SEARCH · USGS Science
Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
An analysis of historical pesticide data (1972-92) for the White River was conducted as part of the U.S. Geological Survey National Water-Quality Assessment Program. Data on the presence of pesticides in streams, bottom sediments, fish, and ground waters were examined. Results are interpreted with respect to spatial, seasonal, and streamflow effects. Concentrations of water-soluble pesticides reach a peak during the first storm following application and remain elevated for 1 to 2 months. The most herbicide loading to the White River occurs during this time, when about 1 percent of the applied herbicides are transported out of the White River Basin. Bottom sediments and fish were analyzed for lipophilic pesticides. Dieldrin, components of technical chlordane, and DDT-related compounds were the most frequently detected. In areas where pesticide concentrations in sediment were high, concentrations in fish were high, indicating that bottom sediments are probably the primary source of lipophilic pesticides in aquatic biota. Ground- water/surface-water interaction and the presence of pesticides in ground waters were examined. The bedrock karst region had the highest degree of ground-water/surface-water interaction, indicating that the shallow ground water is susceptible to contamination from surface sources. Atrazine was the most frequently detected pesticides in ground waters. All wells where pesticides were detected are in karst or alluvial outwash, indicating that these geomorphic units are highly susceptible to contamination.
In 1991 the Trinity River Basin study unit was among the first 20 study units in which work began under full-scale program implementation of the National Water-Quality Assessment Program. A retrospective assessment was undertaken to review and analyze existing pesticide data and related environmental factors. Population and land-use data indicate human modifications to the landscape and hydrologic system of the study area during the period 1968–91. A variety of crops treated with pesticides were identified, with wheat and cotton accounting for the largest number of acres treated annually (541,250 and 519,870 acres, respectively). Agricultural-use estimates for the later period covered by this report (1988–90) indicate that 105 different pesticides were used and that 24 pesticides accounted for 75 percent of average agricultural use in the study area. Sorghum was treated by the largest number of the 24 mostused pesticides, and cotton was treated by the second largest number of those pesticides. Dimethoate and methyl parathion were the most heavily used of the organophosphate class pesticides. The herbicide 2,4–D was the most heavily used chlorophenoxy pesticide. Carbamate pesticides are used extensively in the study area, with carbaryl, carbofuran, methomyl, and thiodicarb accounting for the majority of the use of this class of pesticide. Miscellaneous pesticides included alachlor, arsenic acid, picloram, and glyphosate, among others. The data indicate that herbicide use generally is proportionally higher in the study area than in the Nation, and that insecticide use in the study area generally is proportionally lower than in the Nation. Eight different agencies collected the waterquality data used in this report. Samples were collected by all agencies at a combined total of 155 surface-water sites and 121 ground-water sites. The sampled media included water, bed sediment, and tissues of fish and other aquatic wildlife. Some 273 samples for analysis of the herbicide 2,4–D were collected as part of the city of Arlington’s data-collection program. The herbicide was detected in 74 percent of the samples, but none exceeded the Maximum Contamination Level for drinking water. Dallas Water Utilities collected pesticide samples during a storm in February 1977. Samples were collected at 17 sites with detections of some pesticides in over 50 percent of the samples. Diazinon was detected in 56 percent of samples and 2,4–D was found in 56 percent of samples. Texas Parks and Wildlife Department collected samples from fish tissue for analyses of organochlorine pesticides from 15 sites in the Dallas-Fort Worth area. Chlordane concentrations in some of the samples exceeded the Food and Drug Administration’s action level of 300 micrograms per kilogram. The Texas Water Commission collected ground-water samples in the study area during 1990 for the major types of pesticides and none were detected. No arsenic was detected in samples from 121 wells in or near the study area. Organochlorine and organophosphate samples were collected beginning in 1974 and ending in 1991. Concentrations of organochlorine pesticides in bed sediment decrease with increasing distance downstream from the Dallas-Fort Worth urban area. Pesticide samples collected by the U.S. Geological Survey indicated significant rank correlation between number of detects of chlordane and the percent of the contributing watershed classified as urban land use. Dieldrin in bed sediment samples, and lindane, diazinon, and malathion, in water samples, also were significantly correlated with urban land use. Chlordane and dieldrin were significantly correlated with distance downstream from the Dallas-Fort Worth urban area. Review of all available data showed that pesticides were detected to a substantial degree in various sample media over the time period covered by this report. The authors were able to locate little pesticide-sample data for ground water or for tributary streams because sampling efforts historically have been concentrated on the mainstem Trinity River.
