import tidepredictor as tp
from datetime import datetime, timedelta
path = tp.get_default_constituent_path(tp.PredictionType.level, model_name="DTU10")
repo = tp.NetCDFConstituentRepository(path, model_name="DTU10")
predictor = tp.LevelPredictor(repo)
df = predictor.predict(
lon=-2.75,
lat=56.1,
start=datetime(2021, 1, 1),
end=datetime(2021, 1, 1, 12),
interval=timedelta(hours=1),
)Getting started
Installation
- Install the tidepredictor package from the distributed wheel file. Available on https://github.com/DHI/tidepredictor/releases.
E.g. for version 0.1.0:
pip install https://github.com/DHI/tidepredictor/releases/download/v0.1.0/tidepredictor-0.1.0-py3-none-any.whlCopy constituent files (
.ncfiles) for bothDTU10andFES2014models to the local machine with the following directory tree:~/.local/share/tidepredictor/ ├── DTU10/ │ ├── current.nc │ └── level.nc └── FES2014/ ├── level/ │ ├── 2n2.nc │ ├── eps2.nc │ ├── . │ ├── . │ ├── . │ ├── ssa.nc │ └── t2.nc └── current/ ├── eastward_velocity/ │ ├── 2n2.nc │ ├── eps2.nc │ ├── . │ ├── . │ ├── . │ ├── ssa.nc │ └── t2.nc └── northward_velocity/ ├── 2n2.nc ├── eps2.nc ├── . ├── . ├── . ├── ssa.nc └── t2.nc
Usage
tidepredictor can be used both as a command-line application and as a Python library. The package currently supports two tidal models: DTU10 and FES2014. The desired model can be selected using the model_name keyword argument, with FES2014 set as the default option.
Python library
And similar for depth averaged currents.
path = tp.get_default_constituent_path(tp.PredictionType.current, model_name="FES2014")
repo = tp.NetCDFConstituentRepository(path, model_name="FES2014")
predictor = tp.CurrentPredictor(repo)
df = predictor.predict_depth_averaged(
lon=-2.75,
lat=56.1,
start=datetime(2021, 1, 1),
end=datetime(2021, 1, 1, 12),
interval=timedelta(hours=1)
)And current profiles.
path = tp.get_default_constituent_path(tp.PredictionType.current,model_name="DTU10")
repo = tp.NetCDFConstituentRepository(path,model_name="DTU10")
predictor = tp.CurrentPredictor(repo, alpha=1.0/3)
df = predictor.predict_profile(
lon=-2.75,
lat=56.1,
start=datetime(2021, 1, 1),
end=datetime(2021, 1, 2, 1),
interval=timedelta(hours=1),
levels=[-1.0, -10.0, -27.0]
)from plotnine import *
(
ggplot(df, aes("time","u", color="factor(depth)"))
+ geom_line() + geom_point()
+ labs(color="Depth")
)
Command line
!tidepredictor -x -2.75 -y 56.1 -s "2021-01-01" -e "2021-01-01 02:00:00" -i 30time,level
2021-01-01T00:00:00Z,-0.3465
2021-01-01T00:30:00Z,0.1553
2021-01-01T01:00:00Z,0.5931
2021-01-01T01:30:00Z,0.9517
2021-01-01T02:00:00Z,1.2437
All options are available as command line arguments.
!tidepredictor --help
Usage: tidepredictor [OPTIONS]
Predict the tides for a given location.
╭─ Options ────────────────────────────────────────────────────────────────────╮
│ * --lon -x FLOAT RANGE Longitude │
│ [-180<=x<=180] [default: None] │
│ [required] │
│ * --lat -y FLOAT RANGE Latitude │
│ [-90<=x<=90] [default: None] │
│ [required] │
│ --start -s [%Y-%m-%d|%Y-%m-%dT% Start date │
│ H:%M:%S|%Y-%m-%d [default: None] │
│ %H:%M:%S] │
│ --end -e [%Y-%m-%d|%Y-%m-%dT% End date │
│ H:%M:%S|%Y-%m-%d [default: None] │
│ %H:%M:%S] │
│ --interval -i INTEGER RANGE [x>=1] Interval in minutes │
│ [default: 30] │
│ --output -o PATH Output file, default │
│ is stdout │
│ [default: None] │
│ --model -m TEXT Model name │
│ [default: FES2014] │
│ --format [csv|json] Output format │
│ [default: csv] │
│ --type [level|current] Type of prediction, │
│ 'level' or 'current' │
│ (u,v) │
│ [default: level] │
│ --precision -p INTEGER RANGE [x>=0] Number of decimal │
│ places. (csv only) │
│ [default: 4] │
│ --alpha FLOAT Alpha factor for │
│ current profile │
│ [default: │
│ 0.14285714285714285] │
│ --install-completion Install completion │
│ for the current │
│ shell. │
│ --show-completion Show completion for │
│ the current shell, to │
│ copy it or customize │
│ the installation. │
│ --help Show this message and │
│ exit. │
╰──────────────────────────────────────────────────────────────────────────────╯