Search USGS⌕ Search

SEARCH · Search USGS

Results for “GIS Development”

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.

At least 235 records · Page 13Linked to original sources
USGS2023

Unravelling the influence of landscape alteration from flow alteration on benthic macroinvertebrate assemblage response in the Delaware River Basin

Quantifying the effects of streamflow alteration on assemblage response is central to understanding the role humans play in shaping aquatic environments. These changes represent a level of complexity that impedes developing quantitative links between flow and ecological response because stream hydrology is strongly intertwined with natural and anthropogenic factors. Better management outcomes require disentangling these linkages. Benthic macroinvertebrate data were combined with GIS-derived natural and anthropogenic basin characteristics to identify factors associated with changes in flow processes and assemblage characteristics. Models linking streamflow metrics and macroinvertebrate response at basin and subregion scales were developed using boosted regression tree (BRT) analysis. Basin-scale BRT analyses revealed that links between macroinvertebrate response and flow metrics were often obscured, whereas more homogeneous subregions were better able to discern relations with flow. Urban land cover was the primary factor accounting for changes in flow characteristics. Elevation, land cover, and high flow frequency were the principal variables driving changes in assemblage structure within subregions. Assemblage metrics and traits were equally useful for building response models and were affected similarly by streamflow alteration. Results indicate that response models should be developed based on upland and coastal subregions. However, when defining subregions, care should be taken to maintain data sufficiency. Developing practical flow-protection standards that support a balance between human water requirements and ecological integrity requires models that reduce uncertainty and identify management-relevant drivers. However, effective management often differs by location and models developed at the subregion level may be more applicable to management and stakeholder interests.

Delaware, New Jersey, New York, Pennsylvania↗
USGS2007

Fast algorithm for automatically computing Strahler stream order

An efficient algorithm was developed to determine Strahler stream order for segments of stream networks represented in a Geographic Information System (GIS). The algorithm correctly assigns Strahler stream order in topologically complex situations such as braided streams and multiple drainage outlets. Execution time varies nearly linearly with the number of stream segments in the network. This technique is expected to be particularly useful for studying the topology of dense stream networks derived from digital elevation model data.

Water Resources Bulletin↗
USGS1993

Maps showing mineral resource assessment for porphyry and stockwork deposits of copper, molybdenum, and tungsten and for stockwork and disseminated deposits of gold and silver in the Butte 1° x 2° quadrangle, Montana

This report documents the assessment for potential occurrences of undiscovered porphyry and stockwork deposits of copper, molybdenum, and tungsten (porphyry Cu-Mo-W) and stockwork and disseminated deposits of gold and silver (disseminated Au-Ag) in the Butte 1 °X2° quadrangle. The Butte quadrangle, in west-central Montana, is one of the best known mineral producing regions in the U.S. Mining districts in the quadrangle, including the world famous Butte or Summit Valley district, have produced a variety of metallic and nonmetallic mineral commodities valued at more than $6.4 billion (at the time of production). Because of its importance as a mineral producing region, the Butte quadrangle was selected for study by the U.S. Geological Survey under the Conterminous United States Mineral Assessment Program (CUSMAP). Under this program, new data on geology, geochemistry, geophysics, geochronology, mineral resources, and remote sensing were collected and synthesized. The field and laboratory studies were supported, in part, by funding from the Geologic Framework and Synthesis Program and the Wilderness Program. The methods used in this resource assessment for porphyry Cu-Mo-W and disseminated Au-Ag deposits in the quadrangle include a compilation of all data, the development of descriptive occurrence models, and the analysis of data using techniques provided by a Geographic Information System (GIS). This map is one of several maps on the Butte 1 °X2° quadrangle. Other deposit types have been assessed for the Butte quadrangle, and maps (U.S. Geological Survey (USGS) Miscellaneous Investigation Series Maps) for each of the following have been prepared: Vein and replacement deposits of gold, silver, copper, lead, zinc, manganese, and tungsten (Elliott, Wallace, and others, 1992a) and skarn deposits of gold, silver, copper, tungsten, and iron (Elliott and others, 1992b ). Other publications resulting from this study include linear features map (Rowan and others, 1991 ); limonite and hydrothermal alteration map (Rowan and Segal, 1989); mineral occurrence maps (Elliott and others, 1986; Elliott, Loen, and others, 1992); and geologic maps (Wallace, 1987; Wallace and others, 1987).