The U.S. Geological Survey is conducting an assessment of water quality in the Georgia-Florida Coastal Plain study unit as part of the National Water-Quality Assessment Program. An initial activity of the program is to compile and analyze existing water-quality data for nutrients in each study unit. Ground-water quality data were compiled from three data sources, the U.S. Geological Survey, Florida Department of Environmental Protection, and Georgia Geologic Survey. A total of 2,246 samples of ground water nutrient data for nitrogen and phosphorus species were compiled from these three data sources. Estimates of 1990 nitrogen and phosphorus inputs by county in the study area were calculated from livestock manure, fertilizers, septic tanks, and rainfall. Data for nitrate nitrogen concentrations in ground water were available from the greatest number of wells; samples from 1,233 wells were available in the U.S. Geological Survey, 820 wells from the Florida Department of Environmental Protection, and 680 wells from the Georgia Geologic Survey. The maximum contaminant level for nitrate nitrogen in drinking water of 10 milligrams per liter was exceeded in a higher percentage of samples from the U.S. Geological Survey, mostly because this data contained numerous samples near known contamination areas. The maximum contaminant level for nitrate nitrogen was exceeded in 3 percent of samples from Upper Floridan aquifer and 12 percent of samples from surficial aquifer system in U.S. Geological Survey data and less than 1 percent and 2 percent of samples from the Upper Floridan aquifer and surficial aquifer system, respectively, in Florida Department of Environmental Protection data. In Georgia Geologic Survey data, 1 percent of samples had concentrations of nitrate nitrogen exceeding 10 milligrams per liter. Nutrient concentration data were grouped into categories based on land use, hydrogeology (aquifer and confinement of the Upper Floridan aquifer), and land resource provinces (Central Florida Ridge, Coastal Flatwoods and Southern Coastal Plain) for the surficial aquifer system. The highest median nitrate nitrogen concentrations in the U.S. Geological Survey data were 0.4 milligrams per liter in ground-water samples from the unconfined Upper Floridan aquifer in agricultural areas and 9.0 milligrams per liter in samples from the surficial aquifer system in agricultural areas in the Central Florida Ridge. In Florida Department of Environmental Protection data, the highest median nitrate nitrogen concentrations were much lower and did not exceed 0.2 milligrams per liter in either the Upper Floridan aquifer or the surficial aquifer system. In Georgia Geologic Survey data the highest median nitrate nitrogen concentration was 1.4 milligrams per liter in agricultural areas in the Coastal Flatwoods. Highest median concentrations of total nitrogen of 10 milligrams per liter (includes nitrate, ammonia, and organic nitrogen) were in U.S. Geological Survey data in the surficial aquifer system in agricultural areas in the Central Florida Ridge. Median concentrations of ammonia nitrogen, orthophosphate phosphorus, and total phosphorus did not exceed 0.5 milligrams per liter in all categories from the Upper Floridan aquifer or the surficial aquifer system.
Four devices for splitting water samples into representative aliquots are used by the U.S. Geological Survey's Water Resources Division. A thorough evaluation of these devices (14-liter churn, 8-liter churn, plastic cone, and Teflon cone) encompasses a wide variety of concerns, based on both chemical and physical considerations. This report surveys the existing data (as of April 1994) on cleaning efficiency and splitting capability of these devices and presents the data in a systematic framework for evaluation. From the existing data, some of these concerns are adequately or partially addressed, but the majority of concerns could not be addressed because of the lack of data. In general, the existing cleaning and transport protocols are adequate at the milligram per liter level, but the adequacy is largely unknown for trace elements and organic chemicals at lower concen- trations. The existing data indicate that better results are obtained when the splitters are cleaned in the laboratory rather than in the field. Two conclusions that can be reached on the splitting capability of solids are that more work must be done with all four devices to characterize and quantify their limitations and range of usefulness, and that the 14-liter churn (and by association, the 8-liter churn) is not useful in obtaining representative splits of sand-sized particles.
