of carbon monoxide, hydrogen, hydrogen sulfide, and sulfur dioxide. Aqueous species of these gases are absent from the lake waters as well, and carbon dioxide (CO2) forms 98 to 99 percent of the gas. The Nyos gas is cool; the lake-bottom temperature of 75 °F is no higher than those in other tropical lakes at similar elevation and latitude.

Stratification in lakes is produced by a layer of less dense water overlying a layer of more dense water. Concentrations of dissolved substances in Lake Nyos waters show a definite increase with depth, implying that the lake was stratified and that the dissolved substances were stored in the bottom waters prior to the event. The high concentrations of dissolved ferrous iron (Fe2+) in the presence of bicarbonate (HCO3) indicate that the lake waters were very acidic prior to the event; otherwise, the iron would have been removed by chemical precipitation of the mineral siderite (FeCO3). A large amount of dissolved CO2 is the only likely candidate for producing the lake's acidity.

Oxygen and hydrogen isotopic compositions of Lake Nyos waters indicate that the solutes within the lake were not acquired through the normal process of evaporative concentration. The solutes, including the CO2, likely entered through springs near the bottom of the lake where confining pressures would be sufficient to keep the gas dissolved. The analytical data clearly indicate that the Nyos gas, although released from magma at great depths below the lake, had cooled and lost its reactive constituents (such as sulfur and chlorine compounds) as it moved upward through the permeable diatreme. The gas reached the lake as cool, essentially pure CO2.

Calculations show that Lake Nyos can hold more than a quarter of a cubic mile of dissolved CO2 when fully saturated. Laboratory measurements indicate that the lake currently holds 0.07 mi3 of CO2. The lake apparently released up to 0.29 mi3 of CO2, an amount sufficient to account for the deaths and the drop in lake level immediately following the event. The gas cloud most likely was produced by the rapid exsolution of large amounts of CO2. The gas probably made an explosion-like noise and caused surface waves to form as it escaped from the lake. Because CO2 is 1.5 times more dense than air, the gas cloud tended

to maintain its integrity as it spilled over the crater rim into low-lying areas.

The mechanisms that may have caused the sudden release of the gas from the bottom waters of Lake Nyos include a small earthquake, a landslide into the lake, or the sudden addition of CO2 in the bottom of the lake. Essentially any disturbance of the unstable, CO2-charged bottom waters would cause a chain reaction that could result in the rapid exsolution and explosive release of the CO2.

The people and animals appeared to have died quickly of asphyxia (fig. 4). Very high concentrations of CO2 cause paralysis of the respiratory centers in the brain. The description by many survivors of the odor of "rotten eggs" or "gunpowder," although usually associated with sulfur gases, is a common "olfactory hallucination" described by individuals exposed to high concentrations of CO2.

The results of this investigation provided unprecedented insight into the

Figure 3. Hypothesized schematic cross section of the diatreme beneath Lake Nyos. This structure allowed magmatic gas to migrate into the lake where

it was stored prior to its catastrophic release.

Figure 4. Dead cattle littered the hillsides around Lake Nyos after the event.

conditions associated with the release of lethal gas from crater lakes. This insight will have application to other crater lakes around the world that have the potential for similar disasters.

Costa Rican Mineral Resource Assessment Study

By Richard D. Krushensky

In 1987, the U.S. Geological Survey (USGS), the Direccion General de Minas e Hidrocarburos, and the Universidad de Costa Rica, Republic of Costa Rica, continued a project to assess the non-fuel mineral resources of Costa Rica. Grade and tonnage of known deposits were assessed, and the probability for the discovery, and grade, including quantity, were estimated for presently unknown deposits. Such estimates are a major factor in the analysis of mineral commodities. The assessment was based on an evaluation of existing geologic, geochemical, and geophysical data as well as on new data gathered during intensive field studies. From these data, the team of scientists identified several areas, called "domains," where the geology is conducive to the development of one or more types of mineral deposits. Gold and silver are presently mined in Costa Rica. Occurrences of copper, lead, zinc, manganese, chromium, aluminum, and iron- and titanium-bearing sands are also present. Although some of these latter commodities have been mined in the past, none are now being mined.

Gold, the most important Costa Rican metallic mineral resource, is presently mined from veins and alluvial deposits. Two domains have been delineated for vein gold deposits and two for placer gold deposits. Gold production in Costa Rica has been primarily from veins emplaced near the Earth's surface. USGS scientists predict that there is a 90 percent chance of discovering 4 additional but presently unknown vein gold deposits, and a 10 percent chance of discovering 17 vein gold deposits in a belt that extends from Cañas, along the Cordillera de Tilaran in the north, to Pueblo

Nuevo, just within the Cordillera de Talamanca, in the south (fig. 5).