Montana↗
USGS2016

Importance of regional variation in conservation planning: A rangewide example of the Greater Sage-Grouse

We developed rangewide population and habitat models for Greater Sage-Grouse ( Centrocercus urophasianus ) that account for regional variation in habitat selection and relative densities of birds for use in conservation planning and risk assessments. We developed a probabilistic model of occupied breeding habitat by statistically linking habitat characteristics within 4 miles of an occupied lek using a nonlinear machine learning technique (Random Forests). Habitat characteristics used were quantified in GIS and represent standard abiotic and biotic variables related to sage-grouse biology. Statistical model fit was high (mean correctly classified = 82.0%, range = 75.4–88.0%) as were cross-validation statistics (mean = 80.9%, range = 75.1–85.8%). We also developed a spatially explicit model to quantify the relative density of breeding birds across each Greater Sage-Grouse management zone. The models demonstrate distinct clustering of relative abundance of sage-grouse populations across all management zones. On average, approximately half of the breeding population is predicted to be within 10% of the occupied range. We also found that 80% of sage-grouse populations were contained in 25–34% of the occupied range within each management zone. Our rangewide population and habitat models account for regional variation in habitat selection and the relative densities of birds, and thus, they can serve as a consistent and common currency to assess how sage-grouse habitat and populations overlap with conservation actions or threats over the entire sage-grouse range. We also quantified differences in functional habitat responses and disturbance thresholds across the Western Association of Fish and Wildlife Agencies (WAFWA) management zones using statistical relationships identified during habitat modeling. Even for a species as specialized as Greater Sage-Grouse, our results show that ecological context matters in both the strength of habitat selection (i.e., functional response curves) and response to disturbance.

Ecosphere↗
USGS2011

An approach to regional wetland digital elevation model development using a differential global positioning system and a custom-built helicopter-based surveying system

Accurate topographic data are critical to restoration science and planning for the Everglades region of South Florida, USA. They are needed to monitor and simulate water level, water depth and hydroperiod and are used in scientific research on hydrologic and biologic processes. Because large wetland environments and data acquisition challenge conventional ground-based and remotely sensed data collection methods, the United States Geological Survey (USGS) adapted a classical data collection instrument to global positioning system (GPS) and geographic information system (GIS) technologies. Data acquired with this instrument were processed using geostatistics to yield sub-water level elevation values with centimetre accuracy (±15 cm). The developed database framework, modelling philosophy and metadata protocol allow for continued, collaborative model revision and expansion, given additional elevation or other ancillary data.

Florida↗
USGS2011

Geodatabase of Wyoming statewide oil and gas drilling activity to 2010

The U.S. Geological Survey (USGS) compiled a geographic information system (GIS) of Wyoming statewide historical oil and gas drilling activity for the Wyoming Landscape Conservation Initiative (WLCI). The WLCI is representative of the partnerships being formed by the USGS with other Department of the Interior bureaus, State and local agencies, industry, academia, and private landowners that are committed to maintaining healthy landscapes, sustaining wildlife, and preserving recreational and grazing uses as energy resources development progresses in southwestern Wyoming. This product complements the 2009 USGS publication on oil and gas development in southwestern Wyoming http://pubs.usgs.gov/ds/437/) by approximating, based on database attributes, the time frame of drilling activity for each well (start and stop dates). This GIS product also adds current oil and gas drilling activity not only in the area encompassing the WLCI, but also statewide. Oil and gas data, documentation, and spatial data processing capabilities are available and can be downloaded from the USGS website. These data originated from the Wyoming Oil and Gas Conservation Commission (WOGCC), represent decades of oil and gas drilling (1900 to 2010), and will facilitate a landscape-level approach to integrated science-based assessments, resource management and land-use decision making.

