Overview of Center for Infrastructure Modeling and Management NCIMM for #SWMM6

The Center for Infrastructure Modeling and Management has been envisaged as an entity that will preserve, promote and extend the US EPA SWMM and EPANETsoftware applications, and potentially other applications.  SWMM and EPANET are two foremost tools for assessment of watershed hydrology and pipe network hydraulics, respectively, and are perhaps unparalleled examples of technical community accomplishments conducted as a partnership of stakeholders from all avenues of professional practice.  Build over decades with participation from regulatory, academic, consulting,  owner/operator, vendors, and institutional support under the continuing leadership of the US EPA, these tools are used around the world as 'first choice' software applications in many, many situations.  They will remain as first choice tools for years to come.  A key question, therefore, is how to ensure these hallmarks of professional practice can be maintained, promoted, and developed going forward.  The expectations of users has evolved, needs have evolved, so the tools and their support mechanisms must evolve along with them. 

A visual view of the CDM SWMM5 GUI Circa 2007

This is from a working QA/QC version of SWMM 5.022 showing some feedback from the Simulation Run to the User’s Run Status dialog:

1. You can see the total inflow, outflow, storage and flooding in a graph and text box in the Run Status dialog
2. You can see a running estimate of the Continuity Error (CE) and
3. It stays on the screen when the run as finished.
4. The following images show some examples. 

This is how node interface files works in #SWMM5, Caveats and Tips

The routine readNewIfaceValues reads a line interface flows in SWMM5.  It is a string parser, finds tokens and creates dates, times and flows from the tokens.  It is important to have the correct format for your line else the tokens will not be correctly converted to integers and doubles. Further, it is important to have more than one date/time for each node as the SWMM5 engine interpolates the flow values for each node during the simulation.  One time value or time values out of the simulation date/times will result in no flows.

void readNewIfaceValues()

//

//  Input:   none

//  Output:  none

//  Purpose: reads data from inflows interface file for next date.

//

{

    int    i, j;

    char*  s;

    int    yr = 0, mon = 0, day = 0,

                hr = 0, min = 0, sec = 0;   // year, month, day, hour, minute, second

    char   line[MAXLINE+1];            // line from interface file

    // --- read a line for each interface node

    NewIfaceDate = NO_DATE;

    for (i=0; i<NumIfaceNodes; i++)

    {

        if ( feof(Finflows.file) ) return;

        fgets(line, MAXLINE, Finflows.file);

        // --- parse date & time from line

        if ( strtok(line, SEPSTR) == NULL ) return;

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        yr  = atoi(s);

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        mon = atoi(s);

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        day = atoi(s);

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        hr  = atoi(s);

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        min = atoi(s);

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        sec = atoi(s);

        // --- parse flow value

        s = strtok(NULL, SEPSTR);

        if ( s == NULL ) return;

        NewIfaceValues[i][0] = atof(s) / Qcf[IfaceFlowUnits];

        // --- parse pollutant values

        for (j=1; j<=NumIfacePolluts; j++)

        {

            s = strtok(NULL, SEPSTR);

            if ( s == NULL ) return;

            NewIfaceValues[i][j] = atof(s);

        }

    }

A Visual Studio Compiler for #SWMM5 Note

A Visual Studio Compiler for #SWMM5 note.  If you have many many copies of the SWMM5 code on your PC you can rename the Visual Studio project files from the default SWMM5 names to a more meaningful version name (Figure 2) and using the DLL Properties/General/Output file change the SWMM5.DLL creation directory (Figure 1). 

Figure 1.  DLL Properties/General/Output file change the SWMM5.DLL creation directory

Figure 2.  Visual Studio project files from the default SWMM5 names to a more meaningful version name 

TeeChart Standard v2017 VCL/FMX works with Delphi 10.1 Berlin and 10.2 Tokyo Starter, #SWMM5

For those interested in using the latest Delphi GUI with SWMM 5.1.012 and before.  TeeChart is the needed graphics.  The latest TeeChart Standard v2017 VCL/FMX works with Delphi 10.1 Berlin and 10.2 Tokyo Starter, but there are not specific installers, you should install theTeeChart Standard Starter packages manually following the steps below:

1- Open the Components->Installation Packages and add the DclFMXTeeStd924.bpl and DclTeeStd924.bpl from installation folder, concretely you find the packagesin the %Program Files (x86)%\Steema Software\Steema TeeChart Standard VCL FMX 2016.18\Delphi24\bin

The below image shows you where:

2- Then you should see in the Install Packages window the results below:

3- Create a simple application and run it.