In 1906, the U.S. Geological Survey (USGS) began operating a network of streamflow-gaging stations in Alaska. The primary purpose of the streamflow- gaging network has been to provide peak flow, average flow, and low-flow characteristics to a variety of users. In 1993, the USGS began a study to evaluate the current network of 78 stations. The objectives of this study were to determine the adequacy of the existing network in predicting selected regional flow characteristics and to determine if providing additional streamflow-gaging stations could improve the network's ability to predict these characteristics. Alaska was divided into six distinct hydrologic regions: Arctic, Northwest, Southcentral, Southeast, Southwest, and Yukon. For each region, historical and current streamflow data were compiled. In Arctic, Northwest, and Southwest Alaska, insufficient data were available to develop regional regression equations. In these areas, proposed locations of streamflow-gaging stations were selected by using clustering techniques to define similar areas within a region and by spatial visual analysis using the precipitation, physiographic, and hydrologic unit maps of Alaska. Sufficient data existed in Southcentral and Southeast Alaska to use generalized least squares (GLS) procedures to develop regional regression equations to estimate the 50-year peak flow, annual average flow, and a low-flow statistic. GLS procedures were also used for Yukon Alaska but the results should be used with caution because the data do not have an adequate spatial distribution. Network analysis procedures were used for the Southcentral, Southeast, and Yukon regions. Network analysis indicates the reduction in the sampling error of the regional regression equation that can be obtained given different scenarios. For Alaska, a 10-year planning period was used. One scenario showed the results of continuing the current network with no additional gaging stations and another scenario showed the results of adding gaging stations to the network. With the exception of the annual average discharge equation for Southeast Alaska, by adding gaging stations in all three regions, the sampling error was reduced to a greater extent than by not adding gaging stations. The proposed streamflow-gaging network for Alaska consists of 308 gaging stations, of which 32 are designated as index stations. If the proposed network can not be implemented in its entirety, then a lesser cost alternative would be to establish the index stations and to implement the network for a particular region.
Historical pesticide data from 1970-90 were compiled for 140 surface-water, 92 ground-water, 55 streambed-sediment, and 120 biological-tissue sampling sites within the Ozark Plateaus National Water-Quality Assessment Program study unit. Surface-water, bed-sediment, and biological-tissue sites have drainage basins predominantly in the Springfield and Salem Plateaus; ground-water sites are predominantly located in the Osage Plains and Mississippi Alluvial Plain. Many sites were sampled only once or twice during this period. A large percentage of the samples were collected in the mid-1970's and early 1980's for surface water, 1990 for ground water, the late 1980's for surface water, 1990 for ground water, the late 1980's for bed sediment, and the early 1980's for biological tissue. Pesticide use was approximately 4.2 million pounds per year of active ingredients from 1982-85 in the study unit and was generally greatest in the Springfield and Salem Plateaus pasturelands and in the Osage Plains and Mississippi Alluvial Plain cropland areas. The most frequently applied pesticide in the study unit was 2,4-D. Alachlor was the second most applied pesticide. Corn, pasture, rice, sorghum, and soybeans received approximately 90 percent of the pesticides applied within the study unit. The highest pesticide application rate per acre occurred on these crops in the Osage Plains and Mississippi Alluvial Plain. Pastureland was the predominant crop type in 50 of the 94 counties in the study unit. Toxaphene, the pesticide having the most number of detections in surface water, was found in 17 of 866 samples from 5 of 112 sites. Concentrations ranged from 0.1 to 6.0 micrograms per liter. Six other pesticides or pesticide metabolites were detected in 12 or more surface-water samples: DDE, dieldrin, DDT, aldrin, 2,4-D, and lindane. The maximum concentration for these pesticides was less than 1.0 micrograms per liter. Atrazine, the pesticide having the most number of detections in ground water, was found in 15 of 95 samples from 15 of 79 wells with concentrations ranging from 0.1 to 8.2 micrograms per liter. Metolachlor, alachlor, and prometon were detected more than once with maximum concentrations less than 1.0 micrograms per liter, except for prometon (2.4 micrograms per liter). Chlordane was the pesticide having the most number of detections in bed sediment and biological tissue. Chlordane was detected in 12 of 73 samples from 10 of 45 bed-sediment sites with concentrations ranging from 2.0 to 240 micrograms per kilogram. In biological tissue, chlordane was found in 93 of 151 samples from 39 of 53 sites with concentrations ranging from 0.009 to 8.6 milligrams per kilogram. Other pesticides or pesticide metabolites detected more than once in bed sediment include DDT, DDD, p,p'-DDE, DDE, and hexachlorobenzene and in biological tissue include DDT, p,p'-DDE, and hexachlorobenzene. Quality criteria or standards have been established for 15 of the pesticides detected in the study unit. For surface-water samples, the drinking water maximum contaminant level for alachlor was exceeded in one sample from one site in 1982. For ground-water samples, the drinking water maximum contaminant level for atrazine was exceeded in four samples from four wells in 1990. For biological-tissue samples collected during the years 1982-89, the fish tissue action levels for chlordane (19 sites; 26 samples), heptachlor epoxide (3 sites; 3 samples), p,p'-DDE (2 sites; 2 samples), dieldrin (2 sites, 2 samples), and mirex (1 site; 1 sample) were exceeded. For bed-sediment samples, quality criteria or standards were not exceeded for any pesticide. Pesticides do not pose any widespread or persistent problems in the study unit, based on the limited number of samples that exceeded quality criteria and standards.