Wyoming↗
USGS2016

GIS methodology for geothermal play fairway analysis: Example from the Snake River Plain volcanic province

Play fairway analysis in geothermal exploration derives from a systematic methodology originally developed within the petroleum industry and is based on a geologic and hydrologic framework of identified geothermal systems. We are tailoring this methodology to study the geothermal resource potential of the Snake River Plain and surrounding region. This project has contributed to the success of this approach by cataloging the critical elements controlling exploitable hydrothermal systems, establishing risk matrices that evaluate these elements in terms of both probability of success and level of knowledge, and building automated tools to process results. ArcGIS was used to compile a range of different data types, which we refer to as ‘elements’ (e.g., faults, vents, heatflow…), with distinct characteristics and confidence values. Raw data for each element were transformed into data layers with a common format. Because different data types have different uncertainties, each evidence layer had an accompanying confidence layer, which reflects spatial variations in these uncertainties. Risk maps represent the product of evidence and confidence layers, and are the basic building blocks used to construct Common Risk Segment (CRS) maps for heat, permeability, and seal. CRS maps quantify the variable risk associated with each of these critical components. In a final step, the three CRS maps were combined into a Composite Common Risk Segment (CCRS) map for analysis that reveals favorable areas for geothermal exploration. Python scripts were developed to automate data processing and to enhance the flexibility of the data analysis. Python scripting provided the structure that makes a custom workflow possible. Nearly every tool available in the ArcGIS ArcToolbox can be executed using commands in the Python programming language. This enabled the construction of a group of tools that could automate most of the processing for the project. Currently, our tools are repeatable, scalable, modifiable, and transferrable, allowing us to automate the task of data analysis and the production of CRS and CCRS maps. Our ultimate goal is to produce a toolkit that can be imported into ArcGIS and applied to any geothermal play type, with fully tunable parameters that will allow for the production of multiple versions of the CRS and CCRS maps in order to better test for sensitivity and to validate results.

Idaho↗
USGS2007

ArcView Coal Evaluation User's Guide

Purpose: The objective of the ArcView Coal Evaluation (ACE) is to estimate the amount and location of coal available to be mined by various coal mining technologies, based on the geologic coverages developed in the National Coal Resource Assessment (NCRA) which are the starting coverages used in the Geographic Information Systems (GIS) evaluation of coal resources. The ACE Users Guide provides many examples of how to apply technical limits based upon mining technology. The methods, which are iterative for any given mining technology, should transfer directly by mining technology to other coal beds.

Open-File Report↗
USGS2003

A model for spatially and temporally distributed shallow landslide initiation by rainfall infiltration

We describe a model for regional initiation of shallow landslides based on an approximate analytic solution to Richards equation combined with an infinite-slope calculation. The model applied over digital topography computes pressure heads and factors of safety as functions of depth for geographic information system (GIS) grid cells at any time during and after rainfall events. An example is presented that simulates the progressive development of shallow landslides on steep slopes during a rainfall event. This example shows how this modeling provides insights into transient rainfall-caused processes that trigger shallow slope instability and consequent regionally distributed debris-flow events. Specifically, we infer that the spatial pattern of instability is primarily controlled by topography, while intensity and duration of rainfall, and the subsequent pore-pressure responses control the temporal pattern of instability. ?? 2003 Millpress.

Conference Paper↗
USGS1991

Application and partial validation of a habitat model for moose in the Lake Superior region

A modified version of the dormant-season portion of a Habitat Suitability Index (HSI) model developed for assessing moose ( Alces alces ) habitat in the Lake Superior Region was incorporated in a Geographic Information System (GIS) for 490 km 2 of Minnesota's Superior National Forest. Moose locations (n=235) were plotted during aerial surveys conducted in December 1988 and January 1990-1991. Dormant-season forage and cover quality for 1,000-m, 500-m, and 200-m radii plots around random points and moose locations were compared using U.S. Forest Service stand examination data. Cover quality indices were lower than forage quality indices within all plots. The median value for the average cover quality index was greater (P=0.003) within 200-m plots around cow moose locations than for plots around random points for the most severe winter of the study. The proportion of highest-quality winter cover, such as mixed stands dominated by mid-age class white spruce ( Picea glauca ) and balsam fir ( Abies balsanea ), was greater within 500-m and 200-m plots around cow moose than within similar plots around random points during the two most severe winters. These results indicate that suboptimum ratings of winter habitat quality used in the GIS for dormant-season forage >100 m from cover, as suggested in the original HSI model, are reasonable. Integrating the habitat model with forest stand data using a GIS permitted analysis of moose habitat within a relatively large geographic area. Simulation of habitat quality indicated a potential shortage of late-winter cover in the study area. The effects of forest management actions on moose habitat quality can be simulated without collecting additional data.