Note the steps are for Rad Studio 10.1 Berlin, but you should follow the same steps to install Teechart Standard in Rad Studio 10.2 Tokyo.

EPA SWMM 5.1 RELEASE NOTES (in a different Format)

======================================================

EPA SWMM 5.1 RELEASE NOTES

======================================================
This file contains information concerning Version 5.1 of the EPA Storm Water Management Model (SWMM). A complete Users Manual as well as full source code and other updates/Technical Manuals are available at

www.epa.gov/nrmrl/wswrd/wq/models/swmm

I posted this on the SWMM Group on LinkedIn

World Class Software Documentation for SWMM5 from Lew Rossman and Wayne Huber (Hydrology)

I have noticed based on email questions and postings to the SWMM LIst Sever (a great resource hosted by CHI, Inc.) that many SWMM 5 users do not know about the really outstanding documentation on SWMM 5 posted on the EPA Website https://www.epa.gov/water-research/storm-water-management-model-swmm It consists of two now and in the near future three volumes on Hydrology, Water Quality, LID’s and SuDs and Hydraulics. The documentation is fantastically complete with detailed background on the theory, process parameters and completely worked out examples for all of the processes in SWMM5. It is truly an outstanding aid to modelers and modellers worldwide. It would benefit you to read them (if you have not already downloaded the PDF files). 

======================================================

INSTALLATION

======================================================
To install EPA SWMM 5.1 run the setup program epaswmm5_setup.exe. It will place the following files into the application folder you designate:
  epaswmm5.exe   -- the Windows user interface for SWMM
  epaswmm5.chm  -- the SWMM 5 help file
  swmm5.dll          -- the SWMM 5 computational engine
  swmm5.exe        -- the command line version of SWMM 5
  tutorial.chm        -- an online tutorial for SWMM 5
  notes.txt             -- this file (the source of this blog)

The setup program will also create a Start Menu group named EPA SWMM 5.1 and will register SWMM 5 with Windows. This will allow you to use the Add or Remove Programs option of the Windows Control Panel to un-install EPA SWMM 5.1.

======================================================

EXAMPLE DATA SETS

======================================================
Several example data sets have been included with this package. They are placed in a sub-folder named EPA SWMM Projects\Examples in your My Documents folder. Each example consists of a .INP file that holds the model data and a .TXT file with suggestions on running it.

* EXAMPLE1.INP models runoff quantity and quality from a small watershed and its routing through  a network of storm sewers. It can be run in either single event mode or in continuous mode using the companion rainfall file.

* EXAMPLE2.INP is Example 1 of the 1988 EXTRAN Users Manual. It illustrates how SWMM 5 can graphically compare its results to observed data stored in a text file.

* EXAMPLE3.INP illustrates the use of the rule-  based controls feature in SWMM 5 for simulating real-time control.

* EXAMPLE4.INP shows how the LID controls feature  in SWMM 5 was used to reduce runoff produced from   a 29 acre mixed development site.

======================================================

TERMS OF USE

======================================================
EPA SWMM 5 is public domain software that may be freely copied and distributed.

======================================================

DISCLAIMER

======================================================
The software product is provided on an "as-is" basis. US EPA makes no representations or warranties of any kind and expressly disclaim all other warranties express or implied, including, without limitation, warranties
of merchantability or fitness for a particular purpose. Although care has been used in preparing the software product, US EPA disclaim all liability for its accuracy or completeness, and the user shall be solely responsible for the selection, use,efficiency and suitability of the software product. Any person who uses this product does so at his sole risk and without liability to US EPA.
US EPA shall have no liability to users for the infringement of proprietary rights by the software product or any portion thereof.

Conduits in #SWMM5

Conduits

Conduits are pipes or channels that move water from one node to another in the conveyance system. Their cross-sectional shapes can be selected from a variety of standard open and closed geometries as listed in the following table. Irregular natural cross-section shapes and Dummy links are also supported.