In 1986, Congress appropriated funds for the U.S. Geological Survey (USGS) to develop the pilot phase of the National Water-Quality Assessment (NAWQA) Program. The long-term goals of this program are to (1) provide a nationally consistent description of current water-quality conditions for a large part of the Nation's water resources; (2) define. long-term trends (or lack of trends) in water quality; and (3) identify, describe, and explain, as possible, the major factors that affect the observed water-quality conditions and trends. To fulfill the goals of the NAWQA program, the USGS plans to examine 60 areas (study units) across the United States during full implementation of the program. In 1991, the NAWQA program went into full implementation with the intensive investigation of 20 of these study units; one of these study units is the Western Lake Michigan Drainages (WMIC) study unit.
This report summarizes water-quality conditions, issues of concern, and management efforts underway in southern Florida. The report is designed to provide a conceptual framework for the Southern Florida National Water Quality Assessment (NAWQA) study that began in 1994. The report makes reference to the most important water-quality literature pertaining to southern Florida, to water-quality studies that are underway or planned, and to topics which are of high priority in the study unit. These topics include: the availability and suit ability of water for competing demands; nutrient enrichment of the Everglades; transport, degradation, and effects of pesticides; and the sources and cycling of mercury in the ecosystem. The report also includes a retrospective analysis and conceptual presentation of nutrient loading, which is a high priority for the national NAWQA Program and for regional water-quality managers. Nutrient contributions from point and nonpoint sources are estimated for nine basins in the study area and are discussed in relation to land use. Fertilizer is the dominant source of phosphorus in eight basins and the dominant source of nitrogen in at least five basins. Atmospheric sources of nitrogen contribute more than 20 percent of the total nitrogen input to all basins and are the dominant source of nitrogen input to Lake Okeechobee and the Everglades. Nutrient loads are also estimated in selected canal and river outflows in southern Florida to provide a spatial overview of the magnitude of nutrient loading to coastal waters. Annual phosphorus loads from the Peace River are the highest in the study unit; annual phosphorus loads from the Caloosahatchee River and the major Palm Beach canals are also high, compared to other parts of southern Florida. Estimated annual loads of phosphorus from parts of the Big Cypress Basin and the S-12 water-control structures of the Tamiami Canal are low compared with estimated phosphorus loads in outflows in the northern part of the study unit. Annual nitrogen loads in southern Florida were highest in outflows from the Caloosahatchee River Basin and the major Palm Beach canals. Nitrogen loads in outflows from parts of the Big Cypress subbasin were lower than those estimated to the north.
