Geographic Names Information System Research Articles

Geomorphometric analysis of the Summit and Ridge classes of the Geographic Names Information System

Geomorphometric analysis of the Summit and Ridge classes of the Geographic Names Information System

Principles of Toponyms (Place Names) Classifications

Toponyms (place names) represent language units denoting elements of topographic environment. The relevance of the study is caused by the need to develop the theory of toponymy and to systematize all the existing toponymic classifications.Place names have a complicated (multidimensional) nature, which should be taken into account while classifying them. This research is an attempt to highlight various principles according to which it is possible to group the place names. The research is based on 15000 British and 17000 American toponyms, received from toponymic dictionaries, reference books and electronic toponymic search systems GNIS (Geographic Names Information System) and Get-a-map. Toponyms can be classified according to the following principles: 1) parametric characteristics of an object, 2) ontological characteristics of an object, 3) type of toponymic basis, 4) etymological characteristics of place names, 5) motivational characteristics of place names, 6) chronological characteristics of place names, 7) structural characteristics of toponyms, 8) toponymic polysemy, 9) degree of toponymic nomination, 10) variety of toponymic nomination, 11) localization of an object. It is obvious that it is impossible to create a unified classification of place names, which would reflect the entire multidimensional nature of the toponymic vocabulary. Classifying toponymic nominations, one must consider the multifaceted (intra- and extra-linguistic) principles of allocation of toponymic units.

GeoNat v1.0: A dataset for natural feature mapping with artificial intelligence and supervised learning

AbstractMachine learning allows “the machine” to deduce the complex and sometimes unrecognized rules governing spatial systems, particularly topographic mapping, by exposing it to the end product. Often, the obstacle to this approach is the acquisition of many good and labeled training examples of the desired result. Such is the case with most types of natural features. To address such limitations, this research introduces GeoNat v1.0, a natural feature dataset, used to support artificial intelligence‐based mapping and automated detection of natural features under a supervised learning paradigm. The dataset was created by randomly selecting points from the U.S. Geological Survey’s Geographic Names Information System and includes approximately 200 examples each of 10 classes of natural features. Resulting data were tested in an object‐detection problem using a region‐based convolutional neural network. The object‐detection tests resulted in a 62% mean average precision as baseline results. Major challenges in developing training data in the geospatial domain, such as scale and geographical representativeness, are addressed in this article. We hope that the resulting dataset will be useful for a variety of applications and shed light on training data collection and labeling in the geospatial artificial intelligence domain.

What Is in a “Lake” Name? That Which We Call a Lake by Any Other Name

AbstractGiven how important lakes are to people, it might seem safe to assume that careful thought has been put into the naming of lakes, and that lake names reflect the high societal value people place on lakes. We examined these assumptions by analyzing the official names in the U.S. Geographic Names Information System for the 479,950 lakes ≥ 1 ha in the conterminous U.S. We found that 83% of lakes were unnamed and most of these were small lakes with 80% of unnamed lakes being smaller than 4 ha. Based on the 83,115 named lakes, we found that lake names reflect peoples' everyday lives, that lakes can inspire creativity (although the most common lake name is “Mud”), that Native American and indigenous languages have played a role in lake naming, and that there are regional differences in lake names. Unfortunately, we also found that derogatory terms were part of some lake names. We advocate for thoughtful and inclusive official naming of the 400,000 unnamed lakes in the U.S., as well as renaming of the lakes with derogatory terms to help focus attention on the importance of lakes to local communities and nations.