SWMM5 Lnk Shape

Conduits are pipes or channels that move water from one node to another in the conveyance system. Their cross-sectional shapes can be selected from a variety of standard open and closed geometries as listed in Table 3-1.

Most open channels can be represented with a rectangular, trapezoidal, or user-defined irregular cross-section shape. For the latter, a Transect object is used to define how depth varies with distance across the cross-section (see Section 3.3.5 below). Most new drainage and sewer pipes are circular while culverts typically have elliptical or arch shapes. Elliptical and Arch pipes come in standard sizes that are listed in at the bottom of this page. The Filled Circular shape allows the bottom of a circular pipe to be filled with sediment and thus limit its flow capacity. The Custom Closed Shape allows any closed geometrical shape that is symmetrical about the center line to be defined by supplying a Shape Curve for the cross section (see Section3.3.11 below).

SWMM uses the Manning equation to express the relationship between flow rate (Q), crosssectional area (A), hydraulic radius (R), and slope (S) in all conduits. For standard U.S. units,

where n is the Manning roughness coefficient. The slope S is interpreted as either the conduit slope or the friction slope (i.e., head loss per unit length), depending on the flow routing method used. 

For pipes with Circular Force Main cross-sections either the Hazen-Williams or Darcy-Weisbach formula is used in place of the Manning equation for fully pressurized flow. For U.S. units the Hazen-Williams formula is:

where C is the Hazen-Williams C-factor which varies inversely with surface roughness and is supplied as one of the cross-section’s parameters. The Darcy-Weisbach formula is:

where g is the acceleration of gravity and f is the Darcy-Weisbach friction factor. For turbulent flow, the latter is determined from the height of the roughness elements on the walls of the pipe (supplied as an input parameter) and the flow’s Reynolds Number using the Colebrook-White equation. The choice of which equation to use is a user-supplied option.

A conduit does not have to be assigned a Force Main shape for it to pressurize. Any of the closed cross-section shapes can potentially pressurize and thus function as force mains that use the Manning equation to compute friction losses. 

A constant rate of exfiltration of water along the length of the conduit can be modeled by supplying a Seepage Rate value (in/hr or mm/hr). This only accounts for seepage losses, not infiltration of rainfall dependent groundwater. The latter can be modeled using SWMM’s RDII feature (see Section 3.3.6).

The principal input parameters for conduits are:

• names of the inlet and outlet nodes
• offset heights of the conduit above the inlet and outlet node inverts
• conduit length
• Manning’s roughness
• cross-sectional geometry
• entrance/exit losses
• presence of a flap gate to prevent reverse flow.

A conduit can also be designated to act as a culvert (see Figure 3-2) if a Culvert Inlet Geometry code number is assigned to it. These code numbers are listed in Appendix A.10. Culvert conduits are checked continuously during dynamic wave flow routing to see if they operate under Inlet Control as defined in the Federal Highway Administration’s publication Hydraulic Design of Highway Culverts Third Edition (Publication No. FHWA-HIF-12-026, April 2012). Under inlet control a culvert obeys a particular flow versus inlet depth rating curve whose shape depends on the culvert’s shape, size, slope, and inlet geometry. 

Flow Regulators

Flow Regulators are structures or devices used to control and divert flows within a conveyance system. They are typically used to:

• control releases from storage facilities
• prevent unacceptable surcharging
• divert flow to treatment facilities and interceptors

 InfoSWMM H₂OMap SWMM InfoSWMM SA  can model the following types of flow regulators:

