DDDaniel H. Doctor

Daniel H. Doctor

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Daniel H. Doctor has a total of 52 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Exploring the Influence of Input Feature Space on CNN‐Based Geomorphic Feature Extraction From Digital Terrain Data

    by Aaron E. Maxwell, William E. Odom, Charles M. Shobe, Daniel H. Doctor, Michelle S. Bester, Tobi Ore

    Many studies of Earth surface processes and landscape evolution rely on having accurate and extensive data sets of surficial geologic units and landforms. Automated extraction of geomorphic features using deep learning provides an objective way to consistently map landforms over large spatial extents. However, there is no consensus on the optimal input feature space for such analyses. We explore the impact of input feature space for extracting geomorphic features from land surface parameters (LSPs) derived from digital terrain models (DTMs) using convolutional neural network (CNN)‐based semantic segmentation deep learning. We compare four input feature space configurations: (a) a three‐layer composite consisting of a topographic position index (TPI) calculated using a 50 m radius circular window, square root of topographic slope, and TPI calculated using an annulus with a 2 m inner radius and 10 m outer radius, (b) a single illuminating position hillshade, (c) a multidirectional hillshade, and (d) a slopeshade. We test each feature space input using three deep learning algorithms and four use cases: two with natural features and two with anthropogenic features. The three‐layer composite generally provided lower overall losses for the training samples, a higher F1‐score for the withheld validation data, and better performance for generalizing to withheld testing data from a new geographic extent. Results suggest that CNN‐based deep learning for mapping geomorphic features or landforms from LSPs is sensitive to input feature space. Given the large number of LSPs that can be derived from DTM data and the variety of geomorphic mapping tasks that can be undertaken using CNN‐based methods, we argue that additional research focused on feature space considerations is needed and suggest future research directions. We also suggest that the three‐layer composite implemented here can offer better performance in comparison to using hillshades or other common terrain visualization surfaces and is, thus, worth considering for different mapping and feature extraction tasks.


  • Open Access

    Hydrogeology, karst, and groundwater availability of Monroe County, West Virginia

    by Mark D. Kozar, Daniel H. Doctor, William K. Jones, Nathan Chien, Cheyenne E. Cox, Randall C. Orndorff, David J. Weary, Mitchell R. Weaver, Mitchell A. McAdoo, Mercer Parker

