Geology topics

B. D. Keeland

Publications and source records attributed to B. D. Keeland.

13 recordsLinked to original sources

USGS2009

Delayed tree mortality in the Atchafalaya Basin of Southern Louisiana following Hurricane Andrew

Hurricanes can damage trees in forested wetlands, and the potential for mortality related to these storms exists due to the effects of tree damage over time. In August 1992, Hurricane Andrew passed through the forested wetlands of southern Louisiana with winds in excess of 225 kph. Although more than 78 of the basal area was destroyed in some areas, most trees greater than 2.5 cm dbh were alive and resprouting prolifically the following year (98.8). Survival of most tree species was similarly high two years after the hurricane, but mortality rates of some species increased dramatically. For example, Populus heterophylla (swamp cottonwood) mortality increased from 7.8 to 59.2 (n 76) and Salix interior (sandbar willow) mortality increased from 4.5 to 57.1 (n 21). Stem sprouts on many up-rooted hardwood trees of other species were still alive in 1998, 6 years after the hurricane. Due to the understory tree species composition, regeneration, and high levels of resprouting, there was little change in species composition or perhaps a slight shift toward more shade and flood tolerant species six years following the hurricane event. Triadica sebifera (Chinese tallow) was found on some of the sites heavily disturbed by Hurricane Andrew, and may proliferate at the expense of native tree species. ?? 2009 The Society of Wetland Scientists.

Wetlands↗

USGS2007

Effects of season, rainfall, and hydrogeomorphic setting on mangrove tree growth in Micronesia

Seasonal patterns of tree growth are often related to rainfall, temperature, and relative moisture regimes. We asked whether diameter growth of mangrove trees in Micronesia, where seasonal changes are minimal, is continuous throughout a year or conforms to an annual cycle. We installed dendrometer bands on Sonneratia alba and Bruguiera gymnorrhiza trees growing naturally within mangrove swamps on the islands of Kosrae, Federated States of Micronesia (FSM), Pohnpei, FSM, and Butaritari, Republic of Kiribati, in the eastern Caroline Islands of the western Pacific Ocean. Trees were remeasured monthly or quarterly for as long as 6 yr. Annual mean individual tree basal area increments ranged from 7.0 to 79.6 cm2/yr for all S. alba trees and from 4.8 to 27.4 cm2/yr for all B. gymnorrhiza trees from Micronesian high islands. Diameter increment for S. alba on Butaritari Atoll was lower at 7.8 cm 2/yr for the one year measured. Growth rates differed significantly by hydrogeomorphic zone. Riverine and interior zones maintained up to seven times the annual diameter growth rate of fringe forests, though not on Pohnpei, where basal area increments for both S. alba and B. gymnorrhiza were approximately 1.5 times greater in the fringe zone than in the interior zone. Time-series modeling indicated that there were no consistent and statistically significant annual diameter growth patterns. Although rainfall has some seasonality in some years on Kosrae and Pohnpei and overall growth of mangroves was sometimes related positively to quarterly rainfall depths, seasonal diameter growth patterns were not distinctive. A reduced chance of moisture-related stress in high-rainfall, wetland environments may serve to buffer growth of Micronesian mangroves from climatic extremes. ?? 2007 The Author(s) Journal compilation ?? 2007 by The Association for Tropical Biology and Conservation.

Biotropica↗

USGS2005

Environmental impacts of oil production on soil, bedrock, and vegetation at the U.S. Geological Survey Osage-Skiatook Petroleum Environmental Research site A, Osage County, Oklahoma

