USGS Science⌕ Search

Geology topics

Scott D. Waltemeyer

Publications and source records attributed to Scott D. Waltemeyer.

7 recordsLinked to original sources
USGS2008

Analysis of the Magnitude and Frequency of Peak Discharge and Maximum Observed Peak Discharge in New Mexico and Surrounding Areas

Estimates of the magnitude and frequency of peak discharges are necessary for the reliable design of bridges, culverts, and open-channel hydraulic analysis, and for flood-hazard mapping in New Mexico and surrounding areas. The U.S. Geological Survey, in cooperation with the New Mexico Department of Transportation, updated estimates of peak-discharge magnitude for gaging stations in the region and updated regional equations for estimation of peak discharge and frequency at ungaged sites. Equations were developed for estimating the magnitude of peak discharges for recurrence intervals of 2, 5, 10, 25, 50, 100, and 500 years at ungaged sites by use of data collected through 2004 for 293 gaging stations on unregulated streams that have 10 or more years of record. Peak discharges for selected recurrence intervals were determined at gaging stations by fitting observed data to a log-Pearson Type III distribution with adjustments for a low-discharge threshold and a zero skew coefficient. A low-discharge threshold was applied to frequency analysis of 140 of the 293 gaging stations. This application provides an improved fit of the log-Pearson Type III frequency distribution. Use of the low-discharge threshold generally eliminated the peak discharge by having a recurrence interval of less than 1.4 years in the probability-density function. Within each of the nine regions, logarithms of the maximum peak discharges for selected recurrence intervals were related to logarithms of basin and climatic characteristics by using stepwise ordinary least-squares regression techniques for exploratory data analysis. Generalized least-squares regression techniques, an improved regression procedure that accounts for time and spatial sampling errors, then were applied to the same data used in the ordinary least-squares regression analyses. The average standard error of prediction, which includes average sampling error and average standard error of regression, ranged from 38 to 93 percent (mean value is 62, and median value is 59) for the 100-year flood. The 1996 investigation standard error of prediction for the flood regions ranged from 41 to 96 percent (mean value is 67, and median value is 68) for the 100-year flood that was analyzed by using generalized least-squares regression analysis. Overall, the equations based on generalized least-squares regression techniques are more reliable than those in the 1996 report because of the increased length of record and improved geographic information system (GIS) method to determine basin and climatic characteristics. Flood-frequency estimates can be made for ungaged sites upstream or downstream from gaging stations by using a method that transfers flood-frequency data at the gaging station to the ungaged site by using a drainage-area ratio adjustment equation. The peak discharge for a given recurrence interval at the gaging station, drainage-area ratio, and the drainage-area exponent from the regional regression equation of the respective region is used to transfer the peak discharge for the recurrence interval to the ungaged site. Maximum observed peak discharge as related to drainage area was determined for New Mexico. Extreme events are commonly used in the design and appraisal of bridge crossings and other structures. Bridge-scour evaluations are commonly made by using the 500-year peak discharge for these appraisals. Peak-discharge data collected at 293 gaging stations and 367 miscellaneous sites were used to develop a maximum peak-discharge relation as an alternative method of estimating peak discharge of an extreme event such as a maximum probable flood.

Scientific Investigations Report↗
USGS2006

Analysis of the Magnitude and Frequency of Peak Discharges for the Navajo Nation in Arizona, Utah, Colorado, and New Mexico