Lake Superior region↗
USGS2015

Predicting watershed post-fire sediment yield with the InVEST sediment retention model: Accuracy and uncertainties

Increased sedimentation following wildland fire can negatively impact water supply and water quality. Understanding how changing fire frequency, extent, and location will affect watersheds and the ecosystem services they supply to communities is of great societal importance in the western USA and throughout the world. In this work we assess the utility of the InVEST (Integrated Valuation of Ecosystem Services and Tradeoffs) Sediment Retention Model to accurately characterize erosion and sedimentation of burned watersheds. InVEST was developed by the Natural Capital Project at Stanford University (Tallis et al., 2014) and is a suite of GIS-based implementations of common process models, engineered for high-end computing to allow the faster simulation of larger landscapes and incorporation into decision-making. The InVEST Sediment Retention Model is based on common soil erosion models (e.g., USLE – Universal Soil Loss Equation) and determines which areas of the landscape contribute the greatest sediment loads to a hydrological network and conversely evaluate the ecosystem service of sediment retention on a watershed basis. In this study, we evaluate the accuracy and uncertainties for InVEST predictions of increased sedimentation after fire, using measured postfire sediment yields available for many watersheds throughout the western USA from an existing, published large database. We show that the model can be parameterized in a relatively simple fashion to predict post-fire sediment yield with accuracy. Our ultimate goal is to use the model to accurately predict variability in post-fire sediment yield at a watershed scale as a function of future wildfire conditions.

Conference Paper↗
USGS2008

Change in the forested and developed landscape of the Lake Tahoe basin, California and Nevada, USA, 1940-2002

The current ecological state of the Lake Tahoe basin has been shaped by significant landscape-altering human activity and management practices since the mid-1850s; first through widespread timber harvesting from the 1850s to 1920s followed by urban development from the 1950s to the present. Consequences of landscape change, both from development and forest management practices including fire suppression, have prompted rising levels of concern for the ecological integrity of the region. The impacts from these activities include decreased water quality, degraded biotic communities, and increased fire hazard. To establish an understanding of the Lake Tahoe basin's landscape change in the context of forest management and development we mapped, quantified, and described the spatial and temporal distribution and variability of historical changes in land use and land cover in the southern Lake Tahoe basin (279 km2) from 1940 to 2002. Our assessment relied on post-classification change detection of multi-temporal land-use/cover and impervious-surface-area data that were derived through manual interpretation, image processing, and GIS data integration for four dates of imagery: 1940, 1969, 1987, and 2002. The most significant land conversion during the 62-year study period was an increase in developed lands with a corresponding decrease in forests, wetlands, and shrublands. Forest stand densities increased throughout the 62-year study period, and modern thinning efforts resulted in localized stand density decreases in the latter part of the study period. Additionally forests were gained from succession, and towards the end of the study period extensive tree mortality occurred. The highest rates of change occurred between 1940 and 1969, corresponding with dramatic development, then rates declined through 2002 for all observed landscape changes except forest density decrease and tree mortality. Causes of landscape change included regional population growth, tourism demands, timber harvest for local use, fire suppression, bark beetle attack, and fuels reduction activities. Results from this study offer land managers within the Lake Tahoe basin and in similar regions a basis for making better informed land-use and management decisions to potentially minimize detrimental ecological impacts of landscape change. The perspective to be gained is based on quantitative retrospection of the effects of human-driven changes and the impacts of management action or inaction to the forested landscape. ?? 2008 Elsevier B.V. All rights reserved.

Forest Ecology and Management↗
USGS2009

TerraLook: A GIS-ready time-series of satellite imagery

The goal of TerraLook is to provide satellite images that anyone can use to see changes in the Earth's surface over time. Landsat and other satellites have been collecting images of the Earth from Space for more than 35 years. TerraLook allows users to create their own collections of georeferenced simulated natural-color JPEG images from Landsat (circa 1975, 1990, 2005) and ASTER (2000 to present) at no charge. The simulated natural colors in TerraLook images are consistently processed so that images from these time periods can be easily compared. TerraLook images are bundled with stands- compliant metadata, a world file, and an outline of each image's ground footprint, enabling their use in Geographic Information Systems and Web mapping applications. Different ArcGIS solutions were developed to use with TerraLook Collections. For ArcGIS and ArcGIS Explorer users, project files are available to permit TerraLook Collections to be easily integrated into GIS projects of educational curricula for visual analysis. A Pythion Script is available to extract vector data, such as country boundaries, protected areas, hydrography, and populated places, from global datasets to create extended TerraLook collections. Users can add local data, included ground photography, to the TerraLook Collections, These localized collections can be easily shared. An ArcGIS Publisher tool is provided with the ArcMap project file for publishing documents for use with the free ArcReader application.

