Granato, G.E., 2014, Hydrologic Drought Decision Support System (HyDroDSS): U.S. Geological Survey Open-File Report 2014–1003, 91 p., with CD–ROM. Full CD-ROM and Report On-Line
The hydrologic drought decision support system (HyDroDSS) was developed by the U.S. Geological Survey (USGS) in cooperation with the Rhode Island Water Resources Board (RIWRB) for use in the analysis of hydrologic variables that may indicate the risk for streamflows to be below user-defined flow targets at a designated site of interest, which is defined herein as data-collection site on a stream that may be adversely affected by pumping. Hydrologic drought is defined for this study as a period of lower than normal streamflows caused by precipitation deficits and (or) water withdrawals. The HyDroDSS is designed to provide water managers with risk-based information for balancing water-supply needs and aquatic-habitat protection goals to mitigate potential effects of hydrologic drought.
This report describes the theory and methods for retrospective streamflow-depletion analysis, rank correlation analysis, and drought-projection analysis. All three methods are designed to inform decisions made by drought steering committees and decisionmakers on the basis of quantitative risk assessment. All three methods use estimates of unaltered streamflow, which is the measured or modeled flow without major withdrawals or discharges, to approximate a natural low-flow regime. Retrospective streamflow-depletion analysis can be used by water-resource managers to evaluate relations between withdrawal plans and the potential effects of withdrawal plans on streams at one or more sites of interest in an area. Retrospective streamflow-depletion analysis indicates the historical risk of being below user-defined flow targets if different pumping plans were implemented for the period of record. Retrospective streamflow-depletion analysis also indicates the risk for creating hydrologic drought conditions caused by use of a pumping plan. Retrospective streamflow-depletion analysis is done by calculating the net streamflow depletions from withdrawals and discharges and applying these depletions to a simulated record of unaltered streamflow.
Rank correlation analysis in the HyDroDSS indicates the persistence of hydrologic measurements from month to month for the prediction of developing hydrologic drought conditions and quantitatively indicates which hydrologic variables may be used to indicate the onset of hydrologic drought conditions. Rank correlation analysis also indicates the potential use of each variable for estimating the monthly minimum unaltered flow at a site of interest for use in the drought-projection analysis. Rank correlation analysis in the HyDroDSS is done by calculating Spearman’s rho for paired samples and the 95-percent confidence limits of this rho value. Rank correlation analysis can be done by using precipitation, groundwater levels, measured streamflows, and estimated unaltered streamflows. Serial correlation analysis, which indicates relations between current and future values, can be done for a single site. Cross correlation analysis, which indicates relations among current values at one site and current and future values at a second site, also can be done.
Drought-projection analysis in the HyDroDSS indicates the risk for being in a hydrologic drought condition during the current month and the five following months with and without pumping. Drought-projection analysis also indicates the potential effectiveness of water-conservation methods for mitigating the effect of withdrawals in the coming months on the basis of the amount of depletion caused by different pumping plans and on the risk of unaltered flows being below streamflow targets. Drought-projection analysis in the HyDroDSS is done with Monte Carlo methods by using the position analysis method. In this method the initial value of estimated unaltered streamflows is calculated by correlation to a measured hydrologic variable (monthly precipitation, groundwater levels, or streamflows from an index station identified with the rank correlation analysis). Then a pseudorandom number generator is used to create 251 six-month-long flow traces by using a bootstrap method. Serial correlation of the estimated unaltered monthly minimum streamflows determined from the rank correlation analysis is preserved within each flow trace. The sample of unaltered streamflows indicates the risk of being below flow targets in the coming months under simulated natural conditions (without historic withdrawals). The streamflow-depletion algorithms are then used to estimate risks of flow being below targets if selected pumping plans are used.
This report also describes the implementation of the HyDroDSS. The HyDroDSS was developed as a Microsoft Access® database application to facilitate storage, handling, and use of hydrologic datasets with a simple graphical user interface. The program is implemented in the database by using the Visual Basic for Applications® (VBA) programming language. Program source code for the analytical techniques is provided in the HyDroDSS and in electronic text files accompanying this report. Program source code for the graphical user interface and for data-handling code, which is specific to Microsoft Access® and the HyDroDSS, is provided in the database. An installation package with a run-time version of the software is available with this report for potential users who do not have a compatible copy of Microsoft Access®. Administrative rights are needed to install this version of the HyDroDSS.
A case study, to demonstrate the use of HyDroDSS and interpretation of results for a site of interest, is detailed for the USGS streamgage on the Hunt River (station 01117000) near East Greenwich in central Rhode Island. The Hunt River streamgage was used because it has a long record of streamflow and is in a well-studied basin with a substantial amount of hydrologic and water-use data including groundwater pumping for municipal water supply.
