Geology topics
Research about Maine
Source-linked reports with geographic coverage including Maine.
Structure and variability of the Western Maine Coastal Current
Analyses of CTD and moored current meter data from 1998 and 2000 reveal a number of mechanisms influencing the flow along the western coast of Maine. On occasions, the Eastern Maine Coastal Current extends into the western Gulf of Maine where it takes the form of a deep (order 100 m deep) and broad (order 20 km wide) southwestward flow with geostrophic velocities exceeding 20 cm s -1. This is not a coastally trapped flow, however. In fields of geostrophic velocity, computed from shipboard-CTD data, the core of this current is roughly centered at the 100 m isobath and its onshore edge is no closer than 10 km from the coast. Geostrophic velocity fields also reveal a relatively shallow (order 10 m deep) baroclinic flow adjacent to the coast. This flow is also directed to the southwest and appears to be principally comprised of local river discharge. Analyses of moored current meter data reveal wind-driven modulations of the coastal flow that are consistent with expectations from simple theoretical models. However, a large fraction of the near-shore current variance does not appear to be directly related to wind forcing. Sea-surface temperature imagery, combined with analysis of the moored current meter data, suggests that eddies and meanders within the coastal flow may at times dominate the near-shore current variance. ?? 2005 Elsevier Ltd. All rights reserved.
Steam injection pilot study in a contaminated fractured limestone (Maine, USA): Modeling and analysis of borehole radar reflection data
Steam-enhanced remediation (SER) has been successfully used to remove DNAPL and LNAPL contaminants in porous media. Between August and November 2002, SER was tested in fractured limestone at the former Loring Air Force Base, in Maine, USA. During the SER investigation, the U.S. Geological Survey conducted a series of borehole radar surveys to evaluate the effectiveness of radar methods for monitoring the movement of steam and heat through the fractured limestone. The data were collected before steam injection, 10 days after the beginning of injection, and at the end of injection. In this paper, reflection-mode borehole radar data from wells JBW-7816 and JBW-7817A are presented and discussed. Theoretical modeling was performed to predict the variation of fracture reflectivity owed to heating, to show displacement of water and to assess the effect of SER at the site. Analysis of the radar profile data indicates some variations resulting from heating (increase of continuity of reflectors, attenuation of deeper reflections) but no substantial variation of traveltimes. Spectral content analysis of several individual reflections surrounding the boreholes was used to investigate the replacement of water by steam in the fractures. Observed decrease in radar reflectivity was too small to be explained by a replacement of water by steam, except for two high-amplitude reflectors, which disappeared near the end of the injection; moreover, no change of polarity, consistent with steam replacing water, was observed. The decrease of amplitude was greater for reflectors near well JBW-7817A and is explained by a greater heating around this well.
The role of pH in structuring communities of Maine wetland macrophytes and chironomid larvae (Diptera)
Aquatic vascular plants, or macrophytes, are an important habitat component for many wetland organisms, and larvae of chironomid midges are ubiquitous components of wetland fauna. Many chironomids are primary consumers of algae and detritus and form an essential energetic link between allochthonous and autochthonous primary production and higher trophic levels, while others are predators and feed on smaller invertebrates. Live macrophytes serve mostly as habitat, whereas plant detritus serves as both habitat and as a food source. Assemblages of macrophytes and chrinomid larvae were surveyed in ten Maine wetlands, five with low pH (<5.0) and five with high pH (>5.5), and explained in terms of physical and chemical habitat variables. Macrophyte richness was significantly greater, and richness of chironomid larvae was lower, in low pH wetlands. There was no difference in chironomid abundance related to pH. However, community structure was related to pH, suggesting that competitive dominance of a few taxa was responsible for lower richness in low pH wetlands, whereas competition was weaker in high pH wetlands, making coexistence of more chironomid taxa possible. An examination of individual chironomid taxa by stepwise multiple regression showed that distribution of most taxa was controlled by water chemistry variables and macrophyte habit (i.e., floating, submergent).
