OBS - Input File For All Observations

  • Scroll to top of this topic Scroll to Top of Page

  • Print the current topic Print Topic

  • Expand/collapse expanding elements on page Show/Hide Expanders

Abbreviation in Name file

OBS

Purpose

This short input file contains information applicable to all observations, and needs to be included in the name file using file type OBS if any of the input file types for Observation-Process packages are included.

Documentation

•Hill, M.C., Banta, E.R., Harbaugh, A.W., and Anderman, E.R., 2000, MODFLOW-2000, the U.S. Geological Survey modular ground-water model -- User guide to the Observation, Sensitivity, and Parameter-Estimation Processes and three post-processing programs: U.S. Geological Survey Open-File Report 00-184, 210 p.

•release.txt

•underscore.txt

Related Packages

Observation Process input files

Supported in

•MODFLOW-2000

Input Instructions

This short input file contains information applicable to all observations, and needs to be included in the name file using file type OBS if any of the input file types for Observation-Process packages are included.

Input for the Observation Process for all observations is read from a file listed in the name file with "OBS" as the file type (table 2).

Data Set 0

[#Text]

Item 0 is optional and can include as many lines as desired. Each line needs to begin with the “#” character in the first column.

Text—is a character string (maximum of 79 characters) that starts in column 2. Any characters can be included in Text. The “#” character needs to be in column 1. Text is printed when the file is read and provides an opportunity for the user to include information about the model both in the input file and the associated output file.

Data Set 1

OUTNAM ISCALS [ALLFILES] (free format)

OUTNAM—a string of one to 78 nonblank characters. OUTNAM indicates whether or not the output files listed in tables 5 and 6 are produced by MODFLOW-2000 (table 5) or its post-processing programs RESAN-2000, YCINT-2000, or BEALE-2000 (table 6). If “NONE” is specified (can be any combination of upper and lower case), none of the output files are created. Otherwise, the output files are named using OUTNAM as the base followed by a period and the two- or three-character extensions listed in tables 5 and 6. The specification of lower and upper case in OUTNAM is preserved in generating the file-name base. Extensions for files that are intended for use by a post-processor, graphics program, or other program start with an underscore ( _ ); extensions for files that are intended to be read by the user start with a number sign (#). OUTNAM can include a path; constraints imposed by the operating system regarding file names and paths should be considered when specifying OUTNAM. For compatibility with the post-processing programs, the OUTNAM string should not be changed between the separate model runs used to generate the files to be read by the post processors (see header of table 5 and the discussion in chapter 7).

Table 5: Files produced by MODFLOW-2000 when OUTNAM is not “NONE” that are designed for use by plotting routines and other programs.

[Files are named as OUTNAM followed by a period and an extension that begins with an underscore. For example, if OUTNAM is “gw”, file names would be “gw._os” and so on. The files can be used as described later in this report and in table 16. Shading indicates files that probably require special simulations to be produced properly.]

 

Extension

File contents (The file contents are in the order listed. An * indicates that for each observation the listed items are followed by the OBSNAM and PLOT-SYMBOL. The files can be read as space-delimited free format.)

If the Observation Process is active:

_os

Unweighted simulated equivalents and observations *

_ww

Weighted simulated equivalents and observations *

_ws

Weighted simulated equivalents and residuals *

_r

Unweighted residuals *

_w

Weighted residuals *

_nm

Weighted residuals and probability plotting positions *

If the Sensitivity Process also is active, the following files also are produced:

_sc

Composite scaled sensitivity for each parameter, preceded by the PARNAM.

_sd

Dimensionless scaled sensitivities for each parameter, preceded by OBSNAM and PLOT-SYMBOL. Repeated for each observation.

_s1

One-percent scaled sensitivities for each parameter, preceded by OBSNAM and PLOT-SYMBOL. Repeated for each observation.

If the Parameter-Estimation Process also is active, the following files also are produced:

_ss

Sum of squared weighted residuals for each type of observation and prior information and the total; values are listed for all parameter-estimation iterations.

_pa

Parameter values for each parameter-estimation iteration, formatted for easy production of graphs showing parameter values for each iteration. This file is not produced until the end of the program; to access parameter values as each iteration is performed, use the _b file described below.

_b

Information from each parameter-estimation iteration, including parameter values formatted for easy substitution into the Sensitivity Process input file, sum of squared weighted residuals, maximum calculated fractional parameter change and the associated parameter number (as listed in the SEN file), and value of the Marquardt parameter.

_rs

Input file for post-processing program RESAN-2000 (see chapter 7)

_y0

Input file for post-processing program YCINT-2000 (see chapter 7). Produced when IYCFLG = 0.

_y1

Input file for post-processing program YCINT-2000 (see chapter 7). Produced when IYCFLG = 1.

_y2

Optional input file for post-processing program YCINT-2000 (see chapter 7). Produced when IYCFLG = 2.