Estimates of the magnitude and frequency of peak discharges are necessary for the reliable flood-hazard mapping in the Navajo Nation in Arizona, Utah, Colorado, and New Mexico. The Bureau of Indian Affairs, U.S. Army Corps of Engineers, and Navajo Nation requested that the U.S. Geological Survey update estimates of peak discharge magnitude for gaging stations in the region and update regional equations for estimation of peak discharge and frequency at ungaged sites. Equations were developed for estimating the magnitude of peak discharges for recurrence intervals of 2, 5, 10, 25, 50, 100, and 500 years at ungaged sites using data collected through 1999 at 146 gaging stations, an additional 13 years of peak-discharge data since a 1997 investigation, which used gaging-station data through 1986. The equations for estimation of peak discharges at ungaged sites were developed for flood regions 8, 11, high elevation, and 6 and are delineated on the basis of the hydrologic codes from the 1997 investigation. Peak discharges for selected recurrence intervals were determined at gaging stations by fitting observed data to a log-Pearson Type III distribution with adjustments for a low-discharge threshold and a zero skew coefficient. A low-discharge threshold was applied to frequency analysis of 82 of the 146 gaging stations. This application provides an improved fit of the log-Pearson Type III frequency distribution. Use of the low-discharge threshold generally eliminated the peak discharge having a recurrence interval of less than 1.4 years in the probability-density function. Within each region, logarithms of the peak discharges for selected recurrence intervals were related to logarithms of basin and climatic characteristics using stepwise ordinary least-squares regression techniques for exploratory data analysis. Generalized least-squares regression techniques, an improved regression procedure that accounts for time and spatial sampling errors, then was applied to the same data used in the ordinary least-squares regression analyses. The average standard error of prediction for a peak discharge have a recurrence interval of 100-years for region 8 was 53 percent (average) for the 100-year flood. The average standard of prediction, which includes average sampling error and average standard error of regression, ranged from 45 to 83 percent for the 100-year flood. Estimated standard error of prediction for a hybrid method for region 11 was large in the 1997 investigation. No distinction of floods produced from a high-elevation region was presented in the 1997 investigation. Overall, the equations based on generalized least-squares regression techniques are considered to be more reliable than those in the 1997 report because of the increased length of record and improved GIS method. Techniques for transferring flood-frequency relations to ungaged sites on the same stream can be estimated at an ungaged site by a direct application of the regional regression equation or at an ungaged site on a stream that has a gaging station upstream or downstream by using the drainage-area ratio and the drainage-area exponent from the regional regression equation of the respective region.

Scientific Investigations Report↗
USGS2002

Analysis of the magnitude and frequency of the 4-day annual low flow and regression equations for estimating the 4-day, 3-year low-flow frequency at ungaged sites on unregulated streams in New Mexico

Two regression equations were developed for estimating the 4-day, 3-year (4Q3) low-flow frequency at ungaged sites on unregulated streams in New Mexico. The first, a statewide equation for estimating the 4Q3 low-flow frequency from drainage area and average basin mean winter precipitation, was developed from the data for 50 streamflow-gaging stations that had non-zero 4Q3 low-flow frequency. The 4Q3 low-flow frequency for the 50 gaging stations ranged from 0.08 to 18.7 cubic feet per second. For this statewide equation, the average standard error of estimate was 126 percent and the coefficient of determination was 0.48. The second, an equation for estimating the 4Q3 low-flowfrequency in mountainous regions from drainage area, average basin mean winter precipitation, and average basin slope, was developed from the data for 40 gaging stations located above 7,500 feet in elevation. For this regression equation, the average standard error of estimate was 94 percent and the coefficient of determination was 0.66. A U.S. Geological Survey computer-program interface for a geographical information system (GIS), called the GISWeasel,was used to determine basin and climatic characteristics for 84 gaging stations that were not affected by regulation. Mean monthly precipitation estimates from 1961 to 1990 were used in the GIS Weasel to compute the climatic characteristics of average basin winter precipitation and annual mean precipitation. The U.S. Geological Survey National Elevation Dataset, which currently consists of the 7.5-minute, 30-meter digital elevation model for each State, was used in the GISWeasel to compute the basin characteristics of drainage area, average basin slope, average basin elevation, and average basin aspect. Basin and climatic characteristics that were statistically significant in the regression equation with the 4Q3 lowflow frequency were drainage area, which ranged from 1.62 to 5,900 square miles; average basin mean winter precipitation, which ranged from 3.89 to 19.42 inches; and average basin slope, which ranged from 0.166 to 0.517 percent.