Water level dynamics in wetlands and nesting success of Black Terns in Maine
The Black Tern ( Chlidonias niger ) nests in freshwater wetlands that are prone to water level fluctuations, and nest losses to flooding are common. We examined temporal patterns in water levels at six sites with Black Tern colonies in Maine and determined probabilities of flood events and associated nest loss at Douglas Pond, the location of the largest breeding colony. Daily precipitation data from weather stations and water flow data from a flow gauge below Douglas Pond were obtained for 1960-1999. Information on nest losses from three floods at Douglas Pond in 1997-1999 were used to characterize small (6% nest loss), medium (56% nest loss) and large (94% nest loss) flood events, and we calculated probabilities of these three levels of flooding occurring at Douglas Pond using historic water levels data. Water levels generally decreased gradually during the nesting season at colony sites, except at Douglas Pond where water levels fluctuated substantially in response to rain events. Annual probabilities of small, medium, and large flood events were 68%, 35%, and 13% for nests initiated during 23 May-12 July, with similar probabilities for early (23 May-12 June) and late (13 June-12 July) periods. An index of potential nest loss indicated that medium floods at Douglas Pond had the greatest potential effect on nest success because they occurred relatively frequently and inundated large proportions of nests. Nest losses at other colonies were estimated to be approximately 30% of those at Douglas Pond. Nest losses to flooding appear to be common for the Black Tern in Maine and related to spring precipitation patterns, but ultimate effects on breeding productivity are uncertain.
Negative effects of commercial mussel dragging on eelgrass beds in Maine
A study by the US Geological Survey, the University of New Hampshire, and the Maine Department of Marine Resources showed that commercial mussel dragging poses a severe and long-lasting threat to eelgrass (Zostera marina). Dragging can damage large areas, with individual drag scars up to 79 acres in size found in Maine eelgrass beds. Dragging activity uproots eelgrass plants completely, removing leaves, rhizomes, and roots. Two independent methods were used to predict the rate of eelgrass recovery in heavily dragged areas. Under the best environmental conditions, complete revegetation of a dragged area would require an average of 11 years. Under conditions less favorable for eelgrass growth, such as reduced water quality, dragged areas could require more than 20 years to recover. Protection of eelgrass from commercial shellfish dragging will preserve important coastal habitat.
Surface-geophysical investigation of a formerly used defense site, Machiasport, Maine, February 2003
Surface-geophysical surveys were conducted in February 2003 at a formerly used defense site in Maine, where residual chlorinated solvents are affecting off-site domestic water-supply wells. The U.S. Geological Survey and Argonne National Laboratory used surface-geophysical methods, including ground-penetrating radar and seismic-refraction tomography, to characterize the lithology and structure of the bedrock at the site and to identify highly fractured areas that may provide pathways for ground-water flow and contaminant transport. Multifrequency electromagnetic and inductive terrain-conductivity methods also were evaluated, but these techniques were adversely affected by a nearby naval computer and telecommunications station. Interpretation of the data from ground-penetrating radar indicates that depth to the weathered bedrock surface is approximately 0.5 to 3 meters. Reflections from within the bedrock are visible throughout all ground-penetrating radar profiles, and zones of scattered electromagnetic energy may correlate to zones of highly fractured bedrock. Interpretation of the data from seismic-refraction tomography inversion indicates that zones of relatively low seismic velocity and topographic lows may correlate with fractured and water-producing intervals within the bedrock. Integrated interpretation of the results from ground-penetrating radar and seismic-refraction tomography was used to locate boreholes along the surface-geophysical profiles. An integrated analysis of information obtained from the surface- and borehole-geophysical surveys and test drilling will be used by the U.S. Army Corps of Engineers to develop a conceptual model of ground-water flow and solute transport at the site.
