release.txt MODFLOW-NWT Release Notes This file describes changes introduced into MODFLOW-NWT with each official release; these changes may substantially affect users. Executable files for personal computers are provided as well as the source code at the following URL: http://water.usgs.gov/ogw/modflow-nwt/ MODFLOW Files: Up-to-date descriptions of the input requirements for all MODFLOW-NWT Packages and Processes are provided in the Online Guide to MODFLOW-NWT (http://water.usgs.gov/ogw/modflow-nwt/MODFLOW-NWT-Guide/). NOTE: Any use of trade, product or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government. o Version 1.2.0 03/01/2020: Agricultural (AG) Water Use Package • This is the first version of MODFLOW-NWT that includes this package. Refer to Niswonger (2020) for details. Documentation for the AG Package in included in the pdf document called AG_Package_EM&S.pdf located in the "doc" folder. Additionally, input instructions for the AG Package are included in the document called Input_instructions_AG.pdf located in the "doc" folder. Streamflow Routing (SFR2) Package • A check was added to allow stream reaches to have altitudes below the cell bottom if streambed K (UHC) is zero. • Capability was added to allow a single tabfile to specify flows into multiple stream reaches. Well (WEL) package • Character variable text was initialized in PARSEWELLOPTIONS • A bug was fixed that was setting IUNITRAMP to default value; an incorrect write statement was stating that PSIRAMP was also changed; however, it was not changing from user-specified value. The default value for PSIRAMP was changed to 0.10. Unsatuated-Zone Flow (UZF1) Package • A new option was included for specifying the layer to which UZF recharge is added (NUZTOP=4). If NUZTOP is specified as 4, then the values of IUZFBND are used to define the top of the unsaturated zone, and recharge is added to the uppermost layer in a column that contains a water table. The layer to which recharge is added is updated at the beginning of each new time step and is held constant during a time step. • A small bug was fixed that was causing the counter on trailing waves to be off by one. This bug only affected model in very rare cases; but, this bug could cause the model to stop suddenly. • The calculation for total applied runoff that is output to the UZF gage file option 4 was incorrect. This variable only is used for gage file output and does not affect any other calculated values. • A new option was added to simulate root uptake from the unsaturated zone. This aption simulates root water uptake using a capillary gradient between a user specified root pressure and the simulated capillary pressure in the root zone. This formulation is documented in Lappala and others (1987). Lappala, E. G., Healy, R. W., & Weeks, E. P. (1987). Documentation of computer program VS2D to solve the equations of fluid flow in variably saturated porous media. Water-Resour. Invest. Rep, 83, 4099. Lake (LAK) Package • The interval for which derivatives are smoothed to zero in the calculation of specified outflows from lakes was changed from a constant to a variable. • The dead pool storage for a lake (volume of water below lowest outflow reach) is calculated and printed to the main listing file. Basic (BAS) Package • The size of character variables was increased to print larger words to the LIST file budget table. Link-MT3DMS (LMT) Package • A change was made to make diversions from the last reach of a segment, rather than the first reach as was incorrectly assumed. • Bug fix to prevent specification of linker file in both the name file and the LMT input file. This was creating a potential for conflicting unit numbers • Bug fix in LMT that was wrongly listing the first reach of IUPSEG as the 'from' node when it should have been using the last reach of the IUPSEG as the node from which flow is diverted to the diversion segment. • Bug fix for writing cross-sectional area to linker file for MT3D-USGS. Previously, when a diversion diverts all flow, the code was using the wrong variable for flow for calculating stream cross-sectional area. Gage Package • A minor output problem occurred when writing output for lakes when the Lake Package input specified RNF as a negative value, which is used as a flag, and OUTTYPE = 1, 3, or 4. The Gage output file incorrectly included the negative value rather than the actual computed value of runoff to the lake. This problem was corrected. The model user does not need to do anything differently. Note, all model calculations were correct (and not affected by the fix) and that correct values for runoff to a lake were always printed in the main output (listing) file. Upstream Weighting (UPW) Package • A warning was added indicating that IPHDRY is set to 1 (print HDRY) and observation packages are active. IPHDRY should be set to zero because if IPHDY is set to 1 then observation values could erroneously be calculated using HDRY. o Version 1.1.4 04/01/2018: Well (WEL) package Character variable text was initialized in PARSEWELLOPTIONS Streamflow Routing (SFR2) Package Subroutine CALC_XSA was added to calculate cross-sectional area for MT3D-USGS. Smoothing of conductance in SFR used stream head and was changed to use depth. Added error handling for case of negative flows specified in headwater segments. Added variable IGRID to call list in subroutine SFR7AD. Lake (LAK) Package Error was fixed in calculation for lake volume for the linear case. Drain Return Flow (DRT) Package When a return cell went dry, the program failed to write a budget record for the COMPACT BUDGET option. This resulted in an incorrect number of records. Program was modified so that a zero value is written to the budget file for a return cell if it is inactive. Link-MT3DMS (LMT) Package Changes were made to account for negative values of FLOW in SFR input. Checks were added for DELT=0 before dividing by DELT. MODFLOW-NWT Main file (MF_NWT.f) Added argument IGRID for call to SFR7AD Gage Package Added check to make sure SFR or LAK are active. Upstream Weighting (UPW) Package A warning was added indicating that IPHDRY is set to 1 (print HDRY) and observation packages are active. IPHDRY should be set to zero because if IPHDY is set to 1 then observation values could erroneously be calculated using HDRY. Seawater Intrusion (SWI2) Package Added FIXEDZETA option, which fixes the zeta surfaces during the simulation and does not allow them to move. Multi-Node Well (MNW2) Package These changes were made for MODFLOW-2005 version 1.12 and updated into the version for NWT. These changes were copied from the file "readme_MNW2.txt" included in the MODFLOW-2005 release. A bug was fixed that affected the calculation of intraborehole flow (initiated when PUMPLOC is not = 0) for output purposes only and only for certain conditions. These values are written to the MNWOBS file if QBHflag>0. Additional print statements were also added so that data on the location of the pump intake will always be written to the main listing file. The program will not calculate corrections for partial penetration of a MNW2 well if the losstype=SPECIFYcwc, in part because the required well radius is not specified in the input data for this losstype. If the input for a particular MNW2 well specifies losstype=SPECIFYcwc and PPFLAG>0, the code will reset PPFLAG to a value of 0, and print a related warning message. Code was added to check if the water table drops below the bottom of the open interval in a MNW2 node. If it did, that node is deactivated. Problems related to the partial penetration correction calculation have been reported for MNW2 nodes in convertible cells. To minimize these, the PPFLAG is reset to 0 if the head in the well (hwell) declines to a level that only leaves less than 10% of the open interval saturated or if either hwell or the water table drops below the bottom of the open interval (in which case the node is also deactivated). A bug in the code for the case of a horizontal or slanted well and LOSSTYPE=GENERAL caused the code to ignore the specified value of "B" (the linear well-loss coefficient) and instead assumed that it was zero. This