Add files via upload

Initial upload of files from the zip archive. This does not include the .nc files as they are too large.

ABC_test.sh

@@ -0,0 +1,33 @@
+#!/bin/bash
+## Job Name
+#SBATCH --job-name=chinook_test_9thread
+## Allocation Definition
+#SBATCH --account=stf
+#SBATCH --partition=stf
+## Resources
+## Nodes
+#SBATCH --nodes=1      
+## Tasks per node (Slurm assumes you want to run 28 tasks, remove 2x # and adjust parameter if needed)
+#SBATCH --ntasks-per-node=3
+## Set number of cores per task to allow multithreading
+#SBATCH --cpus-per-task=9
+## Walltime (two hours)
+#SBATCH --time=10:00:00
+# E-mail Notification, see man sbatch for options
+
+
+##turn on e-mail notification
+
+#SBATCH --mail-type=ALL
+
+#SBATCH [email protected]
+
+
+## Memory per node
+#SBATCH --mem=100G
+## Specify the working directory for this job
+#SBATCH --chdir=/gscratch/stf/aebratt/fish_cpp_hyak
+
+module load parallel-20170722
+
+seq 6 | parallel -j 3 -n0 "/gscratch/stf/aebratt/fish_cpp_hyak/bin/headless ABC_test_listings.csv ABCoutput config_9thread.json"

ABC_test_listings.csv

@@ -0,0 +1,8 @@
+Run ID, Status (0: not run yet; 1: claimed), mort const A, mort const C
+0,0,-0.42,0.03096
+1,0,-0.43,0.03096
+2,0,-0.44,0.03096
+3,0,-0.45,0.03096
+4,0,-0.46,0.03096
+5,0,-0.47,0.03096
+6,0,-0.48,0.03096

CONFIG_README.md

@@ -0,0 +1,86 @@
+
+# Configuration
+
+Default model configurations are in the files [default_config_env_sim.json](default_config_env_sim.json) and [default_config_env_from_file.json](default_config_env_from_file.json).
+
+Parameters:
+- `threadCount`: The maximum number of hardware threads to use when running the model (-1 = as many as are available)
+- `envDataType`: string, either `file` or `sim`
+    - if `envDataType` is `file`, the following entries are expected:
+        - `recStartTimestep`: the number of 1-hour timesteps from midnight on January 1 to the start date/time of the recruitment data
+        - `hydroStartTimestep`: the number of 1-hour timesteps from midnight on January 1 to the start date/time of the hydrology input data (tide, flow volume, and air temperature)
+
+            Note: the flow speed data is assumed to start at midnight on January 1 regardless of `hydroStartTimestep`
+
+        - `recruitEntryNodes`: an array of integer map node IDs at which new recruits will originate
+        - `recruitCountsFile`: name of a text file containing one decimal number per line, each representing the number of recruits on a single day
+        - `recruitSizesFile`: name of a CSV file where each row after the first represents the distribution of recruits among 5mm size-classes for a single week
+        - `samplingSitesFile`: name of a CSV file where each row after the first describes a sampling site, giving its location and human-readable name
+        - `mapNodesFile`: name of a CSV file where each row after the first describes a map point, with relevant fields arranged as follows:
+
+            ID, _, _, _, _, area (m^2), habitat type, _, _, _, elev (m), _
+
+            - Elev is the vertical distance to NAVD88 in meters.
+            Habitat type is one of the following:
+                `blind channel`,
+                `impoundment`,
+                `low tide terrace`,
+                `distributary`,
+                `boat harbor`,
+                `nearshore`
+
+            - ID is one-indexed (starts from 1) in map input files but the model converts this to a zero-based index.
+
+        - `mapEdgesFile`: name of a CSV file where each row after the first describes a map edge, with relevant fields arranged as follows:
+
+            _, _, _, _, _, _, _, _, _, _, _, _, _, _, source node ID, target node ID, _, _, length (m), _
+
+            - The relevant fields are the 15th, 16th, and 19th fields.
+
+        - `mapGeometryFile`: name of a CSV file where each row after the first gives the location of a certain node.
+
+            Fields are: x, y, ID
+
+            x and y are UTM Zone 10N coordinates.
+
+        - `tideFile`: name of a text file where each line is a crescent tide height (m) measured every 15 minutes
+        - `flowVolFile`: name of an ASCII file where each line is a flow volume (m^3/s), measured every 15 minutes upstream from the recruitment site
+        - `airTempFile`: name of an ASCII file where each line is an air temperature, in degrees C, measured every 15 minutes
+        - `flowSpeedFile`: name of a NetCDF-3 or NetCDF-4 file where the following variables are present:
+
+            - `x`, `y`: the horizontal and vertical locations where flow velocity is given, in UTM Zone 10N coordinates
+
+                dimensions: node (arbitrary number of locations)
+            
+            - `u`, `v`: the horizontal and vertical flow velocity (m/s)
+
+                dimensions: time (1hr increments, starting at midnight on January 1st), node
+        
+        - `distribWseTempFile`: name of a NetCDF-3 or NetCDF-4 file where the following variables are present:
+
+            - `x`, `y`: the horizontal and vertical locations where flow velocity is given, in UTM Zone 10N coordinates
+
+                dimensions: node (arbitrary number of locations)
+            
+            - `wse`: the water surface elevation (in meters, from NAVD88 datum)
+
+                dimensions: time (1hr increments, starting at midnight on January 1st), node
+
+            - `temp`: the water temperature (in degrees C)
+
+                dimensions: time (1hr increments, starting at midnight on January 1st), node
+
+            Distributary nodes retrieve flow speed, depth, and temperature from their nearest hydro node, so these nodes should be relatively dense spatially in distributary regions.
+
+        - `blindChannelSimplificationRadius`: float; the maximum distance between blind channel nodes that will result in them being merged when the map data is loaded (to speed up model prediction)
+    - if `envDataType` is "sim", the following entries are expected:
+        - `mapParams`: A subgroup of parameters containing the following keys:
+            - `m`: int, the number of distributary nodes per grid row/column
+            - `n`: int, the total number of nodes per grid row/column
+            - `a`: float, the distance between grid nodes in meters
+            - `pBlind`: float between 0 and 1, the probability of deleting a blind channel
+            - `pDist`: float between 0 and 1, the probability of deleting a distributary channel
+        - `recruitSizeMean`: float, the mean recruit size in millimeters
+        - `recruitSizeStd`: float, the standard deviation of recruit sizes in millimeters
+        - `recruitRate`: int, the average number of recruits per timestep
+        - `simLength`: int, the number of timesteps of environmental data to generate