The U.S. Geological Survey is investigating the impacts of oil and gas production on soils, groundwater, surface water, and ecosystems in the United States. Two sites in northeastern Oklahoma (sites A and B) are presently being investigated under the Osage–Skiatook Petroleum Environmental Research project. Oil wells on the lease surrounding site A in Osage County, Oklahoma, produced about 100,000 bbl of oil between 1913 and 1981. Prominent production features on the 1.5-ha (3.7-ac) site A include a tank battery, an oil-filled trench, pipelines, storage pits for both produced water and oil, and an old power unit. Site activities and historic releases have left open areas in the local oak forest adjacent to these features and a deeply eroded salt scar downslope from the pits that extends to nearby Skiatook Lake. The site is underlain by surficial sediments comprised of very fine-grained eolian sand and colluvium as much as 1.4 m (4.6 ft) thick, which, in turn, overlie flat-lying, fractured bedrock comprised of sandstone, clayey sandstone, mudstone, and shale. A geophysical survey of ground conductance and concentration measurements of aqueous extracts (1:1 by weight) of core samples taken in the salt scar and adjacent areas indicate that unusual concentrations of NaCl-rich salt are present at depths to at least 8 m (26 ft) in the bedrock; however, little salt occurs in the eolian sand. Historic aerial photographs, anecdotal reports from oil-lease operators, and tree-ring records indicate that the surrounding oak forest was largely established after 1935 and thus postdates the majority of surface damage at the site. Blackjack oaks adjacent to the salt scar have anomalously elevated chloride ( 400 ppm) in their leaves and record the presence of NaCl-rich salt or salty water in the shallow subsurface. The geophysical measurements also indicate moderately elevated conductance beneath the oak forest adjoining the salt scar.

Oklahoma↗

USGS2002

Evaluation of reforestation in the Lower Mississippi River Alluvial Valley

Only about 2.8 million ha of an estimated original 10 million ha of bottomland hardwood forests still exist in the Lower Mississippi River Alluvial Valley (LMAV) of the United States. The U.S. Fish and Wildlife Service, the U.S. Forest Service, and state agencies initiated reforestation efforts in the late 1980s to improve wildlife habitat. We surveyed restorationists responsible for reforestation in the LMAV to determine the magnitude of past and future efforts and to identify major limiting factors. Over the past 10 years, 77,698 ha have been reforested by the agencies represented in our survey and an additional 89,009 ha are targeted in the next 5 years. Oaks are the most commonly planted species and bare-root seedlings are the most commonly used planting stock. Problems with seedling availability may increase the diversity of plantings in the future. Reforestation in the LMAV is based upon principles of landscape ecology; however, local problems such as herbivory, drought, and flooding often limit success. Broad-scale hydrologic restoration is needed to fully restore the structural and functional attributes of these systems, but because of drastic and widespread hydrologic alterations and socioeconomic constraints, this goal is generally not realistic. Local hydrologic restoration and creation of specific habitat features needed by some wildlife and fish species warrant attention. More extensive analyses of plantings are needed to evaluate functional success. The Wetland Reserve Program is a positive development, but policies that provide additional financial incentives to landowners for reforestation efforts should be seriously considered.

Restoration Ecology↗

USGS2002

Arthropods in decomposing wood of the Atchafalaya River basin

Changes in arthropod populations (numbers of individuals identified to the family level in most cases) were studied during the decomposition of coarse woody debris (CWD) in the Atchafalaya River Basin of Louisiana. The arthropod study was linked with a CWD decomposition study installed after disturbance by Hurricane Andrew. Arthropod numbers were compared between two canopy disturbance classes and between two spatial orientations of CWD (i.e., suspended above- and in contact with the soil). Results during 30 months in the field suggested little influence of canopy disturbance or spatial orientation of CWD on arthropod numbers. Counts were most frequently dominated by Collembola and Acarina and peaked after 18–24 months within larger debris.

Louisiana↗

USGS2002

Effects of canopy gaps and flooding on homopterans in a bottomland hardwood forest

Canopy disturbance is a major factor affecting forest structure and composition and, as a result of habitat alterations, can influence insect communities. We initiated a field study to quantify the effects of canopy disturbance on aerial insect abundance and distribution within a bottomland hardwood forest along the Cache River, Arkansas, USA. We used passive flight-intercept traps to sample insects in canopy gap and forest interior habitats from May to July in 1996, 1997, and 1998. The hydrologic conditions of our study site varied among years: 1996 was relatively dry, 1997 incurred a long-duration flood, and 1998 was moderately wet. Of the 34,000+ Homopterans collected, many groups were distributed in a non-uniform manner among years and between habitats. Total Homopterans, two families of Homopterans, and six morphospecies were more abundant in canopy gaps than interior forest. Many Homopteran taxa were least abundant in 1997 following almost six months of flooding. Alternatively, relatively large Homopteran abundances were associated with the dry conditions of 1996 and the moderately wet conditions of 1998. Differences in Homopteran abundance among years and habitats may be related to differences in vegetation density. Canopy gaps supported more vegetation cover than the interior forest in all but the first sampling interval. In addition, similar to Homopteran abundance, vegetation density was lower in 1997 than in 1998. These results demonstrate that natural disturbance and flooding contribute to Homopteran abundance and distribution patterns in bottomland hardwood forests of the south central United States.