Modelling and Mapping Elusive Locations of Historic Water-Powered Grist Mills

Abstract. Grist mills are structures in which stone wheels are used to grind grain (e.g. corn, wheat) into a powder-like form for human consumption. Circular stone wheels provided the pressure to grind the grain and separate seed components. In most grist mills the energy for turning the heavy stones was derived from water power. The most visible part of some these mills was the externally mounted water wheel. However, other common configurations of the water ‘turbine’ internal to the structure were used (Figure 1). While grist mills existed in Europe at least as early as the 11th century such mills were not present in the United States until the last 1600s. No comprehensive census has been conducted for grist mills in the U.S. although an estimate of all water powered mills in eastern U.S. was 65,000 in 1840. Many of these water powered mills in the census estimate were for textile, sawmills, manufacturing, and other non-grain-grinding applications.Most often, the mill site included a small impoundment for creating a reliable source of water during low-flow stream conditions. These ponds were created in the late 1700s and especially in the 1800s as a reliable water source for turning grist or saw mills. By the middle of the 20th century the grist mills had all but vanished. Except for a few historic relics and conversion to touristic sites, the mill infrastructure has disappeared on the landscape while the mill ponds typically remain, serving other purposes, such as fishing or hunting lakes, or merely aesthetic environments.The local ecosystems around the mills sites have been artificially modified by the presence of these stream impoundments providing a new environment for fish, mammals, avian species, and of course, humans. While numerous positive ecosystem values may be observed from their continued presence, the risk of dam failure to both downstream systems and humans is substantial. In fact, such a catastrophic series of dam failures (51 in South Carolina) occurred in October of 2015 from a heavy rainfall event. In other instances the impoundments now store toxic sediment originating from sources farther upstream such as mining, military or industrial processes.Where are these historic mill sites and mill ponds? No inventory of mills or their ponds exist in South Carolina (and only for a few northeastern states). In this research we developed a systematic approach for identifying and mapping historic mill sites and ponds using a geohistorical framework and applied to South Carolina. The approach relied on 1) numerous geospatial sources, and 2) an analytical model of confidence mapping for predicting sites. The developed approach resulted in a new database of definitive and likely historic mill sites and mill pond locations throughout South Carolina. The sources of geospatial information was largely cartographic in nature including the following:USGS Historic Topographic Map SeriesRobert Mills 1820 atlas covering South CarolinaHistoric aerial photographyNational Hydrography Data (NHD)National Wetlands Inventory (NWI)National Inventory of DamsSC Department of Health and Environmental Control for Regulated DamsToponyms from the Geographic Names Information System (GNIS)U.S. Census TIGER street dataThe design for a predictive model (Figure 2) using diverse historic data with various scales and reliability is problematic. Most of the grist mills have disappeared from the landscape and do not exist in contemporary geospatial databases. Historic maps might include toponyms for the sites or ponds but their map projection, scale, and subsequent distortions in geometry are problematic. In addition, some of the sources for data are derived from a similar source and thus, are not statistically independent. The design of the analytical model for mapping confidence in mill site locations was calibrated and subsequently validated using several independent sources of information. Numerous field visits to known and suspected locations for mills sites were conducted throughout the physiographic regions of South Carolina.The model of likelihood included factors for the reliability of the source, the mapping scale and spatial accuracy of the source, correlation with other sources, and likelihood of correspondence with grist mills.

Automated summit spot height generation for modern topographic mapping

Abstract. Spot elevations published on historical U.S. Geological Survey (USGS) topographic maps were established as needed to enhance information imparted by the quadrangle’s contours. In addition to a number of other features like road intersections, section corners, lakes and water wells, labels were routinely placed on mountain peaks, local high points, passes and saddles. While some elevations were established through field survey triangulation, many were “dropped” during photogrammetric stereocompilation. The average accuracy, which varied with method, terrain, and map vintage, was on the order of ± 10 feet. Today, inexpensive consumer devices deliver comparable accuracy using the Global Positioning System (GPS). Professional equipment can achieve even better accuracies. These methods have replaced the field- and funding-intensive traditional triangulation methods in the geographic sciences. However, since GPS measurements require visiting the feature location, a national dataset containing high-accuracy spot elevations has not yet been created. Consequently, modern US Topo maps are devoid of mountain peak or other spot elevations. Additionally, no government agency is currently mandated to create one. Still, US Topo map users continue to demand the display of spot heights, particularly mountain peak elevations. As lidar data are collected through the 3D Elevation Program (3DEP), the source for higher accuracy spot heights increases. Therefore, a pilot study was conducted to evaluate the feasibility of automatically generating elevation values at summits named in the Board on Geographic Name’s Geographic Names Information System (GNIS), using 3DEP data. As the 3DEP incorporates more lidar data, these values should become increasingly more accurate. The first step in the automation process involved “snapping” GNIS summits to the highest and most accurate nearby point (pixel) in the 3DEP one-third arc second seamless dataset (1/3 a-s). Development of this step was completed in a prior study and will be reported elsewhere. In Step 2, a process similar to Step 1 was implemented to identify the highest pixel in the 3DEP one-meter dataset (1-m) from an area defined by a one-pixel buffer around the snapped point pixel. After finding the highest pixel in the 1-m dataset, the same process for Step 2 was repeated on the lidar point cloud dataset (LPC). In the latter case, the highest lidar point within the buffered area was chosen where there were more than one, which was the case in most instances. In the case that the next higher resolution data were unavailable, the summit spot elevation was set to the value obtained from the lower resolution layer. Hence, if 1-m data were unavailable, but LPC data did exist, the program in its current state was unable use the higher resolution data. This is due to the fact that the 1-m data are employed to drive the software to a more precise location of the summit. Without that intermediate step the method fails to identify the correct LPC point. Work in progress is aimed at overcoming this disadvantage, which is particularly important until the time when the 1-m dataset is populated for the entire nation. The National Geospatial Program estimates the completion of lidar acquisition by 2023. Processing from acquisition to update of the 1-m dataset, including quality assessment and revisions, currently requires between two and three years. Resulting elevation values are compared to those published by Peakbagger.com and TOPOZONE.com. Preliminary results from 40 summits indicate that values derived from lidar are generally higher, whereas those populated from the 1/3 arc-second are generally lower. A thorough understanding of these relationships will require the evaluation of more points.