Geologic study indicates a spatial and genetic relationship between the intrusion of rhyolite and vein gold deposits in the Santa Clara Mine. This study has also indicated many more rhyolitic intrusions than were previously known. As a result, the probability for the discovery of more vein gold in the Santa Clara deposit is greatly increased. Undiscovered vein gold deposits would be most likely to occur near rhyolitic intrusions, probably localized in quartz veins occupying shear or fault zones. Although domains for placer gold have been well explored in the area of the

Osa Peninsula, two newly delineated domains in this area have high potential for hosting fossil placer gold deposits.

The mineral resource assessment also indicated that resources of copper, lead, zinc, manganese, chromium, and aluminum also exist in Costa Rica, and domains for deposits of these commodities have been. delineated. Small deposits of podiform chromite are known from the Santa Elena Peninsula in northwestern Costa Rica, and there is a 50 percent probability of 13 more and a 10 percent chance of 63 or more similar but undiscovered deposits in that same domain. Because all known chromite deposits in Costa Rica contain less than 43

metric tonnes, it is assumed that any additional undiscovered deposits will also be small.

The study also indicates that there is a 50 percent probability for the discovery of at least three porphyry copper deposits in Costa Rica. Such deposits typically contain 140 million metric tonnes or more of ore that ranges from 0.15 to 2 percent copper. It is estimated that there is a 90 percent chance of one such deposit, a 50 percent chance of three, and a 10 percent chance of eight porphyry copper deposits being discovered within one domain in the Cordillera de Talamanca in southern Costa Rica. Detailed geologic mapping, rock and soil sampling, ground geophysical surveys, and drilling of defined target areas need to be carried out in order to identify these potential new deposits.

Volcanogenic manganese is known in more than 90 deposits in two domains in the Nicoya Peninsula, and undiscovered deposits like those known at the surface are believed to exist at depth. Bauxite, an ore of aluminum, has been thoroughly explored for in two domains in the area of the Valle General (valley of the Rio General) and southward, almost to the Panama border. There is a distinct possibility of additional undiscovered bauxite in a similar domain in the area of Guapiles in east-central Costa Rica. Three very small massive-sulfide deposits, with major copper and minor zinc content, occur in the Nicoya Peninsula; there is a 50 percent probability that one other such deposit exists in this same domain.

Other deposit types present in Costa Rica for which domains have not been identified include low-grade offshore placer iron (magnetite), polymetallic veins (veins with three or more metals), copper skarn (deposits formed at the contact between intrusive rocks and sedimentary country rock), lead-zinc skarn, and hot-spring sulfur deposits. The mineral-resource assessment of Costa Rica was carried out with funds provided by the Agency for International Development through the Los Alamos National Laboratory. The assessment supports the President's Caribbean Initiative whereby nonmilitary methods are sought to assist Central American countries in improving their economic conditions.

Coal Resource Assessment in

Pakistan

Exploratory drilling during fiscal year 1987, as a part of the Coal Resource Exploration and Assessment Program (COALREAP), has significantly increased knowledge of the coal resource potential in the Sind Province in southern Pakistan. COALREAP is a technical assistance venture by the U.S. Geological Survey with the Geological Survey of Pakistan under the auspices of the Energy Planning and Development Project of the U.S. Agency for International Development (USAID). Drilling is done by a USAID-contracted Pakistani company, and geophysical logging of holes is done by a U.S. company, also contracted by USAID.

Since April 1986, 32 holes, totaling 25,306 feet (53 percent cored), have been drilled and logged (fig. 6). The holes averaged approximately 800 feet in depth; the deepest hole was 1,247 feet, the shallowest 423 feet. Coal beds that were penetrated by drilling range from a few inches to about 20 feet in thickness; however, in only one hole did a coal bed exceed 7 feet in thickness. Coal beds more than 2 feet thick were drilled in 17 holes; 2 holes contained 4 such beds, 6 holes had 3. Samples of coal were collected from all beds that were a foot or more thick. In all, 141 coal samples were collected from 26 drill holes. The samples are presently being analyzed, but early analysis of the 20-foot-thick bed indicates that it is of good quality. The middle section of the 20-foot bed contains an average of 3.3 percent ash and 0.6 percent sulfur; the parts of the bed above and below contain somewhat higher ash and sulfur percentages.

Preliminary interpretations of recent investigations suggest that:

The presently known and mined coal fields (Lakhra, Sonda-Thatta, MetingJhimpir) are not isolated occurrences but rather are in a region apparently completely underlain by coal-bearing strata.

Several exploratory drill holes did not penetrate coal, but regularly occurring delimitors for individual coal fields were not found.

Most coal beds appear to be discontinuous layers, and, although many different