    Monroe County is in southeastern West Virginia, encompassing an area of 474 square miles. The area consists of karst and siliciclastic aquifers of Ordovician, Silurian, Devonian, and Mississippian age and is in parts of two physiographic provinces: the Valley and Ridge Province to the east of Peters Mountain, and the Appalachian Plateau Province to the west of Peters Mountain. This study was developed in response to inquiries from the Monroe County Commission requesting assessment of the water resources of the county to better understand the quantity of the county’s groundwater resources, for both current [2023] and future demand, and to provide information to support protection and management of the county’s valuable groundwater resources.Various products were developed for this study that provide knowledge with respect to water availability and contamination susceptibility of the karst aquifers within the county. U.S. Geological Survey (USGS) geologists conducted extensive geologic mapping in support of the project, producing (1) a countywide bedrock geologic map, (2) a countywide hydrogeologic map, and (3) a light detection and ranging (lidar)-derived countywide digital elevation model and associated sinkhole map. A significant part of this work was to map in detail the Greenbrier Group at the formation level, which prior to this study had only partially been completed. The report also includes (4) a description of the lithologic units identified as part of the geologic mapping process.U.S. Geological Survey hydrologists completed several additional products for the hydrology part of the effort, including development of (1) a countywide potentiometric surface (water-table) map, (2) a countywide base-flow stream assessment, (3) countywide water-budget estimates, (4) well log surveys for 15 wells to better understand subsurface controls on groundwater flow within the study area, (5) two groundwater tracer tests to better refine the groundwater divide from the northern and southern parts of the karst aquifer in Monroe County; and finally, based on all available data collected for the study including the potentiometric surface map, geologic map, current [2023] and legacy fluorometric groundwater tracer tests, and base-flow stream assessments, (6) groundwater-basin delineations were reassessed for principal groundwater basins within the Greenbrier aquifer.In Monroe County, four principal hydrogeologic settings produce large yields of water for residential, agricultural, and other uses. The most relied upon water-bearing zone with respect to current [2023] public water supply is from springs along Peters Mountain. These springs are derived from intervals of fractured sandstone and resultant alluvial deposits. Groundwater flows downslope through these permeable alluvial deposits and discharges at the contact with less permeable strata, such as the Reedsville Shale. The second most relied upon water-bearing zone in Monroe County is within the karstic Greenbrier Group aquifer, in which the basal Hillsdale Limestone overlies the less permeable Maccrady Shale. This geologic contact between the Hillsdale Limestone and Maccrady Shale is not only targeted as a source of water for agricultural supply but also is targeted as a source of water for residential supply. The third most relied upon water-bearing zone is composed of shallow perched aquifers within the Greenbrier Group. The discontinuous nature of these perched aquifers makes mapping their extent impossible, but they are related to permeable geologic strata, such as karstified limestones with solutionally enhanced permeability that overlies less permeable shale or chert bedrock. During geologic mapping of the county, several of these perched aquifers were documented in the Pickaway, Union, and Alderson Limestones. A fourth zone consists of springs from Ordovician carbonates at the base of Peters Mountain, which are influenced by sinking streams as well as upwelling along faults. In terms of water quantity, the most sustainable springs are those having deeper-sourced flows.Public supplies are a principal source of water used for residential and commercial supply in the region, accounting for 0.49 million gallons per day (Mgal/d) of fresh-water withdrawals (0.14 Mgal/d of groundwater and 0.35 Mgal/d of surface water) for residential and commercial use and serving 6,645 individuals (49.2 percent of the population). An estimated 6,861 people, (50.8 percent of the population) primarily rely on private wells or other unregulated sources, such as springs, and withdraw 0.55 Mgal/d of groundwater for their residential use. Public water supply in the region is primarily (71.4 percent) derived from springs and augmented by stream withdrawals (backup sources mainly during low-flow periods), with the remaining portion (28.6 percent) derived from groundwater withdrawals from wells. For rural residents, however, 100 percent of their withdrawals are derived from groundwater (wells or springs).


  • Open Access

    A multiscale approach for monitoring groundwater discharge to headwater streams by the U.S. Geological Survey Next Generation Water Observing System Program—An example from the Neversink Reservoir watershed, New York

    by Martin A. Briggs, Christopher L. Gazoorian, Daniel H. Doctor, Douglas A. Burns

    Groundwater-stream connectivity across mountain watersheds is critical for supporting streamflow during dry times and keeping streams cool during warm times, yet U.S. Geological Survey (USGS) stream measurements are often sparse in headwaters. Starting in 2019, the USGS Next Generation Water Observing System Program developed a multiscale methods and technology testbed approach to monitoring groundwater discharge to streams in the Neversink Reservoir watershed in the Catskill Mountains of New York. Groundwater discharge dynamics are complex across space and time because of geographic variability, topography, and preferential groundwater flow patterns, and the monitoring of discharge processes necessitates an innovative approach that includes emerging water tracing methods and enhanced local geologic mapping. This fact sheet describes the multiscale monitoring approach applied in the Neversink Reservoir watershed and specifically how the varied data types are complimentary in understanding groundwater-stream connectivity, with elements transferable to other mountain watersheds.