Validating the use of object-based image analysis to map commonly recognized landform features in the United States

ABSTRACTThe U.S. Geological Survey (USGS) National Geospatial Program (NGP) seeks to i) create semantically accessible terrain features from the pixel-based 3D Elevation Program (3DEP) data, and ii) enhance the usability of the USGS Geographic Names Information System (GNIS) by associating boundaries with GNIS features whose spatial representation is currently limited to 2D point locations. Geographic object-based image analysis (GEOBIA) was determined to be a promising method to approach both goals. An existing GEOBIA workflow was modified and the resulting segmented objects and terrain categories tested for a strategically chosen physiographic province in the mid-western US, the Ozark Plateaus. The chi-squared test of independence confirmed that there is significant overall spatial association between terrain categories of the GEOBIA and GNIS feature classes. Contingency table analysis also suggests strong category-specific associations between select GNIS and GEOBIA classes. However, 3D visual analysis revealed that GEOBIA objects resembled segmented regions more than they did individual landform objects, with their boundaries often failing to correspond to match what people would likely perceive as landforms. Still, objects derived through GEOBIA can provide initial baseline landscape divisions that can improve the efficiency of more specialized feature extraction methods.

Establishing Classification and Hierarchy in Populated Place Labeling for Multiscale Mapping for The National Map

Current standards for federal mapping call for use of the Geographic Names Information System (GNIS) point layer for placement of United States populated place labels. However, this point layer contains limited classification information and hierarchy information, resulting in problems of map quality for database-driven, multi-scale, reference mapping, such as maps served by The National Map Viewer from USGS. Database-driven mapping often relies simply on what labels fit best in the map frame. Our research investigates alternative sources for labeling populated places, including polygons defined by the U.S. Census Bureau, such as incorporated place, census designated place (CDP), and economic place. Within each of these polygon layers we investigate relevant attributes from the decennial and economic censuses, such as population for incorporated places and CDPs, and the number of employees for economic places. The data selected are available for the entire country to serve national mapping requirements. This combination of data allows a more refined classification of populated places on maps that better represents relative importance. Visual importance on maps through scale should derive from more than simply residential population, but also economic importance, though comparison is made to this simpler case. We differentiate a fourth category of GNIS populated place points, essentially “neighborhoods” and related features—which are not incorporated places, CDPs, nor economic places. Populated places in this fourth class do not have federally defined boundaries, necessitating an alternative method for determining hierarchy in label presentation through scale.

Spatial Pattern of Uncertainties: An Accuracy Assessment of the TIGER Files

This study discusses the ways how the positional accuracy of the TIGER files can be measured and spatially reported. Many people and companies use the address range of the TIGER files with the geocoding package within a Geographic Information Systems (GIS). However, the problem is that many people have little understanding of the inaccuracy of the TIGER files. This study examines the relationships between the distribution of inaccuracy and physical factors such as stream and urbanity. Next, the inaccuracy of the hydrography shape file of TIGER 2000 files is calculated by comparing it with the stream points data of United States Geological Survey (USGS)’s Geographic Names Information System. Finally, this study examines whether there are individual patterns in each spatial data by comparing the spatial pattern of the inaccuracies of the road and hydrography shape file.

Reviews

References Bright, William. 1993. Colorado Place Names. Boulder, CO.: Johnson. (Rev. 2004). Bright, William. 2004. Native American Placenames of the United States. Norman, OK: University of Oklahoma Press. Bright, William. 2005. “Contextualizing a Grammar.” Studies in Language 30:2. 245–52. Coates, Richard. 2006. “Properhood.” Language 82:2. 356–82. Geographic Names Information System (GNIS). 2002. The current database is available from the US Geological Survey at http://geonames.usgs.gov/index.html, last updated 24 July 2006. Accessed November 19, 2006. Gudde, Erwin G. 1998. California Place Names: The Origin and Etymology of Current Geographical Names. 4th ed. Revised and enlarged by William Bright. Berkeley: University of California Press. Beauchamp, William M. 1907. Aboriginal Place Names of New York. Bulletin 108, Archeology 12, New York State Museum. Bright, William. 2004. Native American Placenames of the United States. University of Oklahoma Press: Norman. Cassidy, Frederic G. 1947. The Place-Names of Dane CountyiWisconsin. American Dialect Society: Greensboro,North Carolina. French, J. H. 1860. 1977. 1860 Gazetteer of the State of New York. Reprinted 1977. The BookMark: Knightstown,Indiana. McArthur, Lewis A., and Lewis L. McArthur. 1992. Oregon Geographic Names. Oregon Historical Society Press: Portland,Oregon. McGoff, Michael F. 1980. Computer-oriented Onomastic Surveys: The Toponyms of New York State. Dissertation: Binghamton University. Nicolaisen, W. F. H. 1975. “Place-Names in Bilingual Communities,” Names 23:167–174. Stewart, George R. 1975. Names on the Globe. Oxford University Press: New York. Zelinsky, Wilbur. 1967. “Classical Town Names in the United States: The Historical Geography of an American Idea.” The Geographical Review 57:463–495.

