochloric acid (HC1). Studies have been carried out to directly compare the mechanism of CN destruction with chlorine and bromine chloride.

Jan 1, 1982

Shaft sinking through an amphibolite-granite rock mass experienced rock bursts on the shaft walls and benches. However, rock burst phenomena was exhibited only in granite, which is e1asto-brittle with ability to store strain energy caused by orogenic and man made stresses. To improve ground conditions, blasting of destressing holes at shaft corners was carried out. By this technique rock burst during shaft sinking was minimized, but not eliminated.

Jan 1, 1982

By Frank L. Peterson

This paper deals with problems of groundwater development in an insular circum-Pacific environment, and is not oriented as such to mining and mineral development. Nonetheless, because a reliable water supply is necessary for most mining and related support activities, and since water supplies are limited in many insular parts of the Pacific Basin, groundwater development is a subject which must be frequently addressed. Because many in the audience may be already familiar with groundwater development for continental environments, I have chosen to deal with groundwater occurrence and development problems which for the most part are unique to insular environments. In such environments, the local geology plays a critical role in determining groundwater occurrence and its suitability for development. The three examples discussed in this paper-Hawaii, Guam, and the Marshall Islands-were selected because they illustrate groundwater occurrence in a wide range of geologic conditions. For example, the Hawaiian aquifers consist primarily of basaltic lavas, while Guam is made up of andesitic lavas and coral reef limestones, and the Marshall Islands are coral atoll deposits. Despite differences in aquifer size and rock types, the principal mode of groundwater occurrence at the three sites is similar; thus, development strategies are also similar. Figure 1 shows the location of the three sites.

Jan 1, 1982

By James Pirie

"The major Archean gold deposits in the Red Lake greenstone belt, which have produced over 374 million grams of gold since 1930, occur near the top of the lower komatiitic to tholeiitic mafic to ultramafic metavolcanic sequence. The upper calc-alkalic mafic to felsic rocks form three discrete and separate ediftces overlymg this lower sequence.The gold deposits are classified according to their stratigraphic or lithologic associations into:1) mafic-volcanic-hosted deposits;2) felsic-intrusive-hosted deposits; and3) stratabound deposits.Group 1 and to a lesser extent Group 2 deposits occur within zones of alteration several square kilometres in extent. The nature of this pervasive alteration can be characterized by comparison of petrochemical data from the altered rocks and their fresher equivalents some distance away. The altered rocks have substantial addition of SiO2> CO2, K2O, As and Sb and subtraction of Na20, CaO, MgO and total Fe. Cr, Ni, Co and Zr appear to have been relatively immobile. Hydrothermal alteration and gold deposition probably accompanied early fumarolic activity and continued during subsequent deformation the metavolcanic sequence, with gold eventually being mobtltzed and concentrated into major vein structures to form the present orebodies. The smaller Group 2 deposits formed by emplacement of felsic intrusions into already altered, anomalously gold-bearing mafic metavolcanics, with the heat from the intrusions setting up parasitic hydrothermal systems which mobilized the gold and deposited it along with minor sulphides in fractures in the cooling intrusions."

Jan 1, 1982

By Paterson R. G, Compston W

The Mine granite at the Ardlethan tinfield is dated by Rb-Sr on a muscovite concentrate at not less than 417 ± 2.5 Ma (2s), consistent with the intrusion of this granitic body into folded Late Ordovician sediments. A combination of Rb-Sr and K-Ar evidence suggests a slightly younger age, 41O ± 2.5 Ma, both for the neighbouring Ardlethan Granite and for the hydrothermal alteration of the Mine granite, with which the tin is associated.On this basis, the somewhat high 206Pb/204Pb of two galena samples from within the hydrothermal alteration zone is best explained by an increase in U/Pb of order 10 per cent during a geochemical event in the Late Precambrian, in the rocks which eventually yielded the Pb to the mineralising fluid. This earlier event probably increased Th/U in these rocks as well, by as much as 5 per cent. If these reconstructions are valid, the lower-than-average U/Pb implied for the original Precambrian source rock bespeaks a mantle component of marginally greater proportion than for "average orogene".