Water-quality data from 23 surfacewater-quality monitoring sites operated by the Indiana Department of Environmental Management and streamflow data from 11 U.S. Geological Survey streamflow-gaging stations in the White River Basin were analyzed to determine recent (1981 90 water years) water-quality conditions, trends, and river loads for ammonia, nitrate, total nitrogen, and total phosphorus. The White River Basin drains 11,349 square miles of central and south-central Indiana and is divided into two nearly equal subbasins the East Fork White River and the White River upstream from its confluence with the East Fork (called the "west fork" of the White River by the State's water-management agencies). Nutrient concentrations generally were higher in the more urbanized west fork than in the more rural east fork because of the much larger volumes of treated municipal sewage, combined-sewer overflows, and urban runoff discharged to the west fork. Concentrations of nutrients, especially ammonia and total phosphorus, were higher downstream from Muncie, Anderson, and Indianapolis than they were upstream from these cities. Nutrient concentrations decreased downstream from Indianapolis in the White River and in the downstream reach of the East Fork White River because of dilution, nitrification, adsorption to stream-bottom sediments, and uptake by aquatic vegetation. Seasonal variations in nutrient concentrations and the relations of nutrient concentrations to streamflow depended on the relative contributions of point and nonpoint sources of the nutrients. Total phosphorus increased with increasing streamflow at monitoring sites on the east fork but decreased with increasing streamflow at sites on the west fork. Increasing concentrations of phosphorus with increasing streamflow were consistent with nonpoint sources of phosphorus that wash off land surfaces, whereas decreasing concentrations of phosphorus with increasing streamflow were consistent with dilution of point sources of phosphorus. Median concentrations of total phosphorus were highest during summer and fall downstream from urban areas on the White River because streamflows that dilute point sources of phosphorus are lowest during summer and fall. Median concentrations of ammonia in the White River were highest in winter because of reduced biological uptake and nitrification of ammonia during cold temperatures.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
As part of the Northern Rockies Intermontane Basins study of the National Water-Quality Assessment Program, data collected between 1974 and 1996 were compiled to describe contaminants in tissue of riverine species. Tissue-contaminant data from 11 monitoring programs and studies representing 28 sites in the study area were summarized. Tissue-contaminant data for most streams generally were lacking. Many studies have focused on and around mining-affected areas on the Clark Fork and Coeur d'Alene Rivers and their major tributaries. DDT and PCBs and their metabolites and congeners were the synthetic organic contaminants most commonly detected in fish tissue. Fish collected from the Spokane River in Washington contained elevated concentrations of PCB arochlors, some of which exceeded guidelines for the protection of human health and predatory wildlife. Tissue samples of fish from the Flathead River watershed contained higher-than-expected concentrations of PCBs, which might have resulted from atmospheric transport. Trace element concentrations in fish and macroinvertebrates collected in and around mining areas were elevated compared with background concentrations. Some cadmium, copper, lead, and mercury concentrations in fish tissue were elevated compared with results from other studies, and some exceeded guidelines. Macroinvertebrates from the Coeur d'Alene River contained higher concentrations of cadmium, lead, and zinc than did macroinvertebrates from other river systems in mining-affected areas. A few sportfish fillet samples, most from the Spokane River in Washington, were collected to assess human health risk. Concentrations of PCBs in these fillets exceeded screening values for the protection of human health. At present, there is no coordinated, long-term fish tissue monitoring program for rivers in the study area, even though contaminants are present in fish at levels considered a threat to human health. Development of a coordinated, centralized national data base for contaminants in fish tissue is needed. The National Water-Quality Assessment Program can provide a framework for other agencies to evaluate tissue contaminants in the Northern Rockies Intermontane Basins study area. As of 1996, there are no fish consumption advisories or fishing restrictions as a result of elevated contaminants on any rivers within the study area.
Although sea turtles have received substantial focus worldwide, research on the immature life stages is still relatively limited. The latter is of particular importance, given that a large proportion of sea turtle populations comprises immature individuals. We set out to identify knowledge gaps and identify the main barriers hindering research in this field. We analyzed the perceptions of sea turtle experts through an online survey which gathered their opinions on the current state of affairs on immature sea turtle research, including species and regions in need of further study, priority research questions, and barriers that have interfered with the advancement of research. Our gap analysis indicates that studies on immature leatherback Dermochelys coriacea and hawksbill Eretmochelys imbricata turtles are lacking, as are studies on all species based in the Indian, South Pacific, and South Atlantic Oceans. Experts also perceived that studies in population ecology, namely on survivorship and demography, and habitat use/behavior, are needed to advance the state of knowledge on immature sea turtles. Our survey findings indicate the need for more inter-disciplinary research, collaborative efforts (e.g. data-sharing, joint field activities), and improved communication among researchers, funding bodies, stakeholders, and decision-makers.