Toponymy and Technology

During the time I was administering Geographic Names Information System (GNIS) contracts in the University of New Mexico Barth Data Analysis Center (BDAC), my mother once took me aside and in a confidential, almost conspiratorial voice, asked, Bob, what do you do? That was the question my colleagues at BDAC often asked as well. BOAC was primarily a GIS shop but also did a lot of image processing, as well as other very high-tech geographic projects. All anyone knew about GNIS was that the equipment it required was from the technological Pleistocene. Thus, they viewed GNIS as irrelevant. At staff meetings I used the word as often as possible, but I fooled no one; I wasn't manipulating bytes and pixels; I was working with names. A quaint but sometimes interesting oddity. But over time that subtly began to change. The GIS people noticed first. They discovered that GNIS wasn't just a bunch of names; as a database it could become a data layer in their GIS's. What's more, it resided not in books and on maps, as name information traditionally had, but on computers and could be transmitted electronically. Almost overnight, I acquired new respect, because names had made the transition from words to data. And· this has led me to ponder this rather momentous shift in the evolution of toponymy. It was a quarter century ago that I first became involved in toponymy, a word foreign to me at the time. I was editing and managing a little weekly newspaper in the White Mountains of northern New Hampshire, and because the business, which included a tiny printing plant, long ago had published a small booklet about the region's myths and legends that still generated revenue from the tourist trade, I reasoned that another booklet might generate still more revenue. But a new booklet about what? Well, the region was filled with interesting, curiosity-compelling place names-why not a booklet about them?

Extracting High Points from Statistical Surfaces: A Case Study Using Digital Elevation Models

We introduce and test an algorithm for extracting high-point locations from statistical surface data. The algorithm uses map algebra and local neighborhood analysis via three key parameters: minimum vertical gain, vertical gain neighborhood, and horizontal separation neighborhood. Though the method is applicable to any x,y,z data set, we tested it on 1:250,000 digital elevation models (DEMs) for Arizona. The resulting high points were compared quantitatively with an independent data set of named summits from the USGS Geographic Names Information System (GNIS). The comparison showed that, on an aggregate basis, the extraction method can approximate the number and spatial pattern of high points when compared to the GNIS points. However, extraction by neighborhood analysis may consistently misdiagnose certain features, such as the edges of troughs (e.g., canyon rims).

The Oregon File of the Geographic Names Information System: A Macro Corpus Typology

AbstractThe Oregon file of the national Geographic Names Information System (GNIS), containing some 51,500 geographic names, is being classified by type. The first part of this article discusses approximately 9,000 of these names, about which many basic facts are known and have been recorded. The second part covers a few general suppositions about various kinds of names, and discusses how certain of these suppositions compare statistically with the researched corpus. It concludes with some comments on name classification problems and suggestions as to how a large-scale study could be of use to various disciplines.

The GNIS and the PC: Two Tools for Today's Toponymic Research

AbstractThe personal computer has revolutionized geographic name research through its ability to store and manipulate vast quantities of data. Oregon has made considerable use of the information made available through the Geographic Names Information System and has combined the basic data with a program designed to aid in researching the origins and history of Oregon names. Procedures have been developed which allow the display of various types of names and naming patterns tailored to specific onomastic interests or the interests of researchers in other disciplines.

Stream Generic Terms as Indicators of Historical Settlement Patterns

AbstractU.S. geographic names may be analyzed with relative ease using data sorting techniques with the Geographic Names Information System. An intensive computer search of the stream names in each state reveals and locates on base maps, using computer-driven plotters, unusual stream generic terms, such as kill, coulee, and cañada. Tables show the relative frequency and the distribution of toponymic generic terms. The study of selected generic terms in placenames demonstrates some degree of correlation with historical patterns of settlement and provides an instructive, but not always reliable, aid to historical research.