Trends in timing, magnitude, and duration of summer and fall/winter streamflows for unregulated coastal river basins in Maine during the 20th century
The U.S. Geological Survey (USGS), in cooperation with the Maine Atlantic Salmon Commission (ASC), began a study in 2003 to examine the timing, magnitude, and duration of summer (June through October) and fall/early winter (September through January) seasonal streamflows of unregulated coastal river basins in Maine and to correlate them to meteorological variables and winter/spring (January through May) seasonal streamflows. This study overlapped the summer seasonal window with the fall/early winter seasonal window to completely bracket the low-streamflow period during July, August, and September between periods of high streamflows in June and October. The ASC is concerned with the impacts of potentially changing meteorological and hydrologic conditions on Atlantic salmon survival. Because winter/spring high streamflows appear to have trended toward earlier dates over the 20th century in coastal Maine, it was hypothesized that the spring/summer recession to low streamflows could have a similar trend toward earlier, and possibly lower, longer lasting, late summer/early fall low streamflows during the 20th century. There were few statistically significant trends in the timing, magnitude, or duration of summer low streamflows for coastal river basins in Maine during the 20th century. The hypothesis that earlier winter/spring high streamflows may result in earlier or lower low streamflows is not supported by the data. No statistically significant trends in the magnitude of total runoff volume during the low-streamflow months of August and September were observed. The magnitude and timing of summer low streamflows correlated with the timing of fall/winter high streamflows and the amount of summer precipitation. The magnitude and timing of summer low streamflows did not correlate with the timing of spring snowmelt runoff. There were few correlations between the magnitude and timing of summer low streamflows and monthly mean surface air temperatures. There were few statistically significant trends in the timing or duration of fall/winter high streamflows for coastal river basins in Maine during the 20th century. The timing of the bulk of fall/winter high streamflows correlated with seasonal precipitation. Earlier fall/winter center-of-volume dates correlated with higher September and October precipitation. In general, little evidence was observed of trends in the magnitude of seasonal runoff volume during fall/winter. The magnitude of fall/winter high streamflows positively correlated with November and December precipitation amounts. There were few correlations between the magnitude and timing of fall/winter high streamflows and monthly mean surface air temperatures.
Analysis of borehole-radar reflection data from Machiasport, Maine, December 2003
In December 2003, the U.S. Geological Survey, in cooperation with the U.S. Army Corps of Engineers, collected borehole-radar reflection logs in two boreholes in Machiasport, Maine. These bedrock boreholes were drilled as part of a hydrogeologic investigation of the area surrounding the former Air Force Radar Tracking Station site on Howard Mountain near Bucks Harbor. The boreholes, MW09 and MW10, are located approximately 50 meters (m) from, and at the site of, respectively, the locations of former buildings where trichloroethylene was used as part of defense-site operations. These areas are thought to be potential source areas for contamination that has been detected in downgradient bedrock wells. This investigation focused on testing borehole-radar methods at this site. Single-hole radar-reflection surveys were used to identify the depth, orientation, and spatial continuity of reflectors that intersect and surround the boreholes. In addition, the methods were used to (1) identify the radial depth of penetration of the radar waves in the electrically resistive bimodal volcanic formation at the site, (2) provide information for locating additional boreholes at the site, and (3) test the potential applications of borehole-radar methods for further aquifer characterization and (or) evaluation of source-area remediation efforts. Borehole-radar reflection logging uses a pair of downhole transmitting and receiving antennas to record the reflected wave amplitude and transit time of high-frequency electromagnetic waves. For this investigation, 60- and 100-megahertz antennas were used. The electromagnetic waves emitted by the transmitter penetrate into the formation surrounding the borehole and are reflected off of a material with different electromagnetic properties, such as a fracture or change in rock type. Single-hole directional radar surveys indicate the bedrock surrounding these boreholes is highly fractured, because several reflectors were identified in the radar-reflection data. There are several steeply dipping reflectors with orientations similar to the fracture patterns observed with borehole imaging techniques and in outcrops. The radar-reflection data showed that the vitrophyre in borehole MW09 was more highly fractured than the underlying gabbroic unit. The velocities of radar waves in the bedrock surrounding the boreholes were determined using single-hole vertical radar profiling. Velocities of 114 and 125 meters per microsecond were used to determine the distance to reflectors, the radial depth of penetration, and the dip of reflectors. The bimodal volcanic units appear to be ideal for radar-wave propagation. For the radar surveys collected at this site, radar reflections were detected up to 40 m into the rock from the borehole. These results indicate that boreholes could conservatively be spaced about 15-20 m apart for hole-to-hole radar methods to be effective for imaging between the boreholes and monitoring remediation. Integrated analysis of drilling and borehole-geophysical logs indicates the vitrophyric formation is more fractured than the more mafic gabbroic units in these boreholes. There does not, however, appear to be a quantifiable difference in the radar-wave penetration in these two rock units.