Arkansas↗

USGS2001

A Guide to Bottomland Hardwood Restoration

During the last century, a large amount of the original bottomland hardwood forest area in the United States has been lost, with losses greatest in the Lower Mississippi Alluvial Valley and East Texas. With a holistic approach in mind, this manual describes methods to restore bottomland hardwoods in the lower Midwest, including the Lower Mississippi Alluvial Valley and the southeastern United States. Bottomland hardwoods in this guide include not only the hardwood species that predominate in most forested floodplains of the area but also the softwood species such as baldcypress that often co-occur. General restoration planning considerations are discussed as well as more specific elements of bottomland hardwood restoration such as species selection, site preparation, direct seeding, planting of seedlings, and alternative options for revegetation. We recognize that most projects will probably fall more within the realm of reforestation or afforestation rather than a restoration, as some site preparation and the planting of seeds or trees may be the only actions taken. Practical information needed to restore an area is provided in the guide, and it is left up to the restorationist to decide how complete the restoration will be. Postplanting and monitoring considerations are also addressed. Restoration and management of existing forests are included because of the extensive areas of degraded natural forests in need of rehabilitation.

Information and Technology Report↗

USGS2000

Downed wood in Micronesian mangrove forests

Dead, downed wood is an important component of upland forest and aquatic ecosystems, but its role in wetland ecosystems, including mangroves, is poorly understood. We measured downed wood in ten sites on the western Pacific islands of Kosrae, Pohnpei, and Yap, all located within the Federated States of Micronesia. Our goals were to examine patterns of variability in the quantity of downed wood in these mangrove ecosystems, provide a general characterization of downed wood in a region with no previously published accounts, and investigate the relationship between harvesting practices and the amount of downed wood. The overall mean volume of downed wood at our study sites was estimated to be 60.8 m3 ha-1 (20.9 t ha-1), which is greater than most published data for forested wetlands. There were significant differences among islands, with the sites on Kosrae (104.2 m3 ha-1) having a much greater mean volume of downed wood than those on Pohnpei (43.1 m3 ha-1) or Yap (35.1 m3 ha-1). Part of the difference among islands may be attributable to differences in stand age and structure, but the most important factor seems to be the greater amount of wood harvesting on Kosrae, coupled with a low efficiency of use of cut trees. Of a total of 45 cut trees examined on Kosrae, no wood had been removed from 18 (40%); these are believed to be trees cut down because other, more valuable, trees were caught on them as they were felled. Of the other 27 trees, only 24 to 42% of the stem volume (to a 10 cm top) was removed from the forest, the amount varying by species. The impacts of current harvesting practices are unknown but may include important effects on tree regeneration and the abundance and species composition of crab populations.

Wetlands↗

USGS1999

Natural regeneration and growth of Taxodium distichum (L.) rich. In Lake Chicot, Louisiana after 44 years of flooding

Lake Chicot, in south central Louisiana, USA, was created in 1943 by the impoundment of Chicot Bayou. Extensive establishment of woody seedlings occurred in the lake during a 1.5 year period, including the growing seasons of both 1986 and 1987, when the reservoir was drained for repair work on the dam. Study plots were established in September 1986 to document woody vegetation establishment and to provide a baseline by which to monitor survival and growth after flooding resumed. Taxodium distichum seedlings were the dominant species after one growing season, with a maximum density of 50 seedlings/m 2 , an average of about 2/m 2 , and an average height of 75 cm. The lake was reflooded at the end of 1987, bringing water depths at the study plots up to about 1.4 m. Temporary drawdowns were again conducted during the fall of 1992 and 1996. In December 1992, the site was revisited, new plots established, and saplings counted and measured. There was an average of 2.1 T. distichum stems/m 2 , and the average height was 315 cm. After the 1996 growing season, there was still an average of about 1.9 stems/m 2 , and the average height had increased to 476 cm. Preservation of T. distichum forests in relatively shallow but continuously flooded areas such as Lake Chicot may be a simple matter of draining the lake after a good seed crop and maintaining the drawdown long enough for the seedlings to grow taller than the typical growing season water level. In the case of Lake Chicot, this period was two growing seasons. This action will mimic natural, drought-related drawdowns of the lake and will allow the seedlings to establish themselves and grow tall enough to survive normal lake water levels.