Sea otters Enhydra lutris are vital keystone predators throughout the North Pacific that were nearly extirpated during the maritime fur trade. Recovery of southern sea otters E. l. nereis has proceeded slowly, with much of their historical range remaining unoccupied, resulting in reduced ecosystem functioning. Numerous studies have used foraging metrics to assess the population status of southern sea otters throughout their current range, but little is known about the northern range extent, where a stall in expansion has limited recovery. Thus, we collected census and foraging data of sea otters at Año Nuevo State Park, California, from 2019 to 2021 to determine sea otter abundance, diet composition, diet diversity, and average energy intake rate at the northern range edge. We then assessed regional population status by comparing values from Año Nuevo with previously collected data from other locations in California, including high-density, range center sites and low-density, range periphery sites. We found that sea otter density at Año Nuevo was greater than surrounding areas at the northern range periphery, and the average (±95% CI) energy intake (9.51 ± 0.91 kcal min -1 ) more closely resembled values observed at high-density sites. Further, dietary diversity (using the Shannon-Wiener index, H ) was intermediate between previously studied high- and low-density populations ( H = 1.81), with crabs making up the largest proportion of the diet (~56%). Overall, this study highlights possible effects of occupation time and range stagnation, identifies unique aspects of the prey resource base at Año Nuevo, and provides insight into the ongoing lack of northern range expansion.
Spatial demographic models can help guide monitoring and management activities targeting at-risk species, even in cases where baseline data are lacking. Here, we provide an example of how site-specific changes in land use and anthropogenic stressors can be incorporated into a spatial demographic model to investigate effects on population dynamics of Golden Eagles ( Aquila chrysaetos ). Our study focused on a population of Golden Eagles exposed to risks associated with rapid increases in renewable energy development in southern California, U.S.A. We developed a spatially explicit, individual-based simulation model that integrated empirical data on demography of Golden Eagles with spatial data on the arrangement of nesting habitats, prey resources, and planned renewable energy development sites. Our model permitted simulated eagles of different stage-classes to disperse, establish home ranges, acquire prey resources, prospect for breeding sites, and reproduce. The distribution of nesting habitats, prey resources, and threats within each individual's home range influenced movement, reproduction, and survival. We used our model to explore potential effects of alternative disturbance scenarios, and proposed conservation strategies, on the future distribution and abundance of Golden Eagles in the study region. Results from our simulations suggest that probable increases in mortality associated with renewable energy infrastructure (e.g., collisions with wind turbines and vehicles, electrocution on power poles) could have negative consequences for population trajectories, but that site-specific conservation actions could reduce the magnitude of negative effects. Our study demonstrates the use of a flexible and expandable modeling framework to incorporate spatially dependent processes when determining relative effects of proposed management options to Golden Eagles and their habitats.
Uncovering the historical and contemporary processes shaping rare species with complex distributions is of growing importance due to threats such as habitat destruction and climate change. Species restricted to specialized, patchy habitat may persist by virtue of life history characteristics facilitating ongoing gene flow and dispersal, but they could also reflect the remnants of formerly widespread, suitable habitat that existed during past climate regimes. If formerly widespread species did not rely upon traits facilitating high dispersibility to persist, contemporary populations could be at high risk of extirpation or extinction. Fortunately, genomic investigations provide an opportunity to illuminate such alternative scenarios while simultaneously offering guidance for future management interventions. Herein, we test the role of these mechanisms in shaping patterns of genomic diversity and differentiation across a highly restricted and rare ecosystem: desert hanging gardens. We focus on Carex specuicola (Cyperaceae), a hanging garden obligate narrowly distributed in the Four Corners region of the southwestern United States that is listed as Threatened under the United States Endangered Species Act. Population structure and diversity analyses reveal that hanging garden populations are shaped by strong genetic drift, but that individuals in gardens are occasionally more closely related to individuals at other gardens than to individuals within the same garden. Similarly, gardens separated by long geographic distances may contain individuals that are more closely related compared to individuals in gardens separated by short geographic distances. Demographic modeling supports historical gene flow between some contemporary garden pairs, which is corroborated by low estimates of inbreeding coefficients and recent divergence times. As such, multiple lines of evidence support dispersal and gene flow across C. specuicola populations at both small and large spatial scales, indicating that even if C. specuicola was formerly more widespread, it may be well suited to persist in hanging gardens so long as suitable habitat remains available. Analyses like those demonstrated herein may be broadly applicable for understanding the short- and long-term evolutionary processes influencing rare species, and especially those having complex distributions across heterogeneous landscapes.