Jan 1, 1982

By Gordon A. Gross

Banded siliceous manganese-iron bearing sediments recognized as distinctive facies of Algoma type iron-formation constitute low grade resources of manganese in many countries. Manganiferous facies associated with oxide and carbonate facies of iron-formation have iron to manganese ratios ranging from 0.2 to 2 and commonly about one. Manganiferous facies up to 30 meters thick occurring separately or with other iron-formation facies are associated with cherts, carbonates, shales, turbidites, tuffs and lavas. Primary cherty manganiferous facies and their gondite metamorphic equivalents represent significant manganese resources and are protores for higher grade deposits derived by secondary enrichment processes.

Jan 1, 1982

By John C. Gould

GENERAL DESCRIPTION The Climax mine is located 161 km (100 miles) southwest of Denver on Fremont Pass astride the Con¬tinental Divide. Climax employs 2800 people to pro¬duce 42 670 t/d (47,000 stpd) of molybdenum ore. The Storke level, the main underground level, produces 19 970 t/d (22,000 stpd); the 600 level, the lowest ac¬tive underground area, produces 7260 t/d (8000 stpd); and the open pit presently produces 15 430 t/d (17,000 stpd), Fig. 1. A panel caving method is possible because of the favorable geology and geometry of the Climax ore body. Geology The Climax deposit is composed of three separate ore bodies formed in a massive Tertiary stockwork which penetrates Precambrian granites and schists. The ore bodies are elliptical in plan with overall dimensions of 914 x 1219 m (3000 x 4000 ft). In section the ore bodies are arcuate with an overall height of 457 m (1500 ft), although the depth is not firmly ascertained. Ore grade decreases rapidly into a highly siliceous cen¬tral core and more gradually toward the outer bounda¬ries of the deposit. The mineralization is relatively continuous in the ore zone and contains few dilution-causing waste inclusions. Molybdenite is contained in pre-ore fractures which were largely recemented during the three mineralization pulses. Postore fractures, with a spacing of a few inches to several meters (feet), render the deposit amenable to caving when it is undercut. The peripheral rock caves similarly to the ore, but the central core is competent and is difficult to cave. Caving Feasibility Successful caving action requires an undercut area of approximately 122 by 122 m (400 by 400 ft). When the undercut area exceeds this minimum distance, the arching effect is lost, and the ore column collapses. Ore extensions from the main cave area less than 122 m (400 ft) wide usually cannot be caved. Waste or low¬grade areas within the deposit can be left if they are large enough to carry weight around openings passing through them. Experience indicates that these areas should be over 30 m (100 ft) in length and width. Smaller unprofitable units of waste, such as postore dikes, are mined to avoid weight problems and result in ore dilution. Successful caving and ore recovery are dependent upon timely ore removal from the undercut zone because columns of immobile ore inhibit caving and concentrate weight onto the underlying workings. Climax experience indicates that a finger (drawpoint) spacing* averaging 10.4 m (34 ft) yields an excellent caving action and a 92.5% ore recovery at the estimated grade. An additional 15% of the ore is recovered-at a diluted grade (Fig. 2). Mine Layout The slusher drift is the primary production facility (Fig. 3). The ore is undercut and caved into the fingers,

Jan 1, 1982

By Larry Dale

An approach for predicting the development of mineral resources on public lands has been developed as an aid to resource agencies. It consists of new techniques for evaluating the economic potential of the resource, the uncertainty about mine cost estimates, and the uncertainty about the occurrance of mineral deposits. New mineral sup- plies from inferred and hypothetical resources can be predicted by using these techniques. A case study illustrates their application in predicting development of copper resources in western Montana.