has been fixed. Note that the algorithm assumes that there is no directional dependence of B and that the specified value applies equally in all directions. This bug would have had only a very small effect on results in most cases. Write statements were modified to help assure that elevations of well screens (open intervals) were printed correctly. Penetration fractions values will be printed as "N/A" for cases where partial penetration calculations are not made. As described on p. 53 of the MNW2 documentation report (specifically, the last sentence describing both "QNDflag" and "QBHflag"), additional data will not be written if the well contains only one node. That is, both of these flags should be equal to 0 if the MNW well only includes a single node. If not, code added to the MNWI routines will detect this inconsistency and reset these flags to zero and also write an information message in the main output file when this triggering condition is found. Several coding changes were made based on suggestions from Scott Boyce (CA WSC) and are labeled with his initials (seb). A bug was fixed for writing header information to the compact budget output file when no wells are active. Other minor format changes were made for the output files. o Version 1.1.3 08/01/2017: Well (WEL) package Added ability to reuse the same well tabfile for more than one well. Changed reading the tabfiles to allow comment lines in tabfiles. Modified output of pumping rates to LST file to separate from printing of wells with reduced pumping. Modified keyword input formats to support backward compatibility of previous keyword input formats. Character variable PHIRAMP can be used instead of SPECIFY for inputing PHIRAMP and optional output file unit number for writing reduced pumping rates. Thus, a phiramp value and outfile file unit number can be specified, respectively, in the first non-comment record in the WEL input file as: OPTIONS phiramp 0.1 15 END Unsaturated-Zone Flow (UZF) package Fixed bug that occurred when IUZFOPT=0 and element 0 accessed in array DELSTOR. Fixed keyword options to support old format. Fixed bug in UZ initialization to allow multiple consecutive steady state stress periods. Fixed bug causing UZF land surface to be incorrectly set for cells underlying lake cells. Streamflow Routing (SFR2) Package Keyword options were modified to allow for backward compatibility of previous input formats of keywords, including single line lists of keywords and keywords not surrounded by OPTIONS and END identifiers. A bug was fixed that affects models that use the LOSSFACTOR option for reducing streambed hydraulic conductivity for losing streams. This bug fix significantly affected stream seepage loss for models that specified a lossfactor value less than 1. Lake (LAK) Package Fixed array index from DELVOLLAK(NSSAR) to DELVOLLAK(NLAKES). Fixed lake package derivatives for better solutions. Fixed instability issue with Lake Package for small lakes and large surface inflow and outflow. Multi-Node Well (MNW2) Package Initialized variable Hwell. Fixed reference to lbotm in botm arrays used for models with quasi-confining units. Link-MT3DMS (LMT) Package Terms appearing in either SFR or LAK were removed from the LMTMODULE and instead placed in SFRMODULE or LAKMODULE. Added IGWET to the LMTMODULE and allocated the variable to avoid memory overwrite errors. Bug fix for binary output in LMT (an extra term was being written that was causing problems to the FMI package in MT3D-USGS). Fixed a bug in LMT related to the linkage of groundwater discharge (or runoff) to a sink (that is, not connected to surface-water through IRUNBND). Bug fix to LMT to use KPER and KSTP from PRMSMODULE instead of GSFMODFLOW MODFLOW-NWT Main file (MF_NWT.f) Changed code to write a warning if model fails to converge and NWT solver character variable CONTINUE is specified. The phrase "Normal termination of simulation" is no longer printed to the screen or LST file if the model fails to converge. o Version 1.1.2 09/15/2016: MODFLOW-NWT version 1.1.2 is based on MODFLOW-2005 version 1.11.0. Unsaturated-Zone Flow (UZF) Package A floating point exception was fixed. WELL (WEL) Package A bug was fixed that would cause the pumping rate to be calculated incorrectly when using the tabfile option and the time values specified in the tabfile did not extend to the end time of the simulation. Multi-Node Well (MNW2) Package A floating point exception was fixed. Streamflow Routing (SFR2) Package An access violation was fixed. o Version 1.1.1 07/28/2016: MODFLOW-NWT version 1.1.1 is based on MODFLOW-2005 version 1.11.0. Unsaturated-Zone Flow (UZF) Package A bug was fixed to initialize a variable. This bug could affect convergence for problems that simulate linear groundwater ET in the UZF Package when using the NWT solver. o Version 1.1.0 07/21/2016: MODFLOW-NWT version 1.1.0 is based on MODFLOW-2005 version 1.11.0. This means that MODFLOW-2005 source files used in MODFLOW-NWT were updated to reflect changes made for MODFLOW-2005 version 1.11.0. Refer to release notes for MODFLOW-2005 version 1.11.0 for these changes. This release provides support for the new transport model called MT3D-USGS. Changes to MODFLOW-NWT to support MT3D-USGS were made to the source file lmt8_NWT.f. These changes include writing output for the UZF, SFR2, and LAK7 packages that is required for simulating transport in the unsaturated zone, streams, and lakes. Refer to the MT3D-USGS documentation report (Bedekar and others, 2016) for details. Refer to the LMT release notes file "LMT8.pdf" located in ..\MODFLOW-NWT_1.1.0\doc Upstream Weighting (UPW) Package An error was fixed in the calculation of the derivative for smoothing conductance during complete wetting of a cell. Testing using several problems indicated that this change has minimal effects on the solution. NWT solver (NWT) Package Previous versions of MODFLOW-NWT reported a warning when the difference between a specified cell top and cell bottom was less than 100*Thickfact, where Thickfact is an input variable specified within the NWT solver input file. These cells were set to inactive by setting IBOUND=0. This conditions is now treated as an error and the model will stop if TOP-BOT<100*Thickfact. An error was fixed in the calculation of the derivative for smoothing conductance during complete wetting of a cell. Testing using several problems indicated that this change has minimal effects on the solution. A minor memory allocation error was fixed in GMRES that affected compilations using the gfortran compiler Default solver input values for the xmd solver (Linmeth=2) were modified to improve convergence. Solver input values for the SIMPLE, MODERATE, and COMPLEX default options are: Simple: DBDTHETA = 0.97 DBDKAPPA = 0.0001 DBDGAMMA = 0.0 MOMFACT = 0.0 BACKFLAG = 0 MAXBACKITER = 20 BACKTOL = 1.5 BACKREDUCE = 0.97 IACL = 1 NORDER = 0 LEVEL = 1 NORTH = 5 IREDSYS = 1 RRCTOLS = 0.0 IDROPTOL = 1 EPSRNS = 5.0e-3 HCLOSEXMD = 1.0e-3 MXITERXMD = 50 Moderate: DBDTHETA = 0.90 DBDKAPPA = 0.00001 DBDGAMMA = 0.0 MOMFACT = 0.1 BACKFLAG = 0 MAXBACKITER = 20 BACKTOL = 1.1 BACKREDUCE = 0.9 IACL = 1 NORDER = 0 LEVEL = 1 NORTH = 10 IREDSYS = 1 RRCTOLS = 0.0 IDROPTOL = 1 EPSRNS = 5.0e-3 HCLOSEXMD = 1.0e-3 MXITERXMD = 100 Complex: DBDTHETA = 0.85 DBDKAPPA = 0.00001 DBDGAMMA = 0.0 MOMFACT = 0.1 BACKFLAG = 1 MAXBACKITER = 50 BACKTOL = 1.1 BACKREDUCE = 0.7 IACL = 2 NORDER = 0 LEVEL = 15 NORTH = 1 IREDSYS = 1 RRCTOLS = 0.0 IDROPTOL = 1 EPSRNS = 1.0e-3 HCLOSEXMD = 1.0e-4 MXITERXMD = 200 Streamflow Routing (SFR2) Package The input format for specifying character variable options has changed. Rather than specifying character variable options in a single line in a particular order, each character variable option is specified on a separate line. If additional input is required along with the character variable option then this value is specified on the same line as the character variable option separated by one or more spaces. Options must be proceeded with the "OPTIONS" specification and followed by the "END" specification. Characters can be specified as upper or lower case. Optional character variables