Error: Symbol ‘pi’ at (1) has no IMPLICIT type

[negar]

Hello,
Im using a model in my project, but unfortunately, I got 2 errors (See below please)
code:
c ------
c compar
c ------
c
c ------------------------
c main parameters for ibis
c ------------------------
c
      integer nlon, ! longitude dimension of domain
     > nlat, ! latitude dimension of domain
     > npoi, ! total number of land points
     > nband, ! number of solar radiation wavebands
     > nsoilay, ! number of soil layers
     > nsnolay, ! number of snow layers
     > npft ! number of plant functional types
c
      real xres, ! longitude resolution (in degrees)
     > yres, ! latitude resolution (in degrees)
     > pi ! you know, that constant thingy
c
c --------------------------
c typical ibis configuration
c --------------------------
c
c
c global 10.0 x 10.0 TEST grid: 36 by 18 array, with 147 land points:
c
c parameter (nlon = 36,
c > nlat = 18,
c > npoi = 147,
c > xres = 10.00,
c > yres = 10.00)
c
c South Asia 0.5 x 0.5 grid: 720 by 360 array, with 58920 land points:
c
c parameter (nlon = 720,
c > nlat = 360,
c > npoi = 1718,
c > xres = 0.50,
c > yres = 0.50)
c
c global 0.5 x 0.5 grid: 720 by 360 array, with 58920 land points:
c
     parameter (nlon = 720,
     > nlat = 360,
     > npoi = 58920,
     > xres = 0.50,
     > yres = 0.50)
c
c
c global 1.0 x 1.0 grid: 360 by 180 array, with 14545 land points:
c
c parameter (nlon = 360,
c > nlat = 180,
c > npoi = 14545,
c > xres = 1.00,
c > yres = 1.00)
c
c global 2.0 x 2.0 grid: 180 by 90 array, with 3728 land points:
c
c parameter (nlon = 180,
c > nlat = 90,
c > npoi = 3728,
c > xres = 2.00,
c > yres = 2.00)
c
c global 4.0 x 4.0 grid: 90 by 45 array, with 912 land points:
c
c parameter (nlon = 90,
c > nlat = 45,
c > npoi = 912,
c > xres = 4.00,
c > yres = 4.00)
c
c -------------------------------
c state description configuration
c -------------------------------
c
      parameter (nband = 2,
     > nsoilay = 6,
     > nsnolay = 3,
     > npft = 12)
c
c --------------
c some constants
c --------------
c
      parameter (pi = 3.1415927)
c
c ----------------------
c required common blocks
c ----------------------
c
      real
     > epsilon, ! small quantity to avoid zero-divides and other
     > ! truncation or machine-limit troubles with small
     > ! values. should be slightly greater than o(1)
     > ! machine precision
     > dtime, ! model timestep (seconds)
     > stef, ! stefan-boltzmann constant (W m-2 K-4)
     > vonk, ! von karman constant (dimensionless)
     > grav, ! gravitational acceleration (m s-2)
     > tmelt, ! freezing point of water (K)
     > hfus, ! latent heat of fusion of water (J kg-1)
     > hvap, ! latent heat of vaporization of water (J kg-1)
     > hsub, ! latent heat of sublimation of ice (J kg-1)
     > ch2o, ! specific heat of liquid water (J deg-1 kg-1)
     > cice, ! specific heat of ice (J deg-1 kg-1)
     > cair, ! specific heat of dry air at constant pressure (J deg-1 kg-1)
     > cvap, ! specific heat of water vapor at constant pressure (J deg-1 kg-1)
     > rair, ! gas constant for dry air (J deg-1 kg-1)
     > rvap, ! gas constant for water vapor (J deg-1 kg-1)
     > cappa, ! rair/cair
     > rhow ! density of liquid water (all types) (kg m-3)
c
      common /compar1/ epsilon, dtime, stef, vonk, grav, tmelt, hfus, hvap, hsub,
     > ch2o, cice, cair, cvap, rair, rvap, cappa, rhow
c
      real
     > garea(npoi), ! area of each gridcell (m**2)
     > vzero(npoi) ! a real array of zeros, of length npoi
c
      common /compar2/ garea, vzero
c
      integer
     > ndaypy, ! number of days per year
     > nlonsub, ! number of longitude points for subsetting
     > nlatsub ! number of latitude points for subsetting
c
      common /compar3/ ndaypy, nlonsub, nlatsub
c
      integer
     > ndaypm(12) ! number of days per month
c
      common /compar4/ ndaypm

error:
gfortran -O0 -O3 -I/usr/include -c -o biogeochem.o biogeochem.f
compar.h:44:21:

      parameter (nlon = 720,
                     1
Error: Expected another dimension in array declaration at (1)
compar.h:88:19:

       parameter (pi = 3.1415927)
                   1
Error: Symbol ‘pi’ at (1) has no IMPLICIT type

[Arjen_Markus]

The code you provide is only the include file - the problem is most probably in the combination with the actual source file, biogeochem.f. Without that we simply cannot reproduce the error. Please provide that as well or some scaled-down version that reproduces the problem.

[negar]

I have sent you my model. Here is my Error.

gfortran -O0 -O3 -I/usr/include -c -o biogeochem.o biogeochem.f
compar.h:44:21:

      parameter (nlon = 720,
                     1
Error: Expected another dimension in array declaration at (1)
compar.h:88:19:

       parameter (pi = 3.1415927)
                   1
Error: Symbol ‘pi’ at (1) has no IMPLICIT type
comsum.h:156:17:

      >  a10ancl3(npoi),     ! 10-day average canopy photosynthesis rate - c3 grasses (mol_co2 m-2 s-1)
                 1
Error: Variable ‘npoi’ cannot appear in the expression at (1)

[negar]

This is my model. In the previous post, it did not post.

[mecej4]

In the file "compar.h", line 44, the keyword "parameter" starts in column-6. Since you are using fixed format, the 'p' should be placed no earlier than column-7.

[negar]

I do not understand what you mean exactly

[mecej4]

Your source files are in fixed format Fortran. In that format, anything beyond column-72 is ignored, but some compilers may, as a non-standard extension, provide a way of processing longer lines without truncation. For Gfortran, the option is -ffixed-line-length-0, please spend some time reading the compiler documentation.

In my earlier post in this thread, I pointed out another error. Adding a blank character to the beginning of line-44 of compar.h, and adding -ffixed-line-length-0 to F77_OPTIONS in your makefile, I find that make completes the compilation of several source files, but stops when it attempts to compile source files that contain "include 'netcdf.inc"",since I do not have that include file.

[negar]

Thank you for answering my question. I completely understand what do you mean. After I made my model,  some object files which end in .o  were made. Could you please tell me how can I run the model?

[mecej4]

If make ran to completion without errors, you should have the executable file ibis in the directory where you ran make. You can run that file as usual by typing in the command ./ibis (the command may need options and arguments -- I know nothing about running that specific program, so you have to read the documentation for it.