Heres a zip file of some CAESAR source and sample files (from the River Swale) - enough to compile and run.. These were actually saved after some 9000 years of sim, and are set to run fomr 1700 to 2500 ad (If I remember correcrtly)

The code is written (poorly) in C, and is pretty basic - probably with some mistakes. It works fine on Linux Boxes, and on a Compaq server we have here. For linux users, compile with the command

gcc tracer21.c -lm -O -o outfilename

For other compilers - I think the key command is the -lm (link modules).

If you run this code, it is set up to generate a series of output files as it goes along, waterdepth.dat (waterdepths), elevdiff.dat (elevation difference - erosion/dep) and some water direction files every 1000 iterations or so.. These can be viewed in arcinfo, by using the asciigrid command (the program is set up to generate them in this format). It dumps the most recent topography at these time steps too.

When it runs there is a series of numbers indicating the initialisation etc.. Then after a couple of mins it stabilises, and the first col. Is Time in years, second time in mins (elapsed) third water discharge, fourth erm a kind of theoretical discharge (ignore!) and the next 9 the volume of sedimetn discharged for the 9 grain sizes.

If you want to put in your own 'terrain', in the first bits of code, the #defines have the number of x cells (xsection) and y cells (reach), cell size is DX and diagonal size root (I think). Its set up so the water and sediment drains to the right (if looking down on the catchment from above) this can be changed, but for now probably easier if you can turn your DEM around so the main exit point (of the trunk stream) is to the right.

A brief explanation of the files needed/attached are:

Tracer.dat.init - the initial DEM elevation file

Tracer21.c - the source

Test4.dat - a bedrock DEM - in this version it is the same as tracer.dat.init (the initial topography), then when it reads in the data it subtracts 2m from this - so there is 2m of 'soil' or sediment covering the whole catchment Surfacebedrock.txt - contains x and y co-ordinates of where there are surface bedrock sections - quite important controls on the swale - but you can always just remove these for your version etc..

Minesites2.txt - x and y co-ordinates of where contaminats can be added (best ignore this for now..) Hourlyrain.dat - 10 years of hourly rainfall data. Climate4.dat - this contains the values that are used to change the rainfaill and hydrological model, 1st col the date (in 50 year sections), second col the 'm' value for the hydrological model - lower m = flashier hydrograph, e.g deforested. Good values to start with are 0.02 for forested, and 0.005 for grassland, third col is a factor by which the rainfall sequence is multiplied - to represet changing climate. Tracerg.dat.init - this contains the saved grainsize data - a large file.