for SFR2 in any order: [OPTIONS] [REACHINPUT] [TRANSROUTE] [TABFILES Numtab Maxval] [LOSSFACTOR Factor] [END] Definitions for new character variable option: LOSSFACTOR -- An optional character variable. When LOSSFACTOR is specified, the real variable Factor is multiplied by STRHC1 or Hc1fact and Hc2fact to calculate seepage loss from streams. Calculation of groundwater seepage to streams is unchanged. Some variables in the SFR2 Package were initialized. Some variables were changed to arrays for use in MT3D-USGS. Unsaturated-Zone Flow (UZF) Package Note that the following changes to the UZF options will no longer support previous versions of UZF input files that used key word options. Previous UZF1 input files must be modified to be consistent with the new character variable input options described below. Some variables in the UZF Package for calculating runoff were initialized. A minor bug was fixed that relates to simulating unsaturated-zone flow beneath lakes. New Options were added to UZF1 to allow the hydraulic conductivity used to calculate rejected infiltration and surface leakage to be different than the vertical hydraulic conductivity of the unsaturated zone (VKS). A new function was added for simulating groundwater ET. This function simulates a constant ET rate over the extinction depth rather than linearly reducing ET as groundwater head decreases. The ET is smoothly reduced to zero using the same polynomial function used to reduce pumping in drying cells (Niswonger and others, 2011). ET is smoothly reduced as groundwater head drops to the extinction depth. The smoothing interval is specified as a factor of the extinction depth. The input format for specifying character variable options has changed. Rather than specifying character variable options in a single line in a particular order, each character variable option is specified on a separate line. If additional input is required along with the character variable option then this value is specified on the same line as the character variable option separated by one or more spaces. Options must be proceeded with the "OPTIONS" specification and followed by the "END" specification. Letters can be specified as upper or lower case. Optional character variables for UZF1 in any order: OPTIONS [SPECIFYTHTR] [SPECIFYTHTI] [NOSURFLEAK] [SPECIFYSURFK] [REJECTSURFK] [SEEPSURFK] [ETSQUARE smoothfact] [NETFLUX unitrech unitdis] END Definitions of new variables: SPECIFYSURFK -- An optional character variable. When SPECIFYSURFK is specified, the variable SURFK is specified in Data Set 4b. REJECTSURFK -- An optional character variable. When REJECTSURFK is specified, SURFK instead of VKS is used for calculating rejected infiltration. REJECTSURFK only is included if SPECIFYSURFK is included. SEEPSURFK -- An optional character variable. When SEEPSURFK is specified, SURFK instead of VKS is used for calculating surface leakage. SEEPSURFK only is included if SPECIFYSURFK is included. ETSQUARE -- An optional character variable. When ETSQUARE is specified, groundwater ET is simulated using a constant potential ET rate, and is smoothed over a specified smoothing interval. This option is recommended only when using the NWT solver. smoothfact -- An optional real variable specified if ETSQAURE is specified. For example, if the interval factor (smoothfact) is specified as smoothfact=0.1 (recommended), then the smoothing interval will be calculated as: SMOOTHINT = 0.1*EXTDP and is applied over the range for groundwater head (h): h < CELTOP-EXTDP, ET is zero; CELTOP-EXTDP < h < CELTOP-EXTDP+SMOOTHINT, ET is smoothed; CELTOP-EXTDP+SMOOTHINT < h, ET is equal to potential ET. NETFLUX -- An optional character variable. When NETFLUX is specified, the sum of recharge (L3/T) and the sum of discharge (L3/T) is written to separate unformatted files using module UBDSV3. Unitrech and Unitdis are the unit numbers to which these values are written when “SAVE BUDGET” is specified in Output Control. Values written to Unitrech are the sum of recharge values for the UZF, SFR2, and LAK packages, and values written to Unitdis are the sum of discharge values for the UZF, SFR2, and LAK packages. Values are averaged over the period between output times. DATA SET 4b: [SURFK (NCOL, NROW)] -- U2DREL SURFK -- An optional array of positive real values used to define the hydraulic conductivity (LT-1). SURFK is used for calculating the rejected infiltration and/or surface leakage. IF SURFK is set greater than VKS then it is set equal to VKS. A new test problem is included in \MODFLOW-NWT_1.1.0\data called UZFtestoptions that provides example input for these new options. Lake (LAK) Package Some variables were changed to arrays in order to save data for MT3D-USGS. A minor bug was fixed that relates to simulating unsaturated-zone flow beneath lakes. WELL (WEL) Package Note that the following changes to the Well Package input no longer supports previous input files that read the key word option SPECIFY in Data Set 2B. The input format for specifying the keyword (character) variable SPECIFY has changed. This character variable is now specified within the OPTIONS block that is specified between Data Set 0 and Data Set 1. A new character variable option was added that allows users to specify pumping rates using separate tabular input files for each well. Each tabular input file consists of a series of simulation times and pumping rates that should begin with time equal to zero. The applied pumping rate will be interpolated from the tabular input file using the times at the end and beginning of the time step. Options must be proceeded with the "OPTIONS" specification and followed by the "END" specification. Characters can be specified as upper or lower case. Optional character variables for WELL in any order: OPTIONS [SPECIFY PHIRAMP IUNITRAMP] [TABFILES NUMTAB MAXVAL] END Definitions for new character variable option: TABFILES -- An optional character variable. When TABFILES is specified, all pumping rates for all wells in the model will be input using separate time series tabular input files. NUMTAB -- An integer variable that is the number of tabular input files that will be read for interpolating pumping rates for each well in the model. A separate tabular input file is required for each well in the model. Wells can be made inactive by specifying pumping rates of zero for times when a well is inactive. MAXVAL -- An integer value equal to the maximum number of time series values specified in the tabular input file, not including header lines. Tabular input files each can have a different number of values and MAXVAL is the number of lines in the tabular input file that has the most lines as compared to all other tabular input files. IF TABFILES is specified then DATA SET 6 is replaced with the following NUMTAB lines for stress period 1 only. DATA SET 6 is not required for stress periods following the first stress period when the TABFILES option is used. [TABUNIT TABVAL TABLAY TABROW TABCOL] #repeat this line NUMTAB times for stress period 1. TABUNIT -- An integer value equal to the unit number for the tabular input file. This unit number must match the unit number specified in the MODFLOW Name file as type data. TABVAL -- An integer value equal to the number of lines in this tabular input file not including header lines. TABLAY -- An integer value equal to the MODFLOW layer for the cell to which the pumping rates will be applied. TABROW -- An integer value equal to the MODFLOW row for the cell to which the pumping rates will be applied. TABCOL -- An integer value equal to the MODFLOW column for the cell to which the pumping rates will be applied. A new test problem was added to the directory .