Makefile

@@ -0,0 +1,72 @@
+#CCEXE = /usr/local/opt/llvm/clang++
+CCEXE = g++
+CFLAGS = -Wall -g -std=c++11
+
+CC = $(CCEXE) $(CFLAGS)
+
+WXFLAGS = $(shell wx-config --cxxflags)
+
+WXLIBS = $(shell wx-config --libs)
+
+RAPIDJSON = -Irapidjson/include
+
+NETCDF = -Ilocal/netcdf-cxx4/include -Ilocal/netcdf-c/include -Ilocal/hdf5/include
+
+ABSLIB_NCCPP = $(dir $(realpath $(firstword $(MAKEFILE_LIST))))local/netcdf-cxx4/lib
+
+ABSLIB_NCC = $(dir $(realpath $(firstword $(MAKEFILE_LIST))))local/netcdf-c/lib
+
+ifeq ($(shell uname -s), Darwin)
+	DL_EXT = dylib
+else
+	DL_EXT = so
+endif
+
+LINKFLAGS = -v -Llocal/netcdf-cxx4/lib -Llocal/netcdf-c/lib -lnetcdf_c++4 -lnetcdf -lpthread -pthread -Wl,-rpath,$(ABSLIB_NCCPP) -Wl,-rpath,$(ABSLIB_NCC) local/netcdf-cxx4/lib/libnetcdf_c++4.$(DL_EXT) local/netcdf-c/lib/libnetcdf.$(DL_EXT)
+
+OBJ = $(addprefix build/, model.o hydro.o load.o fish.o map.o util.o env_sim.o map_gen.o)
+
+headless: $(OBJ) build/headless.o bin
+	$(CC) $(OBJ) build/headless.o -o bin/headless $(LINKFLAGS)
+
+gui: $(OBJ) build/gui.o bin
+	$(CC) $(WXLIBS) $(OBJ) build/gui.o -o bin/gui $(LINKFLAGS)
+
+build/gui.o: src/gui.cpp src/model.h src/fish.h build
+	$(CC) $(WXFLAGS) -c src/gui.cpp -o build/gui.o
+
+bin:
+	mkdir bin
+
+build:
+	mkdir build
+
+build/headless.o: src/headless.cpp src/model.h build
+	$(CC) -c src/headless.cpp -o build/headless.o
+
+build/model.o: src/model.cpp src/model.h src/load.h src/fish.h src/map.h build
+	$(CC) $(RAPIDJSON) $(NETCDF) -c src/model.cpp -o build/model.o
+
+build/hydro.o: src/hydro.cpp src/hydro.h src/map.h build
+	$(CC) -c src/hydro.cpp -o build/hydro.o
+
+build/load.o: src/load.cpp src/load.h src/map.h build
+	$(CC) $(NETCDF) -c src/load.cpp -o build/load.o
+
+build/fish.o: src/fish.cpp src/fish.h src/model.h src/map.h src/util.h build
+	$(CC) -c src/fish.cpp -o build/fish.o
+
+build/map.o: src/map.cpp src/map.h build
+	$(CC) -c src/map.cpp -o build/map.o
+
+build/util.o: src/util.cpp src/util.h build
+	$(CC) -c src/util.cpp -o build/util.o
+
+build/env_sim.o: src/env_sim.cpp src/env_sim.h src/map.h build
+	$(CC) -c src/env_sim.cpp -o build/env_sim.o
+
+build/map_gen.o: src/map_gen.cpp src/map_gen.h src/map.h build
+	$(CC) -c src/map_gen.cpp -o build/map_gen.o
+
+clean:
+	rm -rdf bin/* build/*