Lead isotopic compositions of soil and near-surface till profiles from a watershed containing arsenic-enriched groundwater in coastal Maine
Lead isotope compositions of soils and near-surface tills from an area of coastal Maine known to have groundwater with anomalously high arsenic contents were measured in order to determine the source of the lead and, by inference, possible sources of arsenic. Five soil and till sites were selected for detailed chemical and isotopic analysis. To construct profiles of the soil and till horizons, five samples were collected at 10-cm intervals from the surface to the base of each horizon. Total lead and arsenic concentrations and lead isotopic compositions were measured for 48 leaches and bulk residues. The soils and tills are underlain by sulfidic schists of the Penobscot Formation. Several generations of minerals containing arsenic and lead exist in the regional bedrock, including rock-forming silicates (feldspar and micas), sulfide minerals formed during diagenesis (for example, arsenic-rich pyrite), and sulfide and oxide minerals that formed as a result of Silurian metamorphic and igneous events (for example, arsenopyrite, galena, iron-oxides, and arsenic-sulfides). A young group of secondary minerals (for example, iron-hydroxides, arsenic-hydroxides, lead-sulfate, and arsenic-jarosite) formed from recent weathering and pedogenic processes.
Using hydrogeomorphic criteria to classify wetlands on Mt. Desert Island, Maine – Approach, classification system, and examples
A wetland classification system was designed for Mt. Desert Island, Maine, to help categorize the large number of wetlands (over 1,200 mapped units) as an aid to understanding their hydrologic functions. The classification system, developed by the U.S. Geological Survey (USGS), in cooperation with the National Park Service, uses a modified hydrogeomorphic (HGM) approach, and assigns categories based on position in the landscape, soils and surficial geologic setting, and source of water. A dichotomous key was developed to determine a preliminary HGM classification of wetlands on the island. This key is designed for use with USGS topographic maps and 1:24,000 geographic information system (GIS) coverages as an aid to the classification, but may also be used with field data. Hydrologic data collected from a wetland monitoring study were used to determine whether the preliminary classification of individual wetlands using the HGM approach yielded classes that were consistent with actual hydroperiod data. Preliminary HGM classifications of the 20 wetlands in the monitoring study were consistent with the field hydroperiod data. The modified HGM classification approach appears robust, although the method apparently works somewhat better with undisturbed wetlands than with disturbed wetlands. This wetland classification system could be applied to other hydrogeologically similar areas of northern New England.
Extraordinary size and survival of American black duck, Anas rubripes, broods
Two female American black duck, Anas rubripes, were initially observed during June 1982 with 20 Class Ib or 18-22 Class Ia-b ducklings in two wetlands in Hancock County, Cherryfield, Maine. Fifteen of 20 ducklings (75%) in one brood and 16 of 18-22 ducklings (72-89%) in the other brood survived to fledge. These large broods probably resulted from post-hatch brood amalgamation.