\MODFLOW-NWT_1.1.0\data that includes example input files for using the tabular input file option for specifying pumping. A bug was fixed that sets pumping applied to inactive cells to zero. This bug only affects results printed to the cell by cell budget file. A small error was fixed in the calculation of the derivative used for smoothing the pumping rate to zero for drying cells. Multi-Node Well (MNW2) Package A bug was fixed to correct budget calculations when the UPW Package was used with MNW2. The model stopped and an error was reported when a well was screened in multiple cells and one of the cells was inactive. This error was changed to be a warning if one of the screened cells is active and execution continues. Surface Water Routing (SWR) Process The names of the source files were changed from gwf2swr7_NWT.f to gwf2swr7.f and gsol7.f to gwf2swr7util.f. See SWRProcess_release.txt for a summary of the changes made to SWR Process v 1.04. Source files that were modified for this release are: gwf2uzf1_NWT.f,gwfuzfmodule_NWT.f, gwf2sfr7_NWT.f,gwfsfrmodule_NWT.f,gwf2lak7_NWT.f, NWT1_gmres.f90, gwf2swr7.f, gwf2swr7util.f, gwf2mnw27_NWT.f, gmres.f90, NWT1_solver.f and NWT1_xmd.f, and lmt8_NWT.f. o Version 1.0.9 07/01/2014: Basic (BAS) package Fixed output file control for parallel writing of output to files. Well (WEL) package Initialization of a variable was added as it was not previously getting initialized correctly. Fixed minor error in smoothing function derivative. Unsaturated-Zone Flow (UZF) package A bug was fixed to correct the applied infiltration rate as printed to the UZF1 gage output file summed over the entire model. Previous versions did not include applied infiltration rates in excess of VKS. This bug did not affect calculations or solutions made by UZF, rather only the values for applied infiltration written to UZF1 gage output summed over the entire model were affected. Lake (LAK) Package Fixed compact budget for outputting lake seepage. A small change was made that sets layer indices for calculating lakebed conductance. Multi-Node Well (MNW2) Package A bug was fixed that was causing the cell to well transmissivity to incorrectly be set to zero. Link-MT3DMS (LMT) Package Some changes were made to output model results to be compatible with MT3DMS for simulations that include UZF1. o Version 1.0.8 09/24/2013: A significant change to MODFLOW-NWT for this release was the addition of the Seawater Intrusion (SWI2) Package for MODFLOW (Bakker and others, 2013). Refer to the SWI2 documentation report available at http://pubs.usgs.gov/tm/6a46/ for details regarding the functionality and input required for SWI2. Modifications to MODFLOW-NWT that were made to support SWI2 include the addition of one new source code file (gwf2swi27_NWT.f) and the addition of call statements added to the MODFLOW-NWT main source code file (MF_NWT.f). Lake (LAK) Package A bug was fixed to correctly print lake seepage to unformatted budget files. This bug affects models with lake cells that are surrounded by inactive groundwater cells in the same lake. Small changes were made to the following source code files to remove specification of machine constants for single precision variables: mach_mod.f, modules.f, NWT1_gmres.f, NWT1_ilupc_mod.f, and NWT1_module.f Streamflow Routing (SFR2) Package A bug was fixed to avoid a memory-allocation error that occurred when using the tabfile option for models with more tabfiles than stream segments. Also, an uninitialized variable was initialized. Subsidence (SUB) Package Support was added for the Subsidence Package using the NWT solver and Upstream-Weighting Flow Package. Changes to the code consist of setting storage changes related to subsidence to zero when head falls below the cell bottom. Unsaturated-Zone Flow (UZF1) Package A bug was fixed to allow the vertical hydraulic conductivity to be read when IUZFOPT=-1; that is, when unsaturated-zone storage is ignored and infiltrated water is added directly to the water table. The maximum root depth was reduced from extending through 99% of the cell thickness to 90% of the cell thickness. This improves convergence for models where the root depth extends to the bottom of the cell. Initialization was added for a few variables and checks were added for divide by zero. Surface-Water Routing (SWR1) Process Additional functionality was added to the SWR1 Process incorporated in MODFLOW-NWT. The modifications to the SWR1 Process (version 1.03) are documented in the document titled 'SWRProcess_release.txt' included with this distribution. Complete SWR input instructions for version 1.03 are provided in: '..\MODFLOW-NWT_1.0.8\doc\SWRProcessInputInstructions_v1.03.pdf' included with this distribution. SWR test problems 1-4 have been removed from the distribution but are still available at http://water.usgs.gov/ogw/swr/. SWR test problems 1-4 are surface-water only problems and do not use the functionality of MODFLOW-NWT. SWR test problem 5 is a coupled surface-water/groundwater problem that does use the functionality of MODFLOW-NWT. MODFLOW-NWT Main file At the onset of each time step, groundwater heads were set to their values one iteration prior to convergence after each times step. This change has no effect on the solution but was made for consistancy with the standard Newton solution approach. Link-MT3DMS (LMT) Package A bug was fixed to print the fluxes coming out of the bottom of dry cells. This change will have a significant effect on transport solutions for groundwater flow problems that contain dry cells with fluxes moving through them. Source files that were modified for this release include: gwf2lak7_NWT.f, gwf2sfr7_NWT.f, gwf2sub7_NWT.f, gwf2uzf1_NWT.f, MF_NWT.f, lmt7_NWT.f, gwf2swi27_NWT.f, gwf2swr7_NWT.f, gsol7.f o Version 1.0.7 01/15/2013: Lake (LAK) Package In versions of MODFLOW-NWT prior to 1.0.6, when the Lake Package was used simultaneously with the HUF Package, the effective (combined) aquifer-lakebed conductances were not calculated correctly. In this update, the Lake Package has been modified to compute effective lake-aquifer conductance solely on the basis of the user-specified value of lakebed leakance; aquifer hydraulic conductivities are not used in this calculation. An appropriate informational message is now printed after the lakebed conductances are written to the main output file. There have also been several minor changes to format statements to improve the information content of the output file. Multi-Node Well (MNW2) Package An argument mismatch was corrected between the call to the subroutine GWF2MNW27RP from MF_NWT.f to gwf2mnw27_NWT.f. A bug was fixed that affected simulations in which a composite well screen was generated when specifying multiple open intervals by elevation. In this case, some nodes were inadvertently excluded from the second interval. With the fix, all nodes of the multi-node well will be represented. Some output format adjustments were also made. If a single-node multi-node well with a nonzero specified discharge was located in an inactive cell (IBOUND=0), it could generate a divide-by-zero error and halt execution of the program. The code was fixed to preclude such a floating-point error from occurring. A new warning message is now also written to the output file if this condition is detected. If the well yield (and pumping rate) was reduced because of the seepage face calculations, an incorrect message (indicating erroneously that the cause was the head constraint) was written to the output file. The code was fixed so that possible causes of reduced pumping rates are accurately printed. The code was modified to eliminate separate processing for single-node MNW2 wells, which eliminated some inconsistencies in handling single-node wells. If the computed value for the head in single-node wells drops below the bottom of the cell, it invokes the seepage face calculation, which will reduce or eliminate the desired Q from the well. In cases where the Qdes is thereby reduced, the actual head in the well is indeterminate, but will be reported in separate MNWI output files as the limiting value of the bottom elevation of the cell together with an information note that the actual value may be lower than that reported value. If Qdes is reduced to zero, the well is deactivated and that information is reported in the MNWI optional MNW2 observation well file. At the end of the first paragraph on p. 32 of the documentation report, it states that the alternate calculations for CWC for nonvertical wells will be performed automatically when LOSSTYPE = THIEM, SKIN, or