DispersalKernel.f90 File Reference

Go to the source code of this file.

Functions/Subroutines
subroutine	calculateandstoredispersalkernelofspecies (isp, dofft)
	CalculateAndStoreDispersalKernelOfSpecies
subroutine	dispersalkernel (isp)
	DispersalKernel
subroutine	singlekernel (alpha, cellsidelength, epskernel, rad_c)
	SingleKernel
subroutine	doublekernel (dispfac, alfa1, alfa2, cellsidelength, epskernel, rad_c)
	DoubleKernel
real function	centertocentervalue (alpha, cellsidelength, latwin, lonwin)
	CenterToCenterValue
real function	kernelvalue (dist, alpha)
	KernelValue

Function/Subroutine Documentation

calculateandstoredispersalkernelofspecies()

subroutine calculateandstoredispersalkernelofspecies	(	integer, intent(in)	isp,
		logical, intent(in)	dofft )

CalculateAndStoreDispersalKernelOfSpecies

---

-Calculates the dispersal kernel of one species and stores it to spec.
-sets the dimensions for the FFT (if dispersal by FFT) -calculates the FF transformed kernel (if dispersal by FFT)

• USE: FileListModule, All_par
• IN:

  • Parameters

    isp	!>> species index
    doFFT	!>> flag whether dispersal is done by FFT or not

  • from ALL_par:
    • readDispersalKernel: LOGICAL: read in a previously stored dispersal kernel?
    • spec(maxspc):TYPE(SpecProperties) !>> : species properties; here a.o. all species parameters are put to spec
    • kernelType: INTEGER: 3 different options for shape of kernel: 0: radius, 1: single exponential, 2: double exponential
  • from FileListModule:
    • File: TYPE
    • DispersalKernel_File: TYPE(File): file from which to read in dispersal kernel if wished
    • DispersalKernelNew_File: TYPE(File): file to which to write out dispersal kernel
      
• OUT:
  • via All_par:
    
    • spec(maxspc) :TYPE(SpecProperties): species properties; here a.o. the dispersal kernels are put to spec
• CALLS:
  • DispersalKernel in DispersalKernel.f90
  • AllocateSpecKernel in AllocateAreasAndKernels.f90
  • AllocateSpecTransformedKernel in AllocateAreasAndKernels.f90
  • SetDimsForFFT in GFTConvolution2D.f90
  • CalculateKernelTransformation in GFT_Interact.f90

Definition at line 46 of file DispersalKernel.f90.

centertocentervalue()

real function centertocentervalue	(	real, intent(in)	alpha,
		real, intent(in)	cellsidelength,
		integer, intent(in)	latwin,
		integer, intent(in)	lonwin )

CenterToCenterValue

---

Calculates the value of the dispersal kernel between the centers of a source cell and and a target cell

• IN:

  • Parameters

    latwin,lonwin	: coordinates of target cell
    alpha	: Parameter characterizing slope of kernel; mean dispersal distance
    cellSideLength	: Side-lenght of grid cells

• OUT:

  • Parameters

    CenterToCenterValue

    : value of dispersal kernel in the given target cell

• CALLS:
  • KernelValue in DispersalKernel.f90
• CALLED from:
  • SingleKernel in DispersalKernel.f90

Definition at line 342 of file DispersalKernel.f90.

dispersalkernel()

subroutine dispersalkernel

(

integer, intent(in)

isp

)

DispersalKernel

---

Calculates the generatl dispersal kernel of one species, either by a single or a double kernel Allocates the kernel and stores it and its radius in spec

• USE: FileListModule, All_par
• IN:

  • Parameters

    isp	: species index

  • from ALL_par:
    • spec(maxspc): TYPE(SpecProperties): species properties dispFac, alfa1, alfa2
    • dispersalDifferent: LOGICAL: different kernels for the species? Usually FALSE
    • kernelType: INTEGER: 3 different options for shape of kernel: 0: radius, 1: single exponential, 2: double exponential
    • epsKernel: REAL: threshold for cut-off of kernel , all values smaller than this are set to 0
    • cellSideLength: REAL: Side-lenght of grid cells
      