OUTPUT_README.md

@@ -0,0 +1,110 @@
+
+## Output Files
+
+The model is capable of producing three different kinds of output files:
+
+### State Snapshots
+
+State snapshots can be saved in two ways:
+
+- Pressing Ctrl+C to interrupt the headless model, then entering the 'save' command (with the desired filename)
+- In the GUI, selecting File -> Save from the menu
+
+State snapshots can be loaded from the GUI using the File -> Load menu item in order to load the saved state of the model.
+
+### Sample data
+
+Sample data files are produced by the `headless` executable, and consist of a record of statistics of population samples taken
+at various sampling sites throughout the map. This data is also included in state snapshots.
+Sample data is saved as `output_X.nc`, where X is the run ID.
+
+### Summaries
+
+Summaries are automatically saved as `summary_X.nc` by the headless executable when a simulation is finished, where X is the run ID.
+
+### Tagged Fish Histories
+
+Tagged fish histories can be saved from the GUI using the File -> Save Tagged History menu item. They can also be loaded in the GUI using File -> Load Tagged History in order to view a replay of the history of the tagged fish.
+
+## Formats
+
+All model outputs are saved as [netCDF 4](https://www.unidata.ucar.edu/software/netcdf/) files.
+
+### State Snapshots
+
+- Snapshot files contain the following dimensions:
+    - `n`: Indicates individual, for per-fish variables
+    - `populationHistoryLength`: Indicates timestep for population history entries
+    - `sampleHistoryLength`: Indicates sample number for sample history entries (not equivalent to time)
+
+- Snapshot files contain the following variables:
+    - `modelTime`: int, the current model timestep
+    - `recruitTime[n]`: ints, each fish's entry timestep
+    - `exitTime[n]`: ints, each fish's exit or death timestep (only valid for a given `n` if `status[n]` is not 0, since status 0 is Alive)
+    - `entryForkLength[n]`: floats, fork length of each fish upon entry (mm)
+    - `entryMass[n]`: floats, mass of each fish upon entry (g)
+    - `forkLength[n]`: floats, current/final fork length of each fish (mm)
+    - `mass[n]`: floats, current/final mass of each fish (g)
+    - `status[n]`: ints, current/final status of each fish
+        - 0: Alive
+        - 1: Dead from mortality risk
+        - 2: Dead from stranding
+        - 3: Dead from starvation
+        - 4: Exited
+    - `location[n]`: ints, current/final map node ID where each fish is located
+    - `travel[n]`: floats, most recent timestep's swim distance for each fish (m)
+    - `lastGrowth[n]`: floats, most recent timestep's growth for each fish (g)
+    - `lastMortality[n]`: floats, most recent timestep's mortality risk for each fish (probability value)
+    - `populationHistory[populationHistoryLength]`: ints, per-timestep counts of living individuals in the model
+    - `sampleSiteID[sampleHistoryLength]`: ints, each sample's site ID
+        - Refer to [data/sampling_sites.csv](data/sampling_sites.csv) for a list of sample site names and coordinates. Site IDs correspond to line numbers (site ID 0 is Grain of Sand, 1 is FWP New Site, etc.)
+    - `sampleTime[sampleHistoryLength]`: ints, the timestep when each sample was taken
+    - `samplePop[sampleHistoryLength]`: ints, the number of individuals in each sample