Jan 1, 1982

By J. M. Duke

Significant resources grading less than about 1 % nickel occur in magmatic sulfide deposits of two types. In the Mt. Keith type, extensive zones of disseminated nickel sulfides occur in the central portions of certain synvolcanic dunite-peridotite sills of komatiitic affinity in Archean greens tone terranes. In the Duluth type, disseminated copper-nickel sulfides occur near the basal margins of intrusions of flood basaltic affinity associated with intracontinental rifting. In both types, the parent magmas became saturated with sulfide following the onset of fractional crystallization and the separation of the immiscible sulfide melt was accompanied by the precipitation of a relatively large proportion of silicates. The abundant silicate crystals impeded the coalescense and accumulation of the sulfide droplets with the resulting formation of large but low grade deposits.

Jan 1, 1982

By C. Alan Tapp

Introduction Mine sampling determines the practicality of any mining operation. Improper sampling can result in an incorrect appraisal of present production and future potential. Therefore, the mine department in charge of ore-reserve calculations and mine sampling should be overseen by competent and experienced professionals with technical backgrounds qualifying them to produce accurate results. Sampling is a process by which portions of an ore body are collected and analyzed to estimate the average mineral content of the entire ore body. It is incorrect to assume that a large number of samples eliminates any errors in the sampling method. To obtain unbiased samples, proper sample location with respect to rock type and mineralization is just as important. The sampling procedure must yield correct results for the type of mineral deposit, and careful consideration should be given to whether or not the sampling technique has been developed to an extent sufficient to eliminate as much human error and bias as possible. Only after the ore has been mined and milled is the sampling accuracy known. Sampling also provides information about the bulk composition of the ore for mineralogical and metallurgical tests that determine the economic ore-waste boundaries and the geologic trends for exploration. Actual mining plans can be developed from this information to maximize profits. Accurate sampling is critical, and thus must be approached in a scientific manner. Sampling Techniques Sampling practices and techniques are as varied as the mines in which they are used. The method(s) chosen must be tailored to suit the company and mining needs. For instance, tabular uranium deposits, vein gold deposits, and porphyry copper deposits pose special problems in conducting unbiased sampling. The mine geologist or engineer in charge must develop a sampling method, test it in a sample area, and then critically evaluate the results. If the results from the test area are accurate within the economic limits established by the company, they then may be adopted for general use in the mine. Four routine sampling methods are suitable for specific sampling objectives in the daily mine routine: (1) channel sampling, (2) chip sampling, (3) grab sampling, and (4) bulk sampling. The final sampling results depend upon how the four methods are combined to accurately determine the grade of the ore body. When used in conjunction with each other during different stages of mine development, the channel, chip, grab, and bulk sampling methods provide an in-house check or a comparison by which mining methods and sampling procedures can be evaluated. However, the most valid check is based upon the daily mill production. Channel Sampling Channel samples consist of cuttings collected from a groove cut into the rock about 102 mm (4 in.) wide and 19 mm (0.75 in.) deep. Various tools ranging from a 1.8-kg (4-lb) hammer and moil to a pneumatic chisel can be used to cut the sample. Accessibility and rock hardness determine the applicable sampling tools. Before attempting to take a sample, the rock surface must be cleaned thoroughly; the method of cleaning depends upon the amount of mine dust accumulated on the surface or the degree of alteration of the rock surface. Typical cleaning methods 'employ a wire brush, water, or chipping a fresh surface. Next, the sample outline is marked on the prepared surface, taking care to choose appropriate sample loca- tions. After determining the proper sample outline and length, the sample can be chiseled out, catching the rock fragments on a canvas tarpaulin on the floor, in a powder box, in a canvas bag to avoid contamination, or by some other suitable means to capture all of the sample. Vertical veins present a special problem because the drift's back usually is arched, not square. For example, Fig. 1 shows that if sample intervals are measured from A' to D', the length of the sample interval is greater than the true vein width, resulting in an incorrect calculation of the ore reserves. The correct method would collect three samples at A'-B', B'-C', and C'-D', using either actual measurements in the mine represented by A-B, B-C, and C-D or trigonometric calculations to determine the true ore thickness. Then, each sample can be weighted by grade and true thickness for the vein. Channel lengths usually are a maximum of 1.5 m (5 ft), and it is good practice to divide longer samples into smaller intervals according to structures, changes in rock types, or differences in rock hardnesses. The influence of those features on the mineralization then can be determined. Chip Sampling Chip sampling is a variation of channel sampling used when the rock is too hard to channel sample economically or when little variation in the mineral content indicates that this sampling method will yield results similar to those of channel sampling. Rather than cutting a channel in the rock, small chips are flaked off at regular intervals over the entire face or area being the mine. sampled. Care must be taken to assure that the sample is representative of variations in the rock hardness and type. hi^ method is fast and useful in preliminary evaluations, but it should not be used for quantitative ore-reserve calculations. Grab Sampling Grab sampling is a fast method for double checking either channel or chip-sampling procedures, and, in some instances, mine production can be estimated from carefully taken grab samples. Grab sampling takes equal amounts of material at selected intervals over a mine dump, a muck pile, or from an ore car to estimate its Channel Sampling mineral content. Generally, this method is not considered reliable. Many independent variables can affect this type of sampling process. Thus, if the ore occurs in the softer fraction and a proportional amount of the result-