    • alpha_all: REAL: Alpha-value of kernel if all kernels are the same
    • kernel(:, :): REAL, allocatable: intermediate array for kernel
• OUT:
  • via All_par:
    • TYPE(SpecProperties) : spec(maxspc): species properties kernel and rad
• CALLS:
  • SingleKernel in DispersalKernel.f90
  • DoubleKernel in DispersalKernel.f90
  • AllocateSpecKernel in AllocateAreasAndKernels.f90
• CALLED FROM:
  
  • CalculateAndStoreDispersalKernelOfSpecies in DispersalKernel.f90
    

Definition at line 142 of file DispersalKernel.f90.

doublekernel()

subroutine doublekernel	(	real, intent(in)	dispfac,
		real, intent(in)	alfa1,
		real, intent(in)	alfa2,
		real, intent(in)	cellsidelength,
		real, intent(in)	epskernel,
		integer, intent(out)	rad_c )

DoubleKernel

---

Calculates the double dispersal kernel of one species by adding two single kernels

• USE: All_par
• IN:

  • Parameters

    dispFac	: proportion of short distance dispersal
    alfa1,alfa2	: Parameter characterizing slope of kernel; mean dispersal distance
    cellSideLength	: Side-lenght of grid cells
    epsKernel	: threshold for cut-off of kernel , all values smaller than this are set to 0

• OUT:

  • Parameters

    rad_c

    : radius of combined kernel -*via All_par:* REAL, allocatable:: kernel(:, :) !>> intermediate array for kernel in actual discretization REAL, allocatable:: kernelFine(:, :) !>> intermediate array for kernel in fine discretization

• CALLS:
  • SingleKernel in DispersalKernel.f90
  • AllocateKernel in AllocateAreasAndKernels.f90
• CALLED from:
  • DispersalKernel in DispersalKernel.f90

Definition at line 293 of file DispersalKernel.f90.

kernelvalue()

real function kernelvalue	(	real, intent(in)	dist,
		real, intent(in)	alpha )

KernelValue

---

Calculates the value of the dispersal kernel at a certain distance

• IN:

  • Parameters

    alpha	: Parameter characterizing slope of kernel; mean dispersal distance
    dist	: euclidian distance between source cell and sink cell

• OUT:
  

  • Parameters

    KernelValue

    : value of dispersal kernel

• CALLED from:
  • CenterToCenterValue in DispersalKernel.f90

Definition at line 364 of file DispersalKernel.f90.

singlekernel()

subroutine singlekernel	(	real, intent(in)	alpha,
		real, intent(in)	cellsidelength,
		real, intent(in)	epskernel,
		integer, intent(out)	rad_c )

SingleKernel

---

Calculates the single dispersal kernel of one species

1. dimensions and resolution of the actual kernel and a fine discretization kernel are determined
2. the values of this fine resolution kernel are calculated between the centers of the center cell and of the other grid cells for one quadrant
3. this quadrant is copied to the other 3 quadrants
4. the kernel values all fine discretization cells within one actual discretization cell are summed
5. the actual kernel is normalized The idea of using the fine dispersal kernel is to reduce the discretization error, which would arise by simply calculating the actual dispersal kernel values between the centers of the actual discretization cells

• USE: All_par
• IN:

  • Parameters

    alpha_all	: Parameter characterizing slope of kernel; mean dispersal distance
    cellSideLength	: Side-lenght of grid cells
    epsKernel	: threshold for cut-off of kernel , all values smaller than this are set to 0

• OUT:
  

  • Parameters

    rad	: radius of kernel

  • via All_par:
    
    • REAL, allocatable: kernel(:, :) : intermediate array for kernel in actual discretization
    • REAL, allocatable: kernelFine(:, :) : intermediate array for kernel in fine discretization
• CALLS:
  • CenterToCenterValue in DispersalKernel.f90
  • AllocateKernel in AllocateAreasAndKernels.f90
• CALLED from:
  • DispersalKernel in DispersalKernel.f90

Definition at line 198 of file DispersalKernel.f90.