• Orifices
• Weirs
• Outlets

The following Tables are copied from the EPA Manual on SWMM (Hydraulics) II

SWMM5_NR_ITERATIVE Fortran Routine from 2004

      SUBROUTINE SWMM5_NR_ITERATIVE
C EXTRAN BLOCK test for SWMM 5 beta solution - 12/12/2002
      INCLUDE 'TAPES.INC'
      INCLUDE 'STIMER.INC'
      INCLUDE 'BD.INC'
      INCLUDE 'BND.INC'
      INCLUDE 'HYFLOW.INC'
      INCLUDE 'CONTR.INC'
      INCLUDE 'JUNC.INC'
      INCLUDE 'PIPE.INC'
      INCLUDE 'TIDE.INC'
      INCLUDE 'OUT.INC'
      INCLUDE 'ORF.INC'
      INCLUDE 'WEIR.INC'
      INCLUDE 'FLODAT.INC'
      DOUBLE PRECISION AKON,QNEW,DELQ1,DELQ2,DELQ3,DELQ4,DELQ5
DIMENSION        AS1(NEE)
DOUBLE PRECISION df,f,n_omega
integer          good_nodes
C=======================================================================
C     STORE OLD TIME STEP FLOW VALUES
C=======================================================================
DO  N        = 1,NTL
      QO(N)        = Q(N)
      AT(N)        = A(N)
      VT(N)        = V(N)
enddo
C=======================================================================
C     INITIALIZE CONTINUITY PARAMETERS
C=======================================================================
DO J            = 1,NJ
if(othercom(63).eq.1) then
                     Y(J)      = Y(J) + 0.5 * ( YEX2(J) - 
     +                                  YEX1(J) + YEX1(J) - YO(J))
                     IF(Y(J).LT.FUDGE)      Y(J)=FUDGE
                     IF(Y(J).GT.SURELEV(J)) Y(J)=SURELEV(J)-Z(J)
                     yex2(j)   = yex1(j)
                     yex1(j)   = YO(j)
                     endif
cred  beginning time step value of the node area
asold(j)    = as(j)
      YO(J)       = Y(J) 
      GoodNode(j) = .FALSE.
enddo
good_nodes  = 0
      omega       = input_omega
      n_omega     = node_omega
loop_count  = 0
C=======================================================================
C     HALF-STEP AREA, RADIUS : VELOCITY
C     FULL-STEP FLOW
C=======================================================================
big_loop:  DO while(good_nodes.lt.nj.and.loop_count.le.itmax)
loop_count = loop_count + 1
      if(loop_count.ge.itmax-5) then
        omega   = 0.50 * omega
        n_omega = 0.50 * n_omega 
        endif