Water-Resources Investigations Report↗
USGS2001

Estimates of mountain-front streamflow available for potential recharge to the Tularosa Basin, New Mexico

Streamflow in the Tularosa Basin, New Mexico, infiltrates into alluvial-basin aquifers at or near mountain fronts. Streamflow at or near mountain fronts is a substantial component of potential recharge to these aquifers. Streamflow response from precipitation differs substantially between the streams draining the Sacramento Mountains on the eastern side of the basin and those draining the San Andres Mountains on the western side. Mean annual streamflow at mountain fronts that is available for potential recharge to the Tularosa Basin was estimated using two regional regression methods. The method for estimating mean annual streamflow using basin-climatic characteristics was applied to 46 subbasins in the Tularosa Basin. Drainage areas for the subbasins ranged from 0.87 to 157 square miles, and mean annual precipitation ranged from 11.80 to 24.89 inches. Mean annual streamflow to the basin is estimated to be about 95 cubic feet per second or 68,800 acre-feet using the basin-climatic characteristics method. The method for estimating mean annual streamflow using channel-geometry characteristics was applied to 12 subbasins in the Tularosa Basin. Of the 46 basins, 31 had drainage areas less than 20.7 square miles and 3 had active-channel widths less than 15 feet, which were outside the ranges used to develop the regression equations.

Water-Resources Investigations Report↗
USGS2001

Relations for estimating unit-hydrograph parameters in New Mexico

Data collected from 20 U.S. Geological Survey streamflow-gaging stations, most of which were operated in New Mexico between about 1969 and 1977, were used to define hydrograph characteristics for small New Mexico streams. Drainage areas for the gaging stations ranged from 0.23 to 18.2 square miles. Observed values for the hydrograph characteristics were determined for 87 of the most significant rainfall-runoff events at these gaging stations and were used to define regional regression relations with basin characteristics. Regional relations defined lag time (tl), time of concentration (tc), and time to peak (tp) as functions of stream length and basin shape. The regional equation developed for time of concentration for New Mexico agrees well with the Kirpich equation developed for Tennessee. The Kirpich equation is based on stream length and channel slope, whereas the New Mexico equation is based on stream length and basin shape. Both equations, however, underestimate tc when applied to larger basins where tc is greater than about 2 hours. The median ratio between tp and tc for the observed data was 0.66, which equals the value (0.67) recommended by the Natural Resources Conservation Service (formerly the Soil Conservation Service). However, the median ratio between tl and tc was only 0.42, whereas the commonly used ratio is 0.60. A relation also was developed between unit-peak discharge (qu) and time of concentration. The unit-peak discharge relation is similar in slope to the Natural Resources Conservation Service equation, but the equation developed for New Mexico in this study produces estimates of qu that range from two to three times as large as those estimated from the Natural Resources Conservation Service equation. An average value of 833 was determined for the empirical constant Kp. A default value of 484 has been used by the Natural Resources Conservation Service when site-specific data are not available. The use of a lower value of Kp in calculations generally results in a lower peak discharge. A relation between the empirical constant Kp and average channel slope was defined in this study. The predicted Kp values from the equation ranged from 530 to 964 for the 20 flood-hydrograph gaging stations. The standard error of estimate for the equation is 36 percent.

Water-Resources Investigations Report↗
USGS1997

Methods for estimating magnitude and frequency of floods in the southwestern United States

Methods were developed for estimating magnitude and frequency of floods at gaged and ungaged sites in the Southwestern United States. Estimating equations for ungaged sites were developed by transferring information from gaged to ungaged sites using multiple regression and a hybrid method developed during this study. The methods and analysis are described, and information from more than 1,300 gaged sites is tabulated.

Arizona, California, Colorado, Idaho, Nevada, New ↗