GENERAL. Unfortunately, the code did not check this condition, and erroneously performed these calculations when LOSSTYPE = SPECIFYCWC. The corrected code now uses the specified values of CWC for nonvertical wells when it is supposed to. The code automatically estimates the maximum number of nodes (NODTOT) as required for allocation of arrays. However, if a large number of horizontal wells are being simulated, or possibly for other reasons, this default estimate proves to be inadequate, a new input option has been added to allow the user to directly specify a value for NODTOT. If this is a desired option, then it can be implemented by specifying a negative value for "MNWMAX"--the first value listed in Record 1 (Line 1) of the MNW2 input data file. If this is done, then the code will assume that the very next value on that line will be the desired value of "NODTOT". The model will then reset "MNWMAX" to its absolute value. The value of "IWL2CB" will become the third value on that line, etc. The code included several coding errors in the calculation of cell-to-well conductances for nonvertical wells. There was also an error in the calculation of the length of closed casing between sequential, but non-adjacent, active nodes of a nonvertical MNW2 well (a value only used for informational purposes in the output file). All of these errors have now been fixed. A bug was fixed that affected simulations when both MNW2 and the HUF Packages were used in the same simulation in conjunction with the partial penetration correction. The variable KY is now declared as a Double Precision variable instead of being implicitly assumed (incorrectly) to be an Integer variable. A number of minor changes were made in Format statements to produce a cleaner output file. In Subroutines GWF2MNW27BCF and GWF2MNW27HUF, corrections were made to assure the correct check is made for steady-state or transient conditions and to prevent a divide by zero error if a cell is dry. A bug was fixed for the calculation of angles theta and omega for slanted wells with an orientation into the southwest directional quadrant. A bug was fixed that caused erroneous values to be written to the compact budget file (specifically, for inactive MNW wells), although the numerical solution itself was accurate, as were the budget numbers written to the main listing file. Hydrogeologic-Unit Flow (HUF) Package Small changes were made to make RHS and HCOF double precision when passed as subroutine arguments. Preconditioned Conjugate Gradient (PCG) Package Small changes were made to make RHS and HCOF double precision when passed as subroutine arguments. NWT Main File (MF_NWT.f) The version number and date were made consistent with MODFLOW-2005 1.9.01, 05/01/2012 Streamflow Routing (SFR2) Package An uninitialized variable was initialized. Gage Package Commas were added to some write statements. Modified source files: gwf2mnw27_NWT.f, MF_NWT.f, gwf2lak7_NWT.f, gwf2mnw2i7.f, nogmg.f, gwf2huf7.f, pcg7_NWT.f, gwf2sfr7_NWT.f, gwf2gag7.f o Version 1.0.6 12/05/2012: The most significant change to MODFLOW-NWT for this release was the addition of the Surface- Water Routing Process (Hughes and others, 2012). Refer to the SWR1 documentation report located in MODFLOW-NWT_1.0.6\doc for details regarding the functionality and input required for SWR1. Modifications to MODFLOW-NWT that were made to support SWR1 include the addition of two new source code files (gsol7.f and gwf2swr7_NWT.f) and the addition of call statements added to the MODFLOW-NWT main source code file (MF_NWT.f). Additionally, there are 5 test problems that use SWR1 included with this release for testing the linkage between MODFLOW-NWT and SWR1 located in MODFLOW-NWT_1.0.6\data\SWR_data_files. Refer to the readme files for more details on these test problems and other aspects of SWR1. Newton Solver Changes were made to default input values when the key word options are used. These changes reflect information gained from testing using numerous groundwater models The defaults values for "SIMPLE," "MODERATE," and "COMPLEX" are now as follows: "SIMPLE" IACL = 1, NORDER = 0, LEVEL = 3, NORTH = 5, IREDSYS = 1, All other values remain the same. "MODERATE" DBDTHETA = 0.9, NORDER = 0, LEVEL = 5, NORTH = 5, IREDSYS = 1, RRCTOLS = 0.0, IDROPTOL = 1, EPSRN = 1.0e-4, HCLOSEXMD = 1.0e-4, MXITERXMD = 50, All other values remain the same. "COMPLEX" DBDTHETA = 0.85, BACKFLAG=1, MAXBACKITER=50, BACKTOL=1.1, BACKREDUCE=0.7, LEVEL = 5, IREDSYS = 1,EPSRN = 1.0e-5, HCLOSEXMD = 1.0e-5, MXITERXMD = 100, All other values remain the same. Lake (LAK) Package Moved the read statements for specified lake bathymetry to proceed the stress period information. This change does not affect results or input instructions. Refer to the updated input instructions for using external lake bathymetry input files titled "Lake3_Input_Modified.pdf" located in \MODFLOW-NWT_1.0.6\doc for details. A new test problem was added to the MODFLOW-NWT release 1.0.6 that demonstrates the use of external bathymetry input files. This problem can be run by executing the batch file 'l1b2k_bath.bat' located in MODFLOW-NWT_1.0.6\data. Input for this test problem is located in MODFLOW-NWT_1.0.6\data\Lake_bath_example and output is included in MODFLOW-NWT_1.0.6\output_test. This test problem is identical to the MODFLOW-2005 test problem l1b2k documented in Merritt and Konikow (2000) as test problem 1, except for the use of external bathymetry input files. Thus, the results are unchanged relative to the original results produced by MODFLOW-2005. Also, a bug was found that affects simulations that use the Hydrologic-Unit Flow (HUF) Package (Anderman and Hill, 2000). Due to the difficulty in determining the effective vertical hydraulic conductivity in the HUF Package, the option to use the HUF Package with the Lake package has been removed from MODFLOW-NWT. Thus, an error will be printed and the program will stop if the HUF and LAK Packages are used together in a simulation. WELL (WEL) Package A bug was fixed to allow the reduced pumping information to be written correctly to the LIST file or separate text file. xMD Matrix Solver Some initialization issues associated with Bi-cgstab and Orthomin were resolved, and some minor changes were made. Basic (BAS) package The variables RHS and HCOF were made double precision. Upstream Weighting (UPW) Package Some variables were made double precision and the precision used in calculations was made consistent. Added an option to turn off the checking that a value is defined for all cells when parameters are used to define layer data. This option is implemented by specifying the keyword "NOPARCHECK" as one of the options in input Item 1 of the UPW input file. Streamflow Routing (SFR2) Package Some variables were made double precision and the precision used in calculations was made consistent. Some minor calculations were modified for solution efficiency. Allocation of some variables were changed to reduce the amount of random access memory (RAM) required for simulations. These changes did not affect simulation results. Initialization of some local variables was added. Unsaturated-Zone Flow (UZF1) Package Initialization of some local variables was added. Multi-Node Well (MNW2) Package Checks were added for calculations that divide real numbers. Changes were made to be consistent with version released with MODFLOW-2005 v1.9. General-Head Bounday (GHB) Package Changes were made to be consistent with version release with MODFLOW-2005 v1.9. Strongly Implicit Procedure (PCG) Package The variables RHS and HCOF were made double precision. Preconditioned Conjugate-Gradient (PCG) Package The variables RHS and HCOF were made double precision. Direct Solver (DE) Package The variables RHS and HCOF were made double precision. The Stream Observation Package (obs2str7.f) was added. o Version 1.0.5 05/14/2012: Newton Solver Changes were made to fix a bug that caused Hdry to be printed for confined cells that have heads below the cell bottom altitude. Hdry only will be printed for unconfined layers. Changes were made to default input values when the key word options are used. These changes reflect