Scientific Investigations Report 2008–5077

Prepared in cooperation with the N.J. Department of Environmental Protection

Determination of Baseline Periods of Record for Selected Streamflow-Gaging Stations in New Jersey for Determining Ecologically Relevant Hydrologic Indices (ERHI)

By Rachel A. Esralew and Ronald J. Baker

Scientific Investigations Report 2008–5077

Thumbnail image of cover

This publication is downloadable (15.7 Mb) in PDF format.

To view or print PDF files, the freely downloadable Adobe Acrobat Reader® may be used.


Hydrologic changes in New Jersey stream basins resulting from human activity can affect the flow and ecology of the streams. To assess future changes in streamflow resulting from human activity an understanding of the natural variability of streamflow is needed. The natural variability can be classified using Ecologically Relevant Hydrologic Indices (ERHIs). ERHIs are defined as selected streamflow statistics that characterize elements of the flow regime that substantially affect biological health and ecological sustainability. ERHIs are used to quantitatively characterize aspects of the streamflow regime, including magnitude, duration, frequency, timing, and rate of change. Changes in ERHI values can occur as a result of human activity, and changes in ERHIs over time at various stream locations can provide information about the degree of alteration in aquatic ecosystems at or near those locations. New Jersey streams can be divided into four classes (A, B, C, or D), where streams with similar ERHI values (determined from cluster analysis) are assigned the same stream class.

In order to detect and quantify changes in ERHIs at selected streamflow-gaging stations, a “baseline” period is needed. Ideally, a baseline period is a period of continuous daily streamflow record at a gaging station where human activity along the contributing stream reach or in the stream’s basin is minimal. Because substantial urbanization and other development had already occurred before continuous streamflow-gaging stations were installed, it is not possible to identify baseline periods that meet this criterion for many reaches in New Jersey. Therefore, the baseline period for a considerably altered basin can be defined as a period prior to a substantial human-induced change in the drainage basin or stream reach (such as regulations or diversions), or a period during which development did not change substantially.

Index stations (stations with minimal urbanization) were defined as streamflow-gaging stations in basins that contain less than 15 percent urban land use throughout the period of continuous streamflow record. A minimum baseline period of record for each stream class was determined by comparing the variability of selected ERHIs among consecutive 5-, 10-, 15-, and 20-year time increments for index stations. On the basis of this analysis, stream classes A and D were assigned a minimum of 20 years of continuous record as a baseline period and stream classes B and C, a minimum of 10 years.

Baseline periods were calculated for 85 streamflow-gaging stations in New Jersey with 10 or more years of continuous daily streamflow data, and the values of 171 ERHIs also were calculated for these baseline periods for each station. Baseline periods were determined by using historical streamflow-gaging station data, estimated changes in impervious surface in the drainage basin, and statistically significant changes in annual base flow and runoff.

Historical records were reviewed to identify years during which regulation, diversions, or withdrawals occurred in the drainage basins. Such years were not included in baseline periods of record. For some sites, the baseline period of record was shorter than the minimum period of record specified for the given stream class. In such cases, the baseline period was rated as “poor.”

Impervious surface was used as an indicator of urbanization and change in streamflow characteristics owing to increases in storm runoff and decreases in base flow. Percentages of impervious surface were estimated for 85 streamflow-gaging stations from available municipal population-density data by using a regression model. Where the period of record was sufficiently long, all years after the impervious surface exceeded 10 to 20 percent were excluded from the baseline period. The percentage of impervious surface also was used as a criterion in assigning qualitative ratings to baseline periods.

Changes in trends of annual base flow and runoff were determined by using double-mass curves, in which cumulative discharge at a test station (x-axis) is plotted in relation to cumulative discharge at an index station (y-axis) of the same stream class. The slope of the double-mass curve is expected to remain constant unless there have been changes in the drainage basin of the test station that altered hydrologic processes. The significance of changes in the slope of the relation (breakpoints) was evaluated by analysis of covariance and visual inspection.

A final baseline period was determined for each test station by using a combination of historical records, changes in impervious surface, and double-mass analysis. The baseline period for each station was rated as excellent, good, fair, or poor by using a numerical rating procedure based on length of record, percentage of impervious surface and the results of double-mass analysis. Values for all 171 ERHI were calculated for the baseline periods of each of the 85 stations. Stream class was then determined for each test station.