+    - `sampleMeanMass[sampleHistoryLength]`: floats, each sample's mean individual mass (g)
+    - `sampleMeanLength[sampleHistoryLength]`: floats, each sample's mean individual fork length (mm)
+    - `sampleMeanSpawnTime[sampleHistoryLength]`: floats, each sample's mean individual spawn time (timesteps)
+
+### Sample data
+
+- Snapshot files contain the following dimensions:
+    - `sampleHistoryLength`: Indicates sample number for sample history entries (not equivalent to time)
+- Sample data files contain the following variables:
+    - `sampleSiteID[sampleHistoryLength]`: ints, each sample's site ID
+        - Refer to [data/sampling_sites.csv](data/sampling_sites.csv) for a list of sample site names and coordinates. Site IDs correspond to line numbers (site ID 0 is Grain of Sand, 1 is FWP New Site, etc.)
+    - `sampleTime[sampleHistoryLength]`: ints, the timestep when each sample was taken
+    - `samplePop[sampleHistoryLength]`: ints, the number of individuals in each sample
+    - `sampleMeanMass[sampleHistoryLength]`: floats, each sample's mean individual mass (g)
+    - `sampleMeanLength[sampleHistoryLength]`: floats, each sample's mean individual fork length (mm)
+    - `sampleMeanSpawnTime[sampleHistoryLength]`: floats, each sample's mean individual spawn time (timesteps)
+
+### Summaries
+
+- Summary files contain the following dimensions:
+    - `n`: Indicates individual
+- Summary files contain the following variables:
+    - `recruitTime[n]`: ints, each fish's entry timestep
+    - `exitTime[n]`: ints, each fish's exit or death timestep (only valid for a given `n` if `finalStatus[n]` is not 0, since status 0 is Alive)
+    - `entryForkLength[n]`: floats, fork length of each fish upon entry (mm)
+    - `entryMass[n]`: floats, mass of each fish upon entry (g)
+    - `finalForkLength[n]`: floats, current/final fork length of each fish (mm)
+    - `finalMass[n]`: floats, current/final mass of each fish (g)
+    - `finalStatus[n]`: ints, current/final status of each fish
+
+### Tagged Fish Histories
+
+- Tagged fish histories contain the following dimensions:
+    - `n`: Indicates individual
+    - `t`: Indicates timestep
+- Tagged fish histories contain the following variables:
+    - `recruitTime[n]`: ints, each fish's entry timestep
+    - `taggedTime[n]`: ints, the timestep when each fish was tagged
+    - `exitTime[n]`: ints, each fish's exit or death timestep (only valid for a given `n` if `finalStatus[n]` is not 0, since status 0 is Alive)
+    - `entryForkLength[n]`: floats, fork length of each fish upon entry (mm)
+    - `entryMass[n]`: floats, mass of each fish upon entry (g)
+    - `finalForkLength[n]`: floats, current/final fork length of each fish (mm)
+    - `finalMass[n]`: floats, current/final mass of each fish (g)
+    - `finalStatus[n]`: ints, current/final status of each fish
+    - `locationHistoryOut[n, t]`: ints, the map node ID where each fish was on each timestep (or -1 if a fish wasn't in the model on the timestep in question)
+    - `growthHistoryOut[n, t]`: floats, the growth (g) on each timestep for each fish, or 0 if the fish wasn't in the model on the timestep in question
+    - `mortalityHistoryOut[n, t]`: floats, the mortality risk (probability) on each timestep for each fish, or 0 if the fish wasn't in the model on the timestep in question
+