information gained from testing using numerous groundwater models The defaults values for "SIMPLE," "MODERATE," and "COMPLEX" are now as follows: "SIMPLE" IACL = 1, NORDER = 0, LEVEL = 3, NORTH = 5, IREDSYS = 1, RRCTOLS = 0.0, IDROPTOL = 1 EPSRNS = 1.0e-3, HCLOSEXMDDUM = 1.0e-4, MXITERXMD = 50, All other values remain the same. "MODERATE" IACL = 2, NORDER = 1, LEVEL = 5, NORTH = 5, IREDSYS = 1, RRCTOLS = 0.0, IDROPTOL = 1 EPSRNS = 1.0e-3, HCLOSEXMDDUM = 1.0e-4, MXITERXMD = 50, All other values remain the same. "COMPLEX" IACL = 2, NORDER = 1, LEVEL = 5, NORTH = 7, IREDSYS = 1, RRCTOLS = 0.0, IDROPTOL = 1 EPSRNS = 1.0e-5, HCLOSEXMDDUM = 1.0e-5, MXITERXMD = 50, All other values remain the same. xMD Matrix Solver Some initialization issues associated with Bi-cgstab and Orthomin were resolved, and some minor changes were made. Well Package Changes were made to report wells with reduced pumping due to dewatered conditions to a separate output file or to the main listing file. If the wells with reduced pumping rates are written to a separate output file then a file of type "DATA" must be included in the Name file with a unit number that matches the unit number specified in the Well file (following the variable PHIRAMP in Item 2b). For example, a value of PHIRAMP (see MODFLOW-NWT documentation report) of 0.3 and a unit number of 50 can be specified in the second line of the Well file using the key word "SPECIFY" as: SPECIFY 0.3 50 The following line must be added to the Name file: DATA 50 Reduced_wells.out Thus, wells with reduced pumping will be written for each time step to the output file "Reduced_wells.out" with unit number 50. For example, wells with reduced pumping will be reported to file "reduced_wells.out" as: WELLS WITH REDUCED PUMPING FOR STRESS PERIOD 6 TIME STEP 2 LAY ROW COL APPL.Q ACT.Q GW-HEAD CELL-BOT 4 91 69 -0.104400E+04 -0.104399E+04 0.208142E+04 0.205380E+04 4 91 69 -0.208800E+04 -0.208798E+04 0.208142E+04 0.205380E+04 4 91 69 -0.208800E+04 -0.208798E+04 0.208142E+04 0.205380E+04 WELLS WITH REDUCED PUMPING FOR STRESS PERIOD 6 TIME STEP 3 LAY ROW COL APPL.Q ACT.Q GW-HEAD CELL-BOT 4 91 69 -0.104400E+04 -0.967118E+03 0.207686E+04 0.205380E+04 4 91 69 -0.208800E+04 -0.193424E+04 0.207686E+04 0.205380E+04 4 91 69 -0.208800E+04 -0.193424E+04 0.207686E+04 0.205380E+04 Streamflow-Routing Package A small bug was fixed that stopped the stream width from being set equal to the wetted perimeter for ICALC=2. This change only affects the printout of stream width, and calculations dependent on stream width were made using the correct value in previous versions of SFR2. Changes were made to printout all stream reaches with streambed altitudes below the cell bottom before the model stops due to this error. Previous versions of the code printed out the first reach encountered with an altitude error and stopped. This version prints information for all reaches with altitude errors to the List file before stopping the simulation. An example of this error printout extracted from the List file is shown below. REACHES WITH ALTITUDE ERRORS: LAY ROW COL SEG REACH STR.ELEV. CELL-BOT. 1 4 57 1 57 -1.7500000 0.0000000 1 4 58 1 58 -4.2500000 0.0000000 1 4 59 1 59 -6.7500000 0.0000000 1 4 60 1 60 -9.2500000 0.0000000 1 4 61 1 61 -11.7500000 0.0000000 1 4 62 1 62 -14.2500000 0.0000000 1 4 63 1 63 -16.7500000 0.0000000 1 4 64 1 64 -19.2500000 0.0000000 1 4 65 1 65 -21.7500000 0.0000000 MODEL STOPPING DUE TO REACH ALTITUDE ERROR Multi-Node Well Package version 2 (MNW2) Changes were made to the MNW2 Package by the authors of the package (Konikow and others, 2009) for version 1.9 of MODFLOW-2005. These changes also were made to this release of MODFLOW-NWT. These changes are: A bug was fixed that affected simulations when both MNW2 and the HUF Package were used in the same simulation in conjunction with the partial penetration correction. The variable KY is now declared as a Double Precision variable instead of being implicitly assumed (incorrectly) to be an Integer variable. A number of minor changes were made in Format statements to produce a cleaner output file. In Subroutines GWF2MNW27BCF and GWF2MNW27HUF, corrections were made to assure the correct check is made for steady-state or transient conditions and to prevent a divide by zero error if a cell is dry. A bug was fixed for the calculation of angles theta and omega for slanted wells with an orientation into the southwest directional quadrant. A bug was fixed that caused erroneous values to be written to the compact budget file (specifically, for inactive MNW wells), although the numerical solution itself was accurate, as were the budget numbers written to the main listing file. Additional changes were made to the MNW2 Package for MODFLOW-NWT. These changes do not affect simulations using the MNW2 Package with solvers other than the NWT solver within MODFLOW-NWT. These changes were required to solve instabilities created by seepage face conditions within a MNW2 well and when a cell containing a MNW2 well becomes dry. The first change resets the head in a well to the bottom of the cell when it falls below the bottom of the cell for each node. For certain circumstances this was not occurring. The second change applies smoothing of the cell-to-well conductance as a cell dries. These changes may cause small changes to simulation results if these conditions arise (that is, cell drying or seepage face conditions). Drain (DRN) Package A check was added for drain altitudes set below the cell bottom. If this occurs, then an error is printed to the List file and the model is stopped. General Head Boundary (GHB) Package A check was added for head boundaries set to altitudes below the cell bottom. If this occurs, then an error is printed to the List file and the model is stopped. River (RIV) Package A check was added for river heads set to altitudes below the cell bottom. If this occurs, then an error is printed to the List file and the model is stopped. Link-MT3DMS (LMT) Package Changes were made to the Link-MT3DMS file to accommodate the MNW2 fluxes. There was a bug in the previous version of this file caused by a conflict between MNW1 and MNW2 and the header for the MT3DMS FTL file did not include MNW2. This bug was causing MT3DMS to read the FTL file incorrectly. Source files that were modified include: lmt7_NWT.f, lmt7_NWT.inc, gwf2mnw27_NWT.f, gwf2sfr7_NWT.f gwfsfrmodule_NWT.f, NWT1_xmd.f, and NWT1_xmdlib.f, MF_NWT.f, gwf2drn7_NWT.f, gwf2ghb7_NWT.f, gwf2riv7_NWT.f o Version 1.0.4 01/25/2012: Multi-Node Well (MNW1) Package and the lmt7_NWT.f and lmt7_NWT.inc source files The name of the data module for in MNW1 was changed from GWFMNWMODULE to GWFMNW1MODULE to discriminate it from the module in the MNW2 package. This change was required because data used in the lmt7_NWT.f and lmt7_NWT.inc were being confused with data in the MNW2 Package module. Both the lmt7_NWT.f and lmt7_NWT.inc files were changed to reflect the new module name. Streamflow-Routing Package Modifications were made to allow for negative NSTRM values to be specified in the SFR2 input file. In versions of SFR2 prior to the version released with MODFLOW-NWT 1.0.3, a negative value of NSTRM could be specified to simulate unsaturated flow beneath streams and for specifying streambed information by reach. However, SFR2 released with MODFLOW-NWT 1.0.3 was changed such that negative values specified for NSTRM would cause an error and the program would stop. SFR2 release with MODFLOW-NWT 1.0.4 supports both the key word option and a negative value for NSTRM. See the modified input instructions for SFR2 titled "Updated_SFR2_input_instructions_NWT.pdf" in the "doc" directory of this release. The SFR2 variable NUMSEG was added to the Module GWFSFRMODULE. Source files that were modified include: lmt7_NWT.f, lmt7_NWT.inc, gwf2mnw17_NWT.f, gwf2sfr7_NWT.f and gwfsfrmodule_NWT.f o Version 1.0.3 12/29/2011: NWT Solver The previous version of the NWT solver would take zero outer iterations if the convergence tolerances were met during the first calculation of the residuals. However, this condition could result in a poor solution for some nonlinear problems. At