What is the Australian Weather Dataset dataset?

Australian weather dataset with 366 daily observations and 22 meteorological variables for rain prediction ML models. Features temperature, rainfall, humidity, pressure, and wind data with binary classification target.

How can I download the Australian Weather Dataset dataset?

You can download the Australian Weather Dataset dataset in CSV or Parquet format directly from this page. Each table has its own download buttons.

What format is the Australian Weather Dataset dataset available in?

This dataset is available in CSV and Parquet formats. CSV is great for spreadsheet applications, while Parquet is optimized for data analysis tools like Pandas and DuckDB.

What is the license for the Australian Weather Dataset dataset?

This dataset is available under the Creative Commons Attribution 4.0 license. See the full license at https://creativecommons.org/licenses/by/4.0/

How many tables are in the Australian Weather Dataset dataset?

The Australian Weather Dataset dataset contains 1 table: weather.

Australian Weather Dataset

Last updated 1 weeks ago•January 2, 2026

Source:Australian Bureau of Meteorology

License:Creative Commons Attribution 4.0

Version:1.0

Time:One complete year of daily observations (366 days - leap year)

Location:Australia

Created by Dataset Agent

Dataset Overview

The Australian Weather Dataset provides a complete year of daily meteorological observations from the Australian Bureau of Meteorology, specifically curated for machine learning rain prediction tasks. This dataset captures the diverse climate patterns characteristic of the Australian continent, from cool winter mornings to hot summer days, making it an excellent benchmark for weather forecasting models.

The dataset contains 366 daily weather observations with 22 meteorological variables, providing a complete leap year of weather data.

View Source

SQL

SELECT COUNT(*) AS row_count FROM weather.csv

Data

Row Count
366
1 row

Temperature ranges from -5.3°C to 35.8°C, with an average daily maximum of 20.6°C and average minimum of 12.2°C.

View Source

SQL

SELECT MIN(MinTemp) AS min_temp, MAX(MaxTemp) AS max_temp, ROUND(AVG(MaxTemp), 1) AS avg_max_temp, ROUND(AVG(MinTemp), 1) AS avg_min_temp FROM weather.csv

Data

Min Temp	Max Temp	Avg Max Temp	Avg Min Temp
-5.3	35.8	20.6	12.2
1 row

Total annual rainfall recorded is 522.8mm, with rain occurring on 18% of days (66 days with measurable precipitation).

View Source

SQL

SELECT ROUND(SUM(Rainfall), 1) AS total_rainfall, ROUND( 100.0 * SUM( CASE WHEN RainToday = TRUE THEN 1 ELSE 0 END ) / COUNT(*), 0 ) AS rain_percentage, SUM( CASE WHEN RainToday = TRUE THEN 1 ELSE 0 END ) AS rainy_days FROM weather.csv

Data

Total Rainfall	Rain Percentage	Rainy Days
522.8	18	66
1 row

Data Preview

Below is a sample of the dataset showing key meteorological variables. The data includes both current-day observations and the target variable for next-day rain prediction:

Sample Weather Observations (5 rows)

Min Temp	Max Temp	Rainfall	Wind Gust Dir	Wind Gust Speed	Humidity9am	Humidity3pm	Rain Today	Rain Tomorrow
8	24.3	0	NW	30	68	29	false	true
14	26.9	3.6	ENE	39	80	36	true	true
13.7	23.4	3.6	NW	85	82	69	true	true
13.3	15.5	39.8	NW	54	62	56	true	true
7.6	16.1	2.8	SSE	50	68	49	true	false
5 rows

View Source

SQL

SELECT MinTemp, MaxTemp, Rainfall, WindGustDir, WindGustSpeed, Humidity9am, Humidity3pm, RainToday, RainTomorrow FROM weather.csv LIMIT 5

Data

Min Temp	Max Temp	Rainfall	Wind Gust Dir	Wind Gust Speed	Humidity9am	Humidity3pm	Rain Today	Rain Tomorrow
8	24.3	0	NW	30	68	29	false	true
14	26.9	3.6	ENE	39	80	36	true	true
13.7	23.4	3.6	NW	85	82	69	true	true
13.3	15.5	39.8	NW	54	62	56	true	true
7.6	16.1	2.8	SSE	50	68	49	true	false
5 rows

Climate Patterns and Seasonal Insights

Analysis reveals distinct meteorological patterns characteristic of Australian climate. The data shows clear relationships between atmospheric conditions and rainfall occurrence, with humidity and pressure serving as strong predictive indicators.

View Source

SQL

SELECT CASE WHEN MaxTemp < 15 THEN 'Cold (<