HyDroDSS version 126.96.36.199 - Initial release March 27, 2014
HyDroDSS is written in Visual Basic for Applications and therefore use is limited to Microsoft Windows operating systems. The graphical display forms require a screen resolution exceeding 1024 x 768 pixels.
Output: Example Output Files
The example output text files are provided for inspection without installing or running the model. Input values used in the example are for demonstration only and should not be considered as recommended input values.
DADailyMeanPP.txt -- Daily mean flow plotting-position file
DADailyMeanTS.txt -- Daily mean flow time-series file
DAMonthlyMeanPP.txt -- Monthly mean flow plotting-position file
DAMonthlyMeanTS.txt -- Monthly mean flow time-series file
DAMonthlyMinPP.txt -- Monthly mean flow plotting-position file
DAMonthlyMinTS.txt -- Monthly mean flow time-series file
HuntRiver-RhoGraph.txt -- Spearman's rho graphing-format file for hydrologic data
HuntRiver-RhoTable.txt -- Spearman's rho table-format file for hydrologic data
ProjectionDuration.txt -- Bootstrap projection duration-curve output file.txt
For best results use the DSS installation file, it ensures that you have the proper drivers and settings. If you do not own a copy of Access the installation file will provide a runtime version for use with the HyDroDSS. The installation file was created with the Sagekey Access Deployment Wizard. The uncompiled version is available for advanced Access users. The uncompiled Microsoft Access files listed under source code are provided for future use by model developers.
HyDroDSSInstall.exe -- DSS installation file -- HyDroDSS installation instructions are on the CD-ROM
Source Code (Version 188.8.131.52)
HyDroDSS is a Microsoft Access database application, but the model code can be incorportated into stand-alone software.
Access 2000 format -- Uncompiled version in a .zip file with the application icon (forward compatible)
Access 2003 format -- Uncompiled version in a .zip file with the application icon (forward compatible)
modCalculateDepletions.bas -- Subroutines and functions necessary for calculating streamflow depletions
modSpearmansRho.bas -- Subroutines and functions necessary for calculating Spearman's Rho, confidence limits, and significance
modStochastic.bas -- Subroutines and functions necessary for doing stochastic calculations
modCalcSubroutines.bas -- Subroutines and functions necessary for checking data
modFileFolderFunctions.bas -- Subroutines and functions necessary for handling files and folders
modHandleData.bas -- To hold subroutines and functions necessary for handling different types of data
modPublicInterface.bas -- Subroutines are used throughout the application to standardize actions
modPublicVar.bas -- Public variables used in different forms and modules
32-Bit Microsoft Access vs 64-Bit Microsoft Access
This application is developed for the 32-bit version of Microsoft Access. Microsoft recommends the 32-bit version of Office for most users, because it’s more compatible with most other applications. That is why the 32-bit version of Office is the default installation for 32 and 64 bit versions of Windows. The 32-bit Office client is supported as a Windows-32-on-Windows-64 (WOW64) installation. WOW64 is the x86 emulator that enables 32-bit Windows-based applications to run seamlessly on 64-bit Windows systems. This lets users continue to use existing Microsoft ActiveX Controls and COM add-ins with 32-bit Office. Only people who have to manipulate extremely large recordsets (such as spreadsheets with more than one-million rows) would need to use the 64-bit version of Office. Furthermore, 64-bit files are not backward compatible.
As such, this database application was created and compiled by using the 32-bit version of MS Access. The installed runtime version of the 32-bit application should run, but the compiled version will not run if the user has the 64-bit version of Office installed.
If you have the 64-bit version of MS Office use the decompiled version of the 32-bit database (.accdb), copy it over to a 64-bit version of MS Access. Compact and repair the database, then compile the code. At this point you will need to address any errors (such as missing references) and once you have sucessfully compiled the database, compact and repair it again. You may make an executable version but keep an uncompiled version for future updates.
Limited support is provided for correcting bugs and clarification of how the code is intended to work. Currently, there are no known bugs or suggestions for improvement. For support or to provide feedback or suggestions for improvement, contact Greg Granato.
This software and related material (data and documentation) are made available by the USGS to be used in the public interest and the advancement of science. You may, without any fee or cost, use, copy, modify, or distribute this software, and any derivative works thereof, and its supporting documentation, subject to the USGS Software User Rights Notice
Any use of trade, product, or firm names is for decriptive purposes only and does not imply endorsement by the U.S. Government
|Title||HyDroDSS: Hydrologic Drought Decision Support System - Software page|
|Product Type||Software Release|
|Record Source||USGS Digital Object Identifier Catalog|