least one iteration is now required for all time steps. This change could affect convergence for some problems. There was a bug that was causing the red-black ordering scheme to fail (NWT solver input variable IREDSYS=1).Red-black ordering can help speed convergence for difficult problems. This change could affect convergence for some problems. The header written by the NWT solver to the main Listing file was changed slightly. The header previously titled "RMS," or if residual control is active "RMS1" and "RMS2," was changed to "L2-NORM," and "L2-NEW" and "L2-OLD," respectively. This change reflects the true definition of the residual terms because they are calculated as the L2-NORM and not the root-mean-squared error (RMS error). This change does not affect simulation results. A few small changes were made to initialize variables and remove unused variables. These changes likely would not affect simulation results. Unsaturated-Zone Flow Package A bug was found that was introduced to the UZF1 package after 1.0.1 MODFLOW-NWT release. This bug was causing UZF1 to incorrectly calculate unsaturated-zone ET, which results in much less ET than solution prior to the 1.0.2 release. This change does effect solution results as compared to the 1.0.2 version but not compared to 1.0.1 version. Three new options were added to this package to allow specification of the residual water content (THTR), to provide the option of specifying initial water content (THTI) for simulations that begin with a steady-state recharge period, and to turn off calculation of surface leakage. These changes were made for the following reasons. THTR originally was calculated internally by the UZF Package on the basis of the difference between the saturated water content (THTS) and the specific yield (SY) of the aquifer receiving recharge (Niswonger and others, 2006). However, the ability to specify THTR can be useful for some applications in which the maximum storage in the unsaturated zone (THTS-THTR)is different than the instantaneous drainage from the aquifer (SY). For these cases, specifying THTR based on external calculations provides greater flexibility for parameterizing the unsaturated zone. The option to specify THTR is activated using the key word SPECIFYTHTR. This key word is input on line 1a of the revised UZF1 input file (see revised input instructions below). Values for THTR are input following the variable THTS (see below). Originally, THTI was not specified for simulations that included both a steady-state and one or more transient stress periods. For this case, THTI was calculated internally by the code on the basis of the steady-state infiltration rate and the unsaturated-zone hydraulic properties (FINF, FKS, EPS, THTS, and THTR). However, in well-drained soils, the steady-state recharge rate corresponds to an initial water content that is too large for coarse sediments. Consequently, drainage from the unsaturated zone during the first transient stress period could result in an unrealistically large recharge rate. Additionally, the head dependency on recharge that does not occur during steady-state stress periods but that can occur during transient stress periods can cause a sudden increase in recharge during the transition between these stress periods. For these circumstances it is more realistic to specify the initial water content for the first transient stress period that follows a steady-state stress period. The option to specify THTI in simulations that include both steady-state and one or more transient stress periods is activated using the key word SPECIFYTHTI. This key word is input on line 1a of the UZF1 input file, and follows the key word SPECIFYTHTR if SPECIFYTHTR is specified. Values for THTI are input following variable THTR if it is input; otherwise, THTI follows THTS (see below). UZF1 simulates surface leakage in the uppermost active cell if groundwater head is greater than the top of this cell. Surface leakage is a nonlinear boundary condition that can slow model convergence, and in some cases it is beneficial to inactive this boundary condition. Surface leakage is inactivated by the key word NOSURFLEAK that is specified on line 1a of the UZF1 input file. This key word is specified following the key word SPECIFYTHTI (if SPECIFYTHTI is specified). If the key word NOSURFLEAK is specified then water will not be removed from the uppermost active cell when groundwater head rises above the top of this cell. In addition to these new input options, the default value for SURFDEP was changed. In prior versions of UZF1, if the value for SURFDEP was specified as 0, it was reset internally to 1.0. In this new version of UZF1, SURFDEP is reset to 1.0x10-6. This change could affect simulated results if the value of SURDFEP was specified as 0 in the input file. Modified input instructions for the UZF1 input file for specifying optional key words 0. Data: [#Text] (As in previous versions of the code) 1a. Data: [SPECIFYTHTR] [SPECIFYTHTI] [NOSURFLEAK] SPECIFYTHTR key word for specifying optional input variable THTR. SPECIFYTHTI key word for specifying optional input variable THTI. NOSURFLEAK key word for inactivating calculation of surface leakage. Note that the keywords must be entered in the order shown. For example, if keywords SPECIFYTHTR and NOSURFLEAK are specified, then SPECIFYTHTR must precede NOSURFLEAK. Alternatively, if only SPECIFYTHTI is specified, then the other two keywords should not be specified. 1b. Data: NUZTOP IUZFOPT IRUNFLG IETFLG IUZFCB1 IUZFCB2 NTRAIL2 NSETS2 NUZGAG SURFDEP NUZTOP (As in previous versions of the code) Items 2-5: As in previous versions of the code. 6a. Data: THTS (NCOL, NROW) -- U2DREL Refer to UZF1 documentation report (Niswonger and other, 2006) for definition of THTS. 6b. Data: [THTR (NCOL, NROW)] -- U2DREL THTR An array of positive real values used to define the residual water content for each vertical column of cells in units of volume of water to total volume (L3L-3). THTR is the irreducible water content and the unsaturated water content cannot drain to water contents less than THTR. This variable is not included unless the key word SPECIFYTHTR is specified. 7. Data: [THTI (NCOL, NROW)] -- U2DREL Refer to UZF1 documentation report (Niswonger and others, 2006) for definition of THTI. This variable is not included for simulations with a steady-state stress period unless the key word SPECIFYTHTI is specified. Items 8-16: As in previous versions of the code. Streamflow-Routing Package Modifications were made to correct the effects of lakes inundating stream cells. Seepage in inundated stream cells was not being set to zero in the budget routine. This change could affect mass-balance errors for models that include lakes that grow and inundate stream cells. Modifications were made to correct a bug that occurred when the specific yield specified in the Block-Centered Flow (BCF7) Package was used for calculating the residual water content for the unsaturated zone beneath streams. Changes were made such that the code searches through model layers to find the corresponding layer that stores the specific yield for the case when the BCF7 input variable LAYCON is 2 or 3. Modifications were made to add a smoothing function for the calculation of wetted channel area when the SFR2 input variable ICALC is set to 1 (constant stream width). In previous versions of the package, a discontinuity could form if the stream channel became dry during the outer iteration of a time step, because stream width would oscillate between the specified value and zero. To avoid such conditions, a smoothing function was used in SFR2 to smooth the transition between the constant (specified) stream width and a zero stream width over the interval between zero depth and 1.0x10-5 (units defined on the basis of the DIS file input variable LENUNI). Smoothing is applied automatically if ICALC=1 and the stream depth is between zero and 1.0x10-5 and users should refer to the calculated stream depths to determine if smoothing was applied to a particular reach in a simulation. SFR2 smooths the constant- to-zero-width transition using the same smoothing