_b1

One of two input files for post-processing program BEALE-2000 (see chapter 7). Produced when IBEFLG = 1.

_b2

The second input file for the post-processing program BEALE-2000 (see chapter 7). Produced when IBEFLG = 2.

_pc

This file contains a record for each estimated parameter listing: parameter name, LN, optimized parameter value, BL, BU, lower 95-percent confidence limit on the parameter value, and upper 95-percent confidence limit on the parameter value.

_wt

This file contains observation weights, or the full weight matrix if any of the observation types is entered with a variance-covariance matrix (WTQ in the Observation Process input instructions of USGS Open-File Rep ort 00-184).The first line of this file indicates whether the rest of the file contains a vector of weights or a correlated-weight matrix.  In either case, the ordering of weights is such that weights on all head observations are listed first, followed by weights on observations of the Drain, River, General Head Boundary, and Streamflow Routing (STR) Packages; observations of constant-head flow (Basic Package); and observations of the Drain-Return and Advective-Transport Packages.

The format of this file is described in underscore.txt.

_su

This file contains the sensitivity of the simulated equivalent of each observation to each parameter.

The format of this file is described in underscore.txt.

._mc

This file contains the matrix of parameter correlations.

The format of this file is described in underscore.txt.

 

Table 6: Files produced by the MODFLOW-2000 post-processors RESAN-2000, YCINT-2000, and BEALE-2000 (chapter 7) when OUTNAM is not “NONE”.

[File names are the base specified by OUTNAM followed by a period and an extension. For example, if OUTNAM is defined as “gwmodel,” the file names would be gwmodel._rd, gwmodel.#yc, and so on. Files with extensions that start with an underscore (_) are designed to facilitate plotting; files with extensions that start with a # are designed to be read by the modeler. Use these files as described later in this report and in table 16.]

Extension

File contents (The ‘_’ files contain the items listed in the order listed. An * indicates that for each observation the listed items are followed by the OBSNAM and PLOT-SYMBOL. All ‘_’ files can be read as space-delimited free format.)

If the post-processing program RESANP is executed:

#rs

Main output file.

_rd

Ordered uncorrelated deviates and probability plotting positions. *

_rg

Ordered correlated deviates and probability plotting positions. *

_rc

Cook’s D statistic for each observation. *

See also underscore.txt.

_rb

DFBeta statistics for each parameter, preceded by OBSNAM and PLOT-SYMBOL. Repeated for each observation.

See also underscore.txt.

If the post-processing program YCINT-2000 is executed:

#yc

For the listed predictions or differences, this file contains the linear, 95-percent confidence and prediction intervals. Individual and simultaneous intervals are included.

_yp

Confidence and prediction intervals on predictions. Title lines describing the type of interval are followed by data lines for each prediction. Data lines include lower limit, upper limit, prediction, and standard deviation. *

_yd

Confidence and prediction intervals on differences. Title lines describing the type of interval are followed by data lines for each difference. Data lines include lower limit, upper limit, difference, and standard deviation. *

If the post-processing program BEALE-2000 is executed:

#be

The modified Beale’s measure statistic and auxiliary information.

ISCALS—Controls printing of the observation-sensitivity tables in the primary output files. Creation of the _sc, _sd, and _s1 files is not affected by ISCALS. (ISCALS typically is specified as 1, 2, or 3. Unscaled sensitivities are rarely of interest.) The different types of sensitivities are discussed in Hill (1998, p. 14-16, 33, 38-40, 62-64).

•ISCALS < 0, No observation-sensitivity tables are printed, but a table showing composite scaled sensitivity for each parameter is printed.

•ISCALS = 0, Unscaled sensitivities are printed.

•ISCALS = 1, Dimensionless scaled sensitivities are printed. Sensitivities are scaled by multiplying by the parameter value and the square-root of the weight, which produces dimensionless numbers. If the parameter value is less than BSCAL, which is read from the Sensitivity Process input file for each parameter listed there, the parameter value is replaced by BSCAL for the scaling. The resulting values are dimensionless and equal the number of observation error standard deviations that the simulated value would change given a one-percent change in the parameter value, times 100. Composite scaled sensitivities also are printed.

•ISCALS = 2, One-percent sensitivities are printed. Sensitivities are scaled by multiplying by the parameter value and dividing by 100. If the parameter value is less than BSCAL, which is read from the Sensitivity Process input file for each parameter listed there, the parameter value is replaced by BSCAL for the scaling. The one-percent scaled sensitivities are designed to have the same dimensions as the observations and to equal the amount the associated simulated value will change given a one-percent change in the parameter value.

•ISCALS = 3, Both dimensionless and one-percent sensitivities are printed.

ALLFILES If "ALLFILES" is specified following the ISCALS value, and if the value specified for OUTNAM is not "NONE", up to three additional output files are generated with the filename base defined by OUTNAM.  The "underscore.txt" file in the doc directory describes the new files.