DO J           = 1,NJ
      AS(J)          = AMEN
      AS1(J)         = 0.0
      SUMQ(J)        = QIN(J)
      SUMQS(J)       = QIN(J)
      SUMAL(J)       = 0.0
enddo
CIM   FIRST COMPUTE GATED ORIFICE PARAMETERS
      CALL OGATES(DELT,Y,V)
c
      flow_loop: DO    N      = 1,NTC
      NL            = NJUNC(N,1)
      NH            = NJUNC(N,2)
C=======================================================================
      H(N,1)   = AMAX1(Y(NL) + Z(NL),ZU(N))
      H(N,2)   = AMAX1(Y(NH) + Z(NH),ZD(N))
      CALL nhead(N,NL,NH,H(N,1),H(N,2),Q(N),A(N),V(N),HRAD,
     +           ANH,ANL,RNL,RNH,YNL,YNH,width,IDOIT,LINK(N),AS1)
cred  bypass loop for nodes already converged
      bypass_loop: if(loop_count.gt.2.and.
     +  goodnode(nl).eq..TRUE..and.goodnode(nh).eq..TRUE.) then
bypass = bypass + 1.0
else
IF(HRAD.GT.HMAX(N)) HMAX(N) = HRAD
      IF(A(N).GT.AMAX(N)) AMAX(N) = A(N)
cred  save information for the culvert classification
      HRLAST(N)  = HRAD 
vup(n)     = anl
vdn(n)     = anh
if(loop_count.eq.1) then
                   aup(n)   = anl
                   rup(n)   = rnl
                   rmd(n)   = hrad
                   endif
      positive_flow: IF(IDOIT.gt.0) THEN
c
c       Q/ANH = velocity at downstream end used for exit loss
c       Q/ANL = velocity at upstream end used for entrance loss
c       The loss = 1/2*K*V^2/g is factored into momentum
c       equation similarly to the friction slope
c       The loss momentum term = g*A*[1/2*K*V^2/g]
c                              = g*A*[1/2*K*Q/A*Q/A/g]
C                               =g  *[1/2*K*|Q*A|  /g] * Q
C                               =    [1/2*K*|Q*A|    ] * Q
c       DELQ5 is sum of losses
c
cred  calculate the froude number for every conduit
c
      DELH      = H(N,1) - H(N,2)
      DELZP     = ZU(N)  - ZD(N)
      DIFF_mid  = 0.5 * (H(N,1) - ZU(N) + H(N,2) - ZD(N))
      IF(DIFF_mid.GT.0.0) THEN
           FROUDE_MID = ABS(Q(N))/A(N)/SQRT(GRVT*(DIFF_mid))
                  ELSE
           FROUDE_MID = 0.0
           ENDIF
      bfactor = 0.0
if(FROUDE_MID.ge.1.0) THEN
           bfactor = 0.0
           elseif(froude_mid.gt.0.9) then
bfactor = 1.0 - (10.0*FROUDE_MID-9.0)
           endif
cred  test for zero sloped conduits
      if(delzp.eq.0.0)     then
                    del_ratio   = delh/0.001
                    else
                    del_ratio   = delh/delzp
                    endif
cred  swmm 4 definition for normal flow
      if(del_ratio.le.1.0) then 
                    delfactor = 0.0
cred                       swmm 5 transition definition
                    else if(del_ratio.gt.1.10) then
delfactor = 1.0
                    else
                    delfactor = 10.0 * (del_ratio-1.0)
                      endif
      DELQ5          = 0.0
      IF(ANH.NE.0.0)   DELQ5 = DELQ5 + 0.5 * ABS(Q(N)/ANH)*ENTK(N)
      IF(ANL.NE.0.0)   DELQ5 = DELQ5 + 0.5 * ABS(Q(N)/ANL)*EXITK(N)
      IF(A(N).NE.0.0)  DELQ5 = DELQ5 + 0.5 * ABS(Q(N)/A(N))*OTHERK(N)
      DELQ5 =  DELQ5 * DELT/LEN(N)
c
      DELQ4  = DELT*V(N)*(ANH-ANL)/A(N)/LEN(N)                                          
cred  the mean area travels from the midpoint to the upstream area
cred  as the value of delh/delzp changes
      area_mean =  wt*anl + wd*aup(n)  + ( wt*a(n) + wd*at(n) - 
     +             wt*anl - wd*aup(n)) *   delfactor

      DELQ2  = DELT*GRVT*area_mean*((H(N,2) - H(N,1)))/LEN(N)
      DELQ3  = 2.0*(A(N)-AT(N))/A(N)
cred  the mean hydraulic radius travels from the midpoint to the 
cred  upstream hydraulic radius as the value of delh/delzp changes
      hrad_mean =  wt*rnl + wd*rup(n)  + ( wt*hrad + wd*rmd(n) - 
     +             wt*rnl - wd*rup(n)) *   delfactor
      DELQ1  = DELT*(ROUGH(N)/hrad_mean**1.33333)*ABS(V(N))

QNEW  = QO(N) - delq2
      AKON  = DELQ1 + DELQ5 - delq4*bfactor - DELQ3*bfactor
cred  Newton-Raphson iteration
      F         = Q(N) * ( 1.0 + akon) - qnew
DF        = 1.0 + akon
Q(N)      = Q(N) - omega*f/df
      DQDH      = 1.0/(1.0+AKON)*GRVT*DELT*A(N)/LEN2(N)
dqdh_old(n) = dqdh
C=======================================================================
C     DO NOT ALLOW A FLOW REVERSAL IN ONE TIME STEP
C=======================================================================
      DIRQT = SIGN(1.0,QO(N))
      DIRQ  = SIGN(1.0,Q(N))
      IF(DIRQT/DIRQ.LT.0.0) Q(N) = 0.001*DIRQ

v(n)   = q(n)/A(N)
C=======================================================================
C     COMPUTE CONTINUITY PARAMETERS
C=======================================================================
      SELECT CASE (INGATE(N))
      CASE (1)
      Q(N) = AMAX1(Q(N),0.0)
      CASE (2)
      Q(N) = AMIN1(Q(N),0.0)
      END SELECT