Conference Paper↗
USGS2010

Land-Use Portfolio Modeler, Version 1.0

Natural hazards pose significant threats to the public safety and economic health of many communities throughout the world. Community leaders and decision-makers continually face the challenges of planning and allocating limited resources to invest in protecting their communities against catastrophic losses from natural-hazard events. Public efforts to assess community vulnerability and encourage loss-reduction measures through mitigation often focused on either aggregating site-specific estimates or adopting standards based upon broad assumptions about regional risks. The site-specific method usually provided the most accurate estimates, but was prohibitively expensive, whereas regional risk assessments were often too general to be of practical use. Policy makers lacked a systematic and quantitative method for conducting a regional-scale risk assessment of natural hazards. In response, Bernknopf and others developed the portfolio model, an intermediate-scale approach to assessing natural-hazard risks and mitigation policy alternatives. The basis for the portfolio-model approach was inspired by financial portfolio theory, which prescribes a method of optimizing return on investment while reducing risk by diversifying investments in different security types. In this context, a security type represents a unique combination of features and hazard-risk level, while financial return is defined as the reduction in losses resulting from an investment in mitigation of chosen securities. Features are selected for mitigation and are modeled like investment portfolios. Earth-science and economic data for the features are combined and processed in order to analyze each of the portfolios, which are then used to evaluate the benefits of mitigating the risk in selected locations. Ultimately, the decision maker seeks to choose a portfolio representing a mitigation policy that maximizes the expected return-on-investment, while minimizing the uncertainty associated with that return-on-investment. The portfolio model, now known as the Land-Use Portfolio Model (LUPM), provided the framework for the development of the Land-Use Portfolio Modeler, Version 1.0 software (LUPM v1.0). The software provides a geographic information system (GIS)-based modeling tool for evaluating alternative risk-reduction mitigation strategies for specific natural-hazard events. The modeler uses information about a specific natural-hazard event and the features exposed to that event within the targeted study region to derive a measure of a given mitigation strategy`s effectiveness. Harnessing the spatial capabilities of a GIS enables the tool to provide a rich, interactive mapping environment in which users can create, analyze, visualize, and compare different

Techniques and Methods↗
USGS2014

US Topo Maps 2014: Program updates and research

The U. S. Geological Survey (USGS) US Topo map program is now in year two of its second three-year update cycle. Since the program was launched in 2009, the product and the production system tools and processes have undergone enhancements that have made the US Topo maps a popular success story. Research and development continues with structural and content product enhancements, streamlined and more fully automated workflows, and the evaluation of a GIS-friendly US Topo GIS Packet. In addition, change detection methodologies are under evaluation to further streamline product maintenance and minimize resource expenditures for production in the future. The US Topo map program will continue to evolve in the years to come, providing traditional map users and Geographic Information System (GIS) analysts alike with a convenient, freely available product incorporating nationally consistent data that are quality assured to high standards.

Conference Paper↗
USGS2024

Using surficial geologic maps, vegetation, and monitoring to address erosion impacts from grazing in Channel Islands National Park, California

Employing a map-unit classification scheme based on geomorphic process and age, the U.S. Geological Survey (USGS) mapped Quaternary surficial deposits of the five islands comprising Channel Islands National Park (CHIS), California, as no such maps previously existed. Mapping was motivated through an agreement with the National Park Service (NPS) to aid natural resource assessments, including post-grazing disturbance recovery and identification of mass wasting and tectonic hazards. The resulting detailed (1:12,000 scale) maps portray areas of upland erosional transport processes and alluvial, fluvial, eolian, beach, marine terrace, mass wasting, and mixed depositional processes at a scale commensurate with the process signal on the landscape. Detailed, GPS-located observations of sedimentology, pedogenic development, landscape position, and geomorphology constrain map unit identification. In addition, map-unit boundary delineation was determined in a GIS using a variety of high-resolution (sub-meter) aerial imagery and lidar-based DEMs complemented by derivative raster products. The GIS geodatabase schema used is GeMS compliant. Resolvable map-unit widths are as small as 10-15 meter. Locational accuracy of unit boundaries is estimated to be as low as 10 m or better. Detailed mapping is at a scale allowing determination of geomorphic transport processes operating on the landscape and conveys geologic information to land managers at a resolution commensurate with park resources. Map unit ages are constrained by relative soil chronostratigraphy, vertebrate and invertebrate fossils, and materials collected for dating control using radiogenic, luminescence, and cosmogenic radionuclide techniques. Associated topical studies examined connections between geology, surface water, and vegetation distribution and restoration.