Purpose and scope


Previous and Ongoing Investigations

Use of Hydrologic Indices in Regulatory Planning

Description of the Study Area

Methods of Investigation

Selection of Index Stations and Minimum Period of Record

Baseline Periods of Record for Selected Streams in New Jersey

Use of Historical Stream and Basin Information to Eliminate Non-Baseline Years and Define Preliminary Baseline Periods

Use of Impervious Surface to Eliminate Non-Baseline Years

Double-Mass-Curve Analysis

Analysis of Covariance of the Double-Mass Curve

Interpretation of Double-Mass Curves

Baseline Period of Record Determination from Double-Mass Analysis

Ranking the Baseline Period

Determination of Baseline Period for 85 Selected Streamflow-Gaging Stations

Minimum Period of Record

Historical Analysis

Land-Use/Land-Cover and Impervious Surface

Evaluation of Breakpoints on Double-Mass Curves

Final Baseline Period Determination

Assumptions and Limitations of Methods Used to Determine Baseline Periods

Historical Land Use/Land Cover

Double-Mass Curves

Summary and Conclusions

References Cited



1–2. Maps showing—

1. Location of selected streamflow-gaging stations with stream types and physiographic provinces in New Jersey

2. Impervious surface from the 1995/97 land-use coverage for all land-use polygon segments and locations of selected streamflow-gaging stations in New Jersey

3–4. Graphs showing—

3. Relation of the percentage of impervious surface from 1995–1997 geographic information system (GIS) digital land-use information to population density in persons per square mile for all 570 municipalities in New Jersey

4. Example of double-mass curves for cumulative annual runoff and base flow at sites with low variability of annual values for selected streamflow gaging stations in New Jersey

5. Flowchart showing the steps for determining baseline periods and the quality of baseline period

6–7. Graphs showing—

6. Relation of impervious surface determined from digital geographic information system (GIS) data for 570 municipalities for the years 1995–1997 to impervious surface determined from a mathematical model used to estimate impervious surface using population density for the year 1996

7. Relation of impervious surface determined from digital geographic information system (GIS) data for 85 drainage basins for the years 1995–1997 to impervious surface determined from mathematical model based on population density for the year 1996, for the same 85 drainage basins in New Jersey

8. Boxplot showing slope ratios at selected statistically significant breakpoints on the double-mass curve for test stations and index stations


1. Description of primary and surrogate ecologically relevant hydrologic Indexes (ERHIs) determined by using the New Jersey Hydrologic Integrity Assessment Process for streams in New Jersey

2. Land use at, and continuous period of record for, selected streamflow-gaging stations in New Jersey

3. Description of seven out (type) streamflow-gaging index stations in New Jersey used to determine minimum periods of record for calculating hydrologic indices and used to develop double-mass curves

4. Selected primary Ecologically Relevant Hydrologic Indices (ERHIs) used to compute the minimum period of record for each stream class at selected streamflow-gaging stations in New Jersey

5. Means and standard deviations of the percentages of impervious surface for each Type II land use for each polygon segment from the 1995/97 land-use coverage of New Jersey

6. Baseline period-of-record quality classification based on estimated impervious surface in selected drainage basins in New Jersey

7. Results of the Kruskall-Wallis test to determine the probability of no difference among groups of numbers of years for seven index stations for each primary hydrologic index type for class A, B, C and D, and the minimum number of years for a baseline period

8. Preliminary baseline periods for 85 streamflow-gaging stations in New Jersey and supporting data

9. Estimated percentages of impervious surface and correction factors for drainage basins of selected streamflow-gaging stations in New Jersey

10. Years in which estimated impervious surface in the drainage basins of selected streamflow-gaging stations in New Jersey exceeded 10 percent or 20 percent of the drainage basin area, or had increased by 15 percent

11. Estimated population density in drainage basins of 83 selected streamflow-gaging stations in New Jersey, by decade

12. Description of data for selected streamflow-gaging stations and selected index stations in New Jersey sorted by stream type, used to develop double-mass curves

13. Selected results of analysis of covariance conducted on double-mass curves for 85 streamflow-gaging stations and selected index streamflow-gaging stations on the same stream type in New Jersey

14. Evaluation of breakpoints in double-mass analysis of base-flow and runoff data for selected streamflow-gaging stations in New Jersey

15. Final baseline period and quality ranking and supporting data for 85 selected streamflow-gaging stations in New Jersey

Download: PDF of SIR2008–5077 (15.7 Mb)

U.S. Department of the Interior, U.S. Geological Survey
Persistent URL:
Page Contact Information: Contact USGS
Last modified: Thursday, January 10 2013, 06:52:29 PM
FirstGov button  Take Pride in America button