      IF (NKLASS(N).LE.21) THEN
                      Q(N) = AMAX1(STHETA(N),Q(N))
                           Q(N) = AMIN1(SPHI(N),Q(N))
                    ENDIF
endif positive_flow
      endif bypass_loop
      SUMQS(NL) = SUMQS(NL) - Q(N)*barrels(n)
      SUMAL(NL) = SUMAL(NL) + dqdh_old(n)*barrels(n)
      SUMQS(NH) = SUMQS(NH) + Q(N)*barrels(n)
      SUMAL(NH) = SUMAL(NH) + dqdh_old(n)*barrels(n)

      SUMQ(NL)   = SUMQ(NL)  - WT*Q(N)*barrels(n)  - WD*QO(N)*barrels(n)   
      SUMQ(NH)   = SUMQ(NH)  + WT*Q(N)*barrels(n)  + WD*QO(N)*barrels(n) 
      ENDDO  flow_loop
C=======================================================================
C     SET FULL STEP OUTFLOWS AND INTERNAL TRANSFERS
C=======================================================================
      CALL BOUND(Y,Y,Q,TIME,DELT)
C=======================================================================
      N1       = NTC+1
      DO 370 N = N1,NTL
      NL       = NJUNC(N,1)
      NH       = NJUNC(N,2)
      IF(ABS(Q(N)).LT.1E-10) Q(N) = 0.0
C=======================================================================
      SUMQ(NL)  = SUMQ(NL)  - WT*Q(N)*barrels(n)  - WD*QO(N)*barrels(n) 
      SUMQS(NL) = SUMQS(NL) - Q(N)*barrels(n)    
      IF(NH.NE.0) THEN
         SUMQ(NH)  = SUMQ(NH) + WT*Q(N)*barrels(n) + WD*QO(N)*barrels(n) 
         SUMQS(NH) = SUMQS(NH) + Q(N)*barrels(n)    
         ENDIF
  370 CONTINUE
C=======================================================================
C     CALCULATE THE FULL-STEP HEAD
C=======================================================================
      DO J = 1, NJ
         AS(J)    = AS(J)  + AS1(J) 
      ENDDO
      DO  J   = 1,NJ
      IF(JSKIP(J).le.0) then
C=======================================================================
cred    time weighted area average
      if(othercom(62).eq.1) then
                average_area = WT*as(j) + WD*asold(j)
                else
                average_area = as(j)
                endif
cred    Newton-Raphson iteration
        yt(j)     = y(j)
 if(y(j).le.ycrown(j)) then
 F         =  Y(J) * average_area - YO(J) * average_area   
     +                    - sumq(j)*DELT
   DF        =  average_area 
        ASFULL(J) = AS(J)
 else
   DF        =  sumal(j) 
        IF(Y(J).LT.1.25*YCROWN(J))  DF = DF + 
     +   (ASFULL(J)/DELT-SUMAL(J))*EXP(-15.*(Y(J)-YCROWN(J))/YCROWN(J))
cred       correction from SWMM 5 QA testing - 11/12/2004
cred       if a large delt or if there is a large value of sumal(j)
cred       - usually from a very large conduit connected to the current node - 
cred       the expression  ASFULL(J)/DELT-SUMAL(J) may be negative 
cred       invalidating the whole concept of a "transition" slot
           if(Y(J).LT.1.25*YCROWN(J).le.ASFULL(J)/DELT.le.SUMAL(J)) 
     +     denom = sumal(j)
        CORR     = 1.00
        IF(NCHAN(J,2).EQ.0) CORR  = 0.60
 F       =  Y(J) * DF - YO(J) * DF - CORR * sumqs(j)
cred    WRITE(*,*) ajun(j),df,f,y(j),ycrown(j),sumal(j)
 endif
     Y(J)    =  Y(J) - n_omega*F/DF
        IF(Y(J).LT.0.0) Y(J) = 0.0
        IF((Y(J)+Z(J)).GT.SURELEV(J)) Y(J) = SURELEV(J)-Z(J)
 endif
       enddo