  • Geologic maps of the Stephenson and Winchester quadrangles, Frederick and Clarke Counties, Virginia, and Inwood and White Hall quadrangles, Berkeley and Jefferson Counties, West Virginia

    by David J. Weary, Daniel H. Doctor, Randall C. Orndorff

    The study area consists of four contiguous 7.5-minute quadrangles and is located in Frederick and Clarke Counties, Virginia, and Berkeley and Jefferson Counties, West Virginia. The individual quadrangles are Stephenson, Winchester, Inwood, and White Hall. The study area lies within the Great Valley subprovince of the Valley and Ridge physiographic province where about 23,000 feet (ft) (7,000 meters [m]) of Middle Cambrian to Upper Devonian sedimentary rocks are exposed and are overlain by Holocene and older surficial deposits. The area of the four maps is divided into three geologic regions based on the following primary lithologies: (1) Cambrian and Ordovician carbonate rocks of the Great Valley southeast of the North Mountain fault zone and east and west of the core of the Massanutten synclinorium; (2) shale, graywacke, and calcareous shale of the Ordovician Martinsburg Formation of the Great Valley and Massanutten synclinorium; and (3) Ordovician through Devonian clastic rocks and minor limestone and dolostone northwest of and within the North Mountain fault zone. Rocks of all three regions were folded and faulted during the late Paleozoic Alleghanian orogeny (roughly 320 to 250 million years before present). The terrain of this portion of the Great Valley generally is gently to moderately rolling with low local relief with elevations in the study area ranging from about 425 ft (130 m) where Opequon Creek flows out of the eastern edge of the Inwood quadrangle to about 950 ft (290 m) adjacent to Round Hill in the western part of the Winchester quadrangle. Sinkholes and other karst features are common in the carbonate rocks of the Great Valley. The area west of the North Mountain fault zone is underlain by middle Paleozoic strata and consists of a series of ridges and valleys with higher local relief, with elevations ranging from about 785 ft (240 m) in the vicinity of Green Spring in the central part of the White Hall quadrangle to about 1,435 ft (437 m) at the summit of North Mountain in the northeastern part of the White Hall quadrangle.

  • Karst Geology of the Upper Midwest, USA

    by Daniel H. Doctor, E. Calvin Alexander


  • Open Access

    Geologic map of the Hayfield quadrangle, Frederick County, Virginia

    by Daniel H. Doctor, Ronald A. Parker

    The Hayfield 7.5-minute quadrangle is located within the Valley and Ridge physiographic province of northern Virginia. The quadrangle includes the topographical lowland area of the northern Great Valley to the southeast, the narrow ridge of Little North Mountain along the western edge of the Great Valley, and the broad region of elongated valleys and ridges east of Little North Mountain. The most prominent physiographic feature within the quadrangle is Great North Mountain, which extends across the northwestern portion of the quadrangle. All exposed bedrock units are Paleozoic sedimentary rocks ranging from Middle Cambrian to Late Devonian, approximately 513 to 359 Ma. The clastic and carbonate sedimentary strata in the quadrangle reflect nearshore and offshore marine and deltaic depositional environments. The deposits indicate minor sea level transgression and regression cycles along a passive continental margin during the Late Cambrian to Middle Ordovician, and major sea level changes resulting from tectonic uplift during the Late Ordovician Taconian orogeny and the Middle to Late Devonian Acadian orogeny. Compressive forces caused by the continental collision during the Paleozoic Alleghanian orogeny resulted in folding and faulting of the sedimentary rock strata, with northwestward tectonic transport. The North Mountain fault zone, spanning across the southeastern part of the quadrangle, forms the western border of the Great Valley in northern Virginia and is a series of northeast-trending thrust faults with multiple splays that separate the Silurian and Devonian shales, siltstones, and sandstones from the Cambrian and Ordovician carbonate rocks and shales. Topographic ridges in the quadrangle are primarily held up by sandstones and orthoquartzites that are relatively resistant to erosion. Surficial materials include unconsolidated alluvium, colluvium, debris flow, and terrace deposits that are assumed to be of Quaternary age. Alluvium was mapped along the larger streams; locally, some low alluvial terraces exist but have not been broken out within this unit. Debris flow deposits were mapped where recognized in the lidar-derived topographic imagery on the flanks of Great North Mountain. Colluvium (not mapped separately) covers most of the steeper slopes and fills the bottoms of many of the mountain hollows, and is composed mainly of sandstone boulders and cobbles, and fragments of chert.


  • Karst without Boundaries

    by Daniel H. Doctor