function that is used by the WEL Package to decrease specified pumping rates to zero when unconfined cells are dry (see Niswonger and others, 2011, p. 14). Lake Package A format statement was modified to allow proper printout of the number of dry lake cells to the main Listing file. This change does not affect simulation results. Source files that were modified include: NWT1_solver.f,NWT1_xmd.f,gwf2uzf1_NWT.f,gwfuzfmodule_NWT.f, gwf2sfr7_NWT.f,gwfsfrmodule_NWT.f,gwf2lak7_NWT.f, NWT1_module.f, NWT1_gmres.f90, MF_NWT.f, gwfsfrmodule_NWT.f, gwf2wel7_NWT.f, gwf2upw1.f, gwf2sub7.f, and gwf2hfb7_NWT.f. o Version 1.0.2 10/03/2011: MODFLOW-NWT was compiled again because the last release of the program required a large amount of RAM to run models. The reason for this unnecessary memory usage was not determined but it is assumed that the issue was related to the compiler. The new compiled version requires much less memory. For example, a model that required 2 Gb of memory to run now requires 120 Mb. This new memory usage is consistent with what is required by MODFLOW-2005. Changes were made to the MODFLOW-NWT source files such that residual-control (backtracking) iterations are no longer counted as outer iterations. This change affects the NWT solver iteration information that is printed to the main List file. Another change was made to stop model execution when the convergence criteria are not met. In previous versions of MODFLOW-NWT, the model continued to solve after the model failed to converge. A keyword option was added to the NWT Package input file to allow the user to control whether the model will stop after convergence failure. If the keyword "CONTINUE" is added following the input variable OPTIONS then the model will not stop after convergence failure. If this keyword is not specified then the model will stop after convergence failure. For example, the key word "CONTINUE" can be added to the first line of the NWT input file as: 1.0E-4 1. 500 1.0e-5 2 1 1 SPECIFIED CONTINUE 0.9 0.00001 0.000 0.10 0 20 2.0 0.6 For this input option, if the model fails to converge during a time step then it will continue to the next time step. However, if the "CONTINUE" is removed from this line, then the model will stop after convergence failure. Another change was made to fix a bug that caused the model to print a blank line to the main List file rather than correctly printing the string "SPECIFIC YIELD FOR LAYER..." This change does not affect model results. Another change was made to modify how groundwater heads are reset to the layer- bottom altitude when the calculated heads are below the bottom of the deepest layer. In the previous version of the code there was a condition when heads could be reset to an incorrect value, and this caused small oscillations in some problems. This change may have a small affect to improve model convergence if the NWT input variable IBOTAV is equal to 1. Another change was made to fix oscillatory behavior caused by a feature in the MNW2 and MNW1 packages. The original versions of these packages switched between a specified flux and a head-dependent boundary conditions during outer iterations. The MNW1 and MNW2 packages were modified such that the boundary condition is fixed as a head-dependent boundary condition. A small change was made in the XMD source file to allow for double precision compilations. This change was to change the statement from real*4 to real. Small changes were made to the UZF1 and SFR2 source files to improve standard output options. Changes were made to the MT3DMS linker files (lmt7_NWT.f, and lmt7_NWT.inc) to include water balance information for the MNW2 and MNW1 packages. Source files that were modified include: MF_NWT.f, NWT1_solver.f, NWT1_xmd.f, gwf2MNW27_NWT.f, gwf2MNW17_NWT.f, gwf2sfr7_NWT.f, gwf2upw1.f, gwfsfrmodule_NWT.f, gwf2uzf1_NWT.f, gwfuzfmodule_NWT.f, lmt7_NWT.f, and lmt7_NWT.inc. o Version 1.0.1 05/27/2011: Changes were made to the MODFLOW-NWT source files to correct an error in how the Horizontal-Flow (HFB) Barrier Package was implemented in MODFLOW-NWT. Additionally, changes were made to deallocate arrays in the XMD and GMRES solvers at the termination of a simulation. Source files that were modified include: MF_NWT.f, NWT1_solver.f, NWT1_xmd.f, gwf2hfb7_NWT.f, NWT1_gmres.f90, and gwf2bas7.f o Version 1.0 05/11/2011: This version is the initial release. However, a number of changes were required for four additional MODFLOW-2005 Packages that were not explained in the MODFLOW-NWT documentation report (Niswonger and others, 2011). The modified packages include the Horizontal-Flow Barrier, Lake, and Multi-Node Well Packages versions 1 and 2. Changes to the Horizontal-Flow (HFB) Barrier Package were made to the subroutine GWF2HFB7FM. If the Newton method is used in MODFLOW-NWT, then horizontal conductances do not include the aquifer thickness for unconfined conditions. Thus, equations were changed such that the conductances modified in subroutine GWF2HFB7FM are not multiplied by the aquifer thickness. This change required that a new argument be added to the call statement for subroutine GWF2HFB7FM to include the file unit number for the UPW Package. Changes to the Lake (LAK) Package were made to the subroutine GWF2LAK7RP and a new subroutine was added to the LAK Package called SGWF2LAK7UPW1RPS for calculating lakebed conductance values. This change required that a new argument be added to the call statement for subroutine GWF2LAK7RP to include the file unit number for the UPW Package. Changes to Versions 1 and 2 of the multi-Node Well (MNW1 and MNW2) Packages were made to the subroutines GWF2MNW17RP and GWF2MNW27RP. A new subroutine was added to the MNW1 and MNW2 Packages called CEL2WELUPW for calculating well conductance values. A new argument was added to the call statement for subroutines GWF2MNW17RP and GWF2MNW27RP to include the file unit number for the UPW Package. References Anderman, E. R., & Hill, M. C. (2000). MODFLOW-2000: The US Geological Survey Modular Ground-water Model--documentation of the Hydrogeologic-Unit Flow (HUF) Package. US Department of the Interior, US Geological Survey. Available online at http://pubs.er.usgs.gov/publication/ofr03347 Bakker, Mark, Schaars, Frans, Hughes, J.D., Langevin, C.D., and Dausman, A.M., 2013, Documentation of the seawater intrusion (SWI2) package for MODFLOW: U.S. Geological Survey Techniques and Methods, book 6, chap. A46, 47 p. Available online at http://pubs.er.usgs.gov/publication/tm6A46 Bedekar, V., Morway, E.D., Langevin, C.D., Tonkin, M.T., 2016, MT3D-USGS Version 1: A U.S. Geological Survey Release of MT3DMS Updated with New and Expanded Transport Capabilities for MODFLOW, U.S. Geological Survey, Techniques and Methods 6-A53 variously p. Hughes, J.D., Langevin, C.D., Chartier, K.L., and White, J.T., 2012, Documentation of the Surface-Water Routing (SWR1) Process for modeling surface-water flow with the U.S. Geological Survey Modular Ground-Water Model (MODFLOW-2005): U.S. Geological Survey Techniques and Methods, book 6, chap. A40 (Version 1.0), 113 p. Available online at http://pubs.er.usgs.gov/publication/tm6A40 Konikow, L.F., Hornberger, G.Z., Halford, K.J., and Hanson, R.T., 2009, Revised multi-node well (MNW2) package for MODFLOW ground-water flow model: U.S. Geological Survey Techniques and Methods 6–A30, 67 p. Available online at http://pubs.er.usgs.gov/publication/tm6A30 Niswonger, R.G., Panday, Sorab, and Ibaraki, Motomu, 2011, MODFLOW-NWT, A Newton formulation for MODFLOW-2005: U.S. Geological Survey Techniques and Methods 6-A37, 44 p. Available online at http://pubs.er.usgs.gov/publication/tm6A37 Niswonger, R.G., Prudic, D.E., and Regan, R.S., 2006, Documentation of the Unsaturated-Zone Flow (UZF1) Package for modeling unsaturated flow between the land surface and the water table with MODFLOW-2005: U.S. Geological Survey Techniques and Methods 6-A19, 62 p. Available online at http://pubs.er.usgs.gov/publication/tm6A19 Niswonger, R. G. (2020). An Agricultural Water Use Package for MODFLOW and GSFLOW. Environmental Modelling & Software, 104617. https://www.sciencedirect.com/science/article/pii/S1364815219305080