cred  converge until all nodes meet the convergence criteria
good_nodes = 0 
error_node = 0
i_was_bad  = 1 
DO j       = 1,nj
if(jskip(j).le.0) then
                  if(abs(y(j)-yt(j)).le.node_toler) then
                                     good_nodes = good_nodes + 1
                                     GoodNode(j) = .TRUE.
                              endif
                   if(abs(y(j)-yt(j)).gt.error_node) then
       error_node = abs(y(j)-yt(j))
                          i_was_bad  = j
                   endif
                 else
                  good_nodes  = good_nodes + 1
           GoodNode(j) = .TRUE.
endif
enddo 
c     write(*,669) loop_count,nj-good_nodes,error_node,
c    +               delt,ajun(i_was_bad),y(i_was_bad),  
c    +               sumq(i_was_bad),omega
669   format(2I6,F10.4,F8.2,1x,a12,3F9.3)
enddo big_loop

cred  culvert information - 8/6/2002
cred  culvert information for the culvert comparison file - 8/6/2002
cred  culvert information - 8/6/2002
      if(othercom(90).eq.1) then
       do n = 1,nc
       if(nklass(n).eq.1.or.nklass(n).eq.21) then
       if(abs(q(n)).gt.culvert_loss(n,5).and.abs(v(n)).lt.20.0)then
                    HW          = H(N,1)   - ZU(N)
                    TW          = H(N,2)   - ZD(N)
                    vup(n)      = v(n)
                    vdn(n)      = v(n)
                    head_loss   = 0.5*ENTK(N)*vup(N)*vup(N)/GRVT  +
     +                            0.5*EXITK(N)*vdn(N)*vdn(N)/GRVT +
     +                            0.5*OTHERK(N)*v(N)*v(N)/GRVT 
                    sfloss  = ROUGH(N)/GRVT * 
     +                        abs(v(N))*ABS(v(N))/hrlast(n)**1.33333
cred                sfloss  = ROUGH(N)/GRVT * 
cred +                   Q(N)*ABS(Q(N))/(A(N)**2*HRLAST(N)**1.33333)
             culvert_loss(n,1) = len(n)*sfloss
             culvert_loss(n,2) = head_loss
             culvert_loss(n,3) = h(n,1)
             culvert_loss(n,4) = h(n,2)
             culvert_loss(n,5) = q(n)  
   endif 
             CALL DEPTHX(N,NKLASS(N),q(n),YC,YNORM)
cred                type 1     
                    IF(HW.LT.1.5*DEEP(N).AND.YC.LT.YNORM.and.
     +                         TW.LE.YC) then
                               culverted(N,1) = culverted(N,1) + DELT
                        culverted(n,8) = 1
                        endif
cred                type 2
                    IF(HW.LT.1.5*DEEP(N).AND.YC.LT.YNORM.AND.
     +                         TW.GT.YC.AND.TW.LE.DEEP(N)) then
                               culverted(N,2) = culverted(N,2) + DELT
                        culverted(n,8) = 2
                        endif
cred                 type 3
                     IF(YC.GE.YNORM.AND.TW.LT.ZU(N)-ZD(N)) then
                               culverted(N,3) = culverted(N,3) + DELT
                        culverted(n,8) = 3
                        endif
cred                 type 4
                     IF(YC.GE.YNORM.AND.TW.GE.ZU(N)-ZD(N)+YC.
     +                         AND.TW.LT.ZU(N)-ZD(N)+DEEP(N)) then
                               culverted(N,4) = culverted(N,4) + DELT
                        culverted(n,8) = 4
                        endif
cred                 type 5
                     IF(YC.LT.YNORM.AND.TW.GE.DEEP(N)) then
                               culverted(N,5) = culverted(N,5) + DELT
                        culverted(n,8) = 5
                        endif
cred                 type 6
                     IF(YC.GE.YNORM.AND.TW.GE.ZU(N)-ZD(N)+
     +                         DEEP(N)) then
                               culverted(N,6) = culverted(N,6) + DELT
                        culverted(n,8) = 6
                        endif
cred                 type 7
                     IF(HW.GE.1.5*DEEP(N).AND.TW.LT.DEEP(N)) then
                               culverted(N,7) = culverted(N,7) + DELT
                        culverted(n,8) = 7
                        endif
                      endif
                      enddo 

      endif
      RETURN
      END