Louisiana↗

USGS1998

Beaver lodge distributions and damage assessments in a forested wetland ecosystem in the southern United States

Caddo Lake, USA, a Ramsar Wetland of International Importance, is a lacustrine wetland complex consisting of stands of flooded baldcypress intermixed with open water and emergent wetland habitats. Recently, concern has been expressed over a perceived increase in the beaver population and the impact of beaver on the long-term sustainability of the baldcypress ecosystem. We used intensive beaver lodge surveys to determine the distribution and relative abundance of beaver and the amount, type, and distribution of beaver damage to mature trees and seedlings at Caddo Lake. A total of 229 lodges were located with a combination of aerial and boat/ground surveys. Most lodges were located in open water and edge habitats. About 95% of the lodges were occupied by beaver or nutria. Some form of damage was exhibited by one or more trees near 85% of the lodges. Intensive damage assessments around 35 lodges indicated that most damage to trees, baldcypress in particular, was restricted to peeling or stripping of bark which is believed to have minimal effect on tree survival. Surveys of regeneration indicated that baldcypress seedlings were very abundant; however, over 99.9% were less than 30 cm tall. The lack of recruitment into the larger size classes appears to be a result of high stand densities and water management practices. At this time, the young age and density of the baldcypress forests suggest that recruitment is not a major concern and herbivore damage appears to be having a minimal effect on the forest.

Forest Ecology and Management↗

USGS1998

Natural establishment of woody species on abandoned agricultural fields in the lower Mississippi Valley: first- and second-year results

The natural establishment of woody seedlings on abandoned agricultural fields was investigated at sites in Louisiana and Mississippi. Series of disked and undisked plots originating at forest edges and oriented in cardinal directions were established on fields at each site. During the firest 2 years, seedling recruitment was dominated by sweetgum, sugarberry, and elms at both sites. Seedling establishment was strongly affected by direction from mature forest and disking, and to a slightly lesser degree by distance from mature forest. Slightly under half of the variation in seedling numbers per plot was explained by the effects of direction, distance, and disking, indicating that other factors also may play an important role in seedling recruitment.

General Technical Report SRS↗

USGS1997

Long-term growth trends of baldcypress (Taxodium distichum (L.) rich.) at Caddo Lake, Texas

Caddo Lake, situated on the border of northeast Texas and northwest Louisiana, USA is a medium-sized lake dominated by stands of baldcypress (Taxodiwn distichum). A study of tree growth was initiated at Caddo Lake to address concerns about the health of the baldcypress ecosystem. The lake has been subjected to several dramatic water-level changes over the past 200 years, including water-level stabilization following dam construction in 1914. To assess the long-term growth trends of baldcypress trees and determine if a recent growth decline is occurring at Caddo Lake, increment cores were taken from 52 trees. The cores were crossdated and rings between the years 1900 and 1992 measured to the nearest 0.01 mm. Most cores were characterized by high variation in year-to-year growth. Although increasing growth rates were observed at most locations, trees from two backwater areas of the lake had recent growth rates lower than their long-term average. Growth amounts at these two sites were, however, within the historic range of variation. No recruitment was observed. From these data, we can conclude that the historic, extreme changes in hydrologic regime and the current stabilized water levels have not resulted in an overall decline in baldcypress growth at Caddo Lake.

Texas↗

USGS1997

Use of inexpensive pressure transducers for measuring water levels in wells

Frequent measurement of below ground water levels at multiple locations is an important component of many wetland ecosystem studies. These measurements, however, are usually time consuming, labor intensive, and expensive. This paper describes a water-level sensor that is inexpensive and easy to construct. The sensor is placed below the expected low water level in a shallow well and, when connected to a datalogger, uses a pressure transducer to detect groundwater or surface water elevations. Details of pressure transducer theory, sensor construction, calibration, and examples of field installations are presented. Although the transducers must be individually calibrated, the sensors have a linear response to changing water levels ( r 2 ≈ .999). Measurement errors resulting from temperature fluctuations are shown to be about 4 cm over a 35°C temperature range, but are minimal when the sensors are installed in groundwater wells where temperatures are less variable. Greater accuracy may be obtained by incorporating water temperature data into the initial calibration (0.14 cm error over a 35C temperature range). Examples of the utility of these sensors in studies of groundwater/surface water interactions and the effects of water level fluctuations on tree growth are provided.

Wetlands Ecology and Management↗