Jan 1, 1982

By Arthur H. Wilson

INTRODUCTION The stratified deposits of the coal measures are strongly laminated and exhibit a high degree of anisotropy. The stronger rocks in the sequence possess joints and shear planes, the coal itself has a closely spaced cleat pattern, and the softer rocks are capable of flowage. To calculate the general distribution of stress in such a medium by analytical means based on the theories of elasticity is difficult, and has not yet been satisfactorily achieved. Nevertheless, some estimate is required to aid the design of supporting pillars and permit the correct siting of main roadways. This is important when laying out districts for longwall extraction, particularly if the results of previous experience are not available in the area to be exploited. Although elastic stress analyses, including finite element analysis, may not be very appropriate, the basic equilibrium conditions will apply. These, combined with logical deductions based on observed fact, can be used to give appropriate solutions which experience has shown to be capable of giving acceptable results in the design of pillars associated with longwall workings. Constancy of cover load is one of the fundamental equilibrium conditions. When mining below ground, the cover is not removed, therefore the average vertical load after mining must equal the initial cover load. If mining produces a local decrease in load in any particular area, it must be balanced by an equivalent increase in load nearby. Bending moments in the strata above are capable of compensating for the turning moments which this redistribution of load may produce, and the existence of a "rear abutment", much discussed between 1950 and 1960, is not an essential equilibrium requirement. If the distribution of vertical stress is plotted three-dimensionally, then the load on any small element of area 6S will be o.6S, where o is the average stress. This also represents the volume of the "stress envelope" which lies above 6s. By making the individual areas very small and summing together, it is possible to show that the total load on any particular area can be found by calculating the volume of the stress distribution envelope which lies above it. A similar consideration applies to the stress distribution in a two-dimensional situation. If the stress distribution curve is drawn, then the area below any portion will represent the load per unit depth of section. For example, Figure 1 represents the possible stress distribution across a caved waste and associated ribside. If AW is the load deficiency [ ] These loads must balance, therefore area AW = AS, and if one can be deduced, the other will then be known. Attempts have been made to measure stress distributions in the ribside and in the waste. Stress meters in soft rocks such as the coal measures have not proved very effective. Those based on the measurement of strain have experienced difficulty in strata capable of flowage, high modulus stress meters suffer from cross sensitivity, and overcoming to obtain absolute stress is almost impossible. Load measuring instruments left in the waste have also proved largely ineffective, mainly as a result of the difficulties of protecting the leads in a caving environment. Certain logical deductions, however, can be made about the conditions in the waste. CONDITIONS IN THE WASTE Over the area of the waste the surface trough which develops is remarkably consistent, despite the geological variations which occur from coalfield to coalfield. In Britain, an accuracy of 210% is claimed for subsidence prediction based on width of extraction and depth of cover (National Coal Board, 1975). The subsidence in the USA is more variable (Adamek and Jeran, 1981), particularly with respect to the point of inflexion, but the same general pattern occurs. At the shallower depths the somewhat harder roof strata involved in the caving in the USA may suffer less compaction, giving reduced values of subsidence. Subsidence at intermediate horizons below the surface

Jan 1, 1982