Dynamic use of wetlands by black ducks and mallards: Evidence against competitive exclusion
The decline of the American black duck ( Anas rubripes ) has been attributed to competition from mallards ( A. platyrhynchos ) that led to exclusive use of fertile wetlands by mallards. Data from annual breeding waterfowl surveys provide instantaneous, single observations of breeding pairs, which are used to estimate breeding population size and evaluate the condition of habitat. Data from these surveys have been used to document habitat use by black ducks and mallards. We used quiet-observation surveys from elevated platforms to study sympatric black ducks and mallards in northern Maine during the breeding season. Our objectives were to document occupancy of wetlands by breeding black ducks and mallards throughout the day during prenesting and early nesting periods to determine whether 1) wetlands were occupied by only a single species, 2) pairs of the same species occupied wetlands throughout the period, and 3) single observations of short duration adequately determine numbers and species using a wetland. We observed ducks at 5-minute intervals from elevated platforms on wetland margins to determine numbers and species of indicated pairs using each wetland over time. We visited 80% of the wetlands ≥2 times, with mean total time per wetland averaging 267 minutes. For each wetland we determined the most frequently observed grouping of black ducks and mallards from all combinations recorded during all intervals (e.g., 1 black duck [BO] pair during 9 intervals; 2 mallard [MA] pairs and 1 BO pair during 22 intervals; 0 pairs during 3 intervals). A single pair, a lone male, or no ducks were recorded during 34% of the 5-minute intervals. For wetlands with >2 hours of observations ( n =65 ), all but 2 were used by ≥2 different combinations of ducks. On most wetlands, the most frequent grouping was observed during <40% of the intervals. To simulate aerial surveys, we randomly selected 1 5-minute interval for each wetland. On average, the number of indicated pairs recorded during random 5-minute intervals was less than half of the total black duck pairs (2.0 vs. 4.4, P = 0.009 ), total mallard pairs (1.1 vs. 2.6, P= 0.0001), and pairs of both species combined (3.2 vs. 7.0, P= 0.0001) determined for each wetland based on total observations. On wetlands used by both species, random counts detected one or both species 49% of the time. Although 53 of the 65 wetlands observed ≥2 hours were used by both species, random visits detected both species on only 27 wetlands. Our data do not support assertions that the mallard has caused the decline of black ducks through interspecific competition for habitat, or that wetlands are occupied continuously by single pairs that aggressively exclude conspecifics. Our data indicated that single, short-duration visits with disturbance to wetlands are unreliable and inappropriate to document seasonal use of wetlands by breeding black ducks and mallards.
Canada lynx Lynx canadensis habitat and forest succession in northern Maine, USA
The contiguous United States population of Canada lynx Lynx canadensis was listed as threatened in 2000. The long-term viability of lynx populations at the southern edge of their geographic range has been hypothesized to be dependent on old growth forests; however, lynx are a specialist predator on snowshoe hare Lepus americanus, a species associated with early-successional forests. To quantify the effects of succession and forest management on landscape-scale (100 km 2 ) patterns of habitat occupancy by lynx, we compared landscape attributes in northern Maine, USA, where lynx had been detected on snow track surveys to landscape attributes where surveys had been conducted, but lynx tracks had not been detected. Models were constructed a priori and compared using logistic regression and Akaike's Information Criterion (AIC), which quantitatively balances data fit and parsimony. In the models with the lowest (i.e. best) AIC, lynx were more likely to occur in landscapes with much regenerating forest, and less likely to occur in landscapes with much recent clearcut, partial harvest and forested wetland. Lynx were not associated positively or negatively with mature coniferous forest. A probabilistic map of the model indicated a patchy distribution of lynx habitat in northern Maine. According to an additional survey of the study area for lynx tracks during the winter of 2003, the model correctly classified 63.5% of the lynx occurrences and absences. Lynx were more closely associated with young forests than mature forests; however, old-growth forests were functionally absent from the landscape. Lynx habitat could be reduced in northern Maine, given recent trends in forest management practices. Harvest strategies have shifted from clearcutting to partial harvesting. If this trend continues, future landscapes will shift away from extensive regenerating forests and toward landscapes dominated by pole-sized and larger stands. Because Maine presently supports the only verified populations of this federally threatened species in the eastern United States, changes in forest management practices could affect recovery efforts throughout that region.
