GitHub

Quick Start

This page walks through the shortest useful ClimateTools workflow: open a dataset, inspect its dimensions, compute a summary, and prepare for scenario-building work.

1. Open a Dataset

ClimateTools operates on YAXArrays-native cubes.

using ClimateTools
using YAXArrays

cube = Cube(open_dataset("tasmax.nc"))

When a dataset contains multiple variables, select the one you need first.

ds = open_dataset("tasmax.nc")
tasmax = Cube(ds[:tasmax])

2. Inspect Axes and Variable Shape

Before computing anything, check which dimensions are present.

axes(tasmax)
size(tasmax)

Typical dimensions are time, longitude, and latitude, but some datasets use alternative names such as Ti, rlon, or rlat.

3. Select a Study Period

Use DimensionalData selectors to subset in time.

using Dates
using DimensionalData

hist = tasmax[time=DateTime(1981, 1, 1)..DateTime(2010, 12, 31)]

If your dataset uses a different time dimension name, inspect axes(tasmax) and adapt the selector accordingly.

4. Compute a Summary or Index

Aggregation functions work directly on cubes.

annual_hot = annualmax(hist)

Many xclim-style indices also operate on cubes directly.

txx = tx_max(hist)
hot = hot_days(hist; thresh=30.0, freq="YS")

5. Visualize a Slice

ClimateTools returns ordinary Julia arrays once you materialize the data.

using Plots

arr = Array(txx)
heatmap(arr[:, :, 1], title="Annual maximum tasmax")

6. Bias-Correct a Simulation

If you have observations, a historical simulation, and a future simulation on comparable grids, you can apply a basic bias-correction workflow.

obs = Cube(open_dataset("obs.nc"))
ref = Cube(open_dataset("historical.nc"))
fut = Cube(open_dataset("future.nc"))

qq = qqmap(obs, ref, fut; method="additive", detrend=true)
txx_future = tx_max(qq)

This is the core ClimateTools pattern: open, align, correct, then compute indicators.

7. Next Steps