California↗
USGS2009

Basinsoft, a computer program to quantify drainage basin characteristics

Surface water runoff is a function of many interrelated factors including climate, soils, landuse, and the physiography of the drainage basin. A practical and effective method to quantify drainage basin characteristics would allow analysis of the interrelations of these factors, leading to an improved understanding of the effects of drainage basin characteristics on surface-water runoff. Historically, the quantification of drainage basin characteristics has been a tedious and time-consuming process. Recent improvements in computer hardware and software technology have enabled the developers of a program called Basinsoft to automate this process. Basinsoft requires minimal preprocessing of data and provides an efficient, automated procedure for quantifying selected morphometric characteristics and the option to area-weight characteristics for a drainage basin. The user of Basinsoft is assumed to have a limited amount of experience in the use of ARC/INFO, a proprietary geographic information system (GIS). (The use of brand names in this chapter is for identification purposes only and does not constitute endorsement by the U.S. Geological Survey [USGS].) In 1988, the USGS began developing a program called Basinsoft. The initial program quantified 16 selected drainage basin characteristics from three source-data layers that were manually digitized from topographic maps using the versions of ARC/INFO, Fortran programs, and prime system Command Programming Language (CPL) programs available in 1988 (Majure and Soenksen, 1991). By 1991, Basinsoft was enhanced to quantify 27 selected drainage-basin characteristics from three source-data layers automatically generated from digital elevation model (DEM) data using a set of Fortran programs (Majure and Eash, 1991: Jenson and Dominique, 1988). Due to edge-matching problems encountered in 1991 with the preprocessing

Book chapter↗
USGS2010

Flood-frequency estimates for streams on Kauai, Oahu, Molokai, Maui, and Hawaii, State of Hawaii

This study provides an updated analysis of the magnitude and frequency of peak stream discharges in Hawai`i. Annual peak-discharge data collected by the U.S. Geological Survey during and before water year 2008 (ending September 30, 2008) at stream-gaging stations were analyzed. The existing generalized-skew value for the State of Hawai`i was retained, although three methods were used to evaluate whether an update was needed. Regional regression equations were developed for peak discharges with 2-, 5-, 10-, 25-, 50-, 100-, and 500-year recurrence intervals for unregulated streams (those for which peak discharges are not affected to a large extent by upstream reservoirs, dams, diversions, or other structures) in areas with less than 20 percent combined medium- and high-intensity development on Kaua`i, O`ahu, Moloka`i, Maui, and Hawai`i. The generalized-least-squares (GLS) regression equations relate peak stream discharge to quantified basin characteristics (for example, drainage-basin area and mean annual rainfall) that were determined using geographic information system (GIS) methods. Each of the islands of Kaua`i,O`ahu, Moloka`i, Maui, and Hawai`i was divided into two regions, generally corresponding to a wet region and a dry region. Unique peak-discharge regression equations were developed for each region. The regression equations developed for this study have standard errors of prediction ranging from 16 to 620 percent. Standard errors of prediction are greatest for regression equations developed for leeward Moloka`i and southern Hawai`i. In general, estimated 100-year peak discharges from this study are lower than those from previous studies, which may reflect the longer periods of record used in this study. Each regression equation is valid within the range of values of the explanatory variables used to develop the equation. The regression equations were developed using peak-discharge data from streams that are mainly unregulated, and they should not be used to estimate peak discharges in regulated streams. Use of a regression equation beyond its limits will produce peak-discharge estimates with unknown error and should therefore be avoided. Improved estimates of the magnitude and frequency of peak discharges in Hawai`i will require continued operation of existing stream-gaging stations and operation of additional gaging stations for areas such as Moloka`i and Hawai`i, where limited stream-gaging data are available.

Hawaii↗