Estimating monthly, annual, and low 7-day, 10-year streamflows for ungaged rivers in Maine
Regression equations to estimate monthly, annual, and low 7-day, 10-year (7Q10) streamflows were derived for rivers in Maine. The derived regression equations for estimating mean monthly, mean annual, median monthly, median annual, and low 7Q10 streamflows for ungaged rivers in Maine presented in this report supersede those derived in previous studies. Twenty-six U.S. Geological Survey streamflow-gaging stations on unregulated, rural rivers in Maine with 10 years or more of recorded streamflow were used to develop the regression equations. Ordinary least squares (OLS) regression techniques were used to select the explanatory variables (basin and climatic characteristics) that would appear in the final regression equations. OLS regression of all possible subsets was done with 62 explanatory variables for each of 27 response variables. Five explanatory variables were chosen for the final regression equations: drainage basin area, areal fraction of the drainage basin underlain by sand and gravel aquifers, distance from the coast to the drainage basin centroid, mean drainage basin annual precipitation, and mean drainage basin winter precipitation (the sum of mean monthly precipitation for December, January, and February). Generalized least-squares regression techniques were used to derive the final coefficients and measures of uncertainty for the regression equations. The forms of many of the derived regression equations indicate some physical, mechanistic processes. Drainage basin area is the most statistically important explanatory variable and appears in all derived regression equations. Monthly streamflows are related inversely to the distance from the coast to the drainage basin centroid during December, January, February, and March; that is, the closer a river basin is to the coast, the higher monthly streamflows are per unit drainage basin area during the winter. The relation reverses in May when higher streamflows are attributed to basins farther from the coast. These relations are consistent with colder, inland drainage basins storing more water in snowpack during the winter and releasing it in the spring. The monthly streamflows (and low 7Q10) during July, August, September, and October are related positively to areal fraction of the drainage basin underlain by sand and gravel aquifers. In general, sand and gravel aquifers underlying Maine river basins have excellent water-yielding characteristics and can provide water to streams during low-flow conditions in the summer and early fall.
Hydraulic-geometry relations for rivers in coastal and central Maine
Hydraulic-geometry relations (curves) were derived for 15 sites on 12 rivers in coastal and central Maine on the basis of site-specific (at-a-station) hydraulic-geometry relations and hydraulic models. At-a-station hydraulic-geometry curves, expressed as well-established power functions, describe the relations between channel geometry, velocity, and flow at a given point on a river. The derived at-a-station hydraulic-geometry curves indicate that, on average, a given increase in flow at a given river cross section in the study area will be nearly equally conveyed by increases in velocity and channel cross-sectional area. Regional curves describing the bankfull streamflow and associated channel geometry as functions of drainage area were derived for use in stream-channel assessment and restoration projects specific to coastal and central Maine. Regional hydraulic-geometry curves were derived by combining hydraulic-geometry information for 15 river cross sections using bankfull flow as the common reference streamflow. The exponents of the derived regional hydraulic-geometry relations indicate that, in the downstream direction, most of the conveyance of increasing contribution of flow is accommodated by an increase in cross-sectional area?with about 50 percent of the increase in flow accommodated by an increase in channel width, and 32 percent by an increase in depth. The remaining 18 percent is accommodated by an increase in streamflow velocity. On an annual-peak-series basis, results of this study indicate that the occurrence of bankfull streamflow for rivers in Maine is more frequent than the 1.5-year streamflow. On a flow-duration basis, bankfull streamflow for rivers in coastal and central Maine is equaled or exceeded approximately 8.1 percent of the time on mean?or about 30 days a year. Bankfull streamflow is roughly three times that of the mean annual streamflow for the sites investigated in this study. Regional climate, snowmelt hydrology, and glacial geology may play important roles in dictating the magnitude and frequency of occurrence of bankfull streamflows observed for rivers in coastal and central Maine.
Water budget for Lake Auburn, Maine, May 1, 2000 through April 30, 2003
Annual water budgets were developed for Lake Auburn in southwestern Maine for three water-budget years, May 1, 2000 through April 30, 2003. The measured inflow components of the water budget are direct precipitation to the surface of the lake and surface-water inflow. Mean annual inflow (precipitation and surface water) to Lake Auburn during water-budget years 2001-03 was 816 million cubic feet. The measured outflow components of the water budget are evaporation from the surface of the lake, municipal water-supply withdrawals by the Auburn and Lewiston Water Districts, and surface-water outflow. Mean annual outflow (evaporation, withdrawals, and surface water) during water-budget years 2001-03 was 834 million cubic feet. The two largest components of the water budget are the surface-water inflow and municipal withdrawals. Surface-water inflow was about 62 percent of the mean annual inflow budget and municipal withdrawals were about 47 percent of the mean annual outflow budget. Changes in lake storage also were included in the water budget. Dry conditions were present in Maine from 1999 to 2002, with a severe drought occurring in 2001 and 2002. On February 20, 2002, Auburn Water District recorded Lake Auburn?s lowest recorded water level, 256.7 ft, during the period of record from April 1, 1940 to June 20, 2003. The total inflow and outflow volumes, adjusted for changes in lake storage, nearly balance for each water year. Any non-zero remainder volume for a given water-budget year is the residual. The mean annual residual for the three water-budget years is -1.06 million cubic feet (-7.93 million gallons). The changing sign of the computed residual from water-budget year to water-budget year during the 3-year study period indicates that no systematic errors were made in estimating inflows to or outflows from Lake Auburn. Because ground-water flow is unaccounted for in this water budget, any net ground-water flow is assumed to be contained in the residual term.
August median streamflow on ungaged streams in eastern coastal Maine
Methods for estimating August median streamflow were developed for ungaged, unregulated streams in eastern coastal Maine. The methods apply to streams with drainage areas ranging in size from 0.04 to 73.2 square miles and fraction of basin underlain by a sand and gravel aquifer ranging from 0 to 71 percent. The equations were developed with data from three long-term (greater than or equal to 10 years of record) continuous-record streamflow-gaging stations, 23 partial-record streamflow- gaging stations, and 5 short-term (less than 10 years of record) continuous-record streamflow-gaging stations. A mathematical technique for estimating a standard low-flow statistic, August median streamflow, at partial-record streamflow-gaging stations and short-term continuous-record streamflow-gaging stations was applied by relating base-flow measurements at these stations to concurrent daily streamflows at nearby long-term continuous-record streamflow-gaging stations (index stations). Generalized least-squares regression analysis (GLS) was used to relate estimates of August median streamflow at streamflow-gaging stations to basin characteristics at these same stations to develop equations that can be applied to estimate August median streamflow on ungaged streams. GLS accounts for different periods of record at the gaging stations and the cross correlation of concurrent streamflows among gaging stations. Thirty-one stations were used for the final regression equations. Two basin characteristics-drainage area and fraction of basin underlain by a sand and gravel aquifer-are used in the calculated regression equation to estimate August median streamflow for ungaged streams. The equation has an average standard error of prediction from -27 to 38 percent. A one-variable equation uses only drainage area to estimate August median streamflow when less accuracy is acceptable. This equation has an average standard error of prediction from -30 to 43 percent. Model error is larger than sampling error for both equations, indicating that additional or improved estimates of basin characteristics could be important to improved estimates of low-flow statistics. Weighted estimates of August median streamflow at partial- record or continuous-record gaging stations range from 0.003 to 31.0 cubic feet per second or from 0.1 to 0.6 cubic feet per second per square mile. Estimates of August median streamflow on ungaged streams in eastern coastal Maine, within the range of acceptable explanatory variables, range from 0.003 to 45 cubic feet per second or 0.1 to 0.6 cubic feet per second per square mile. Estimates of August median streamflow per square mile of drainage area generally increase as drainage area and fraction of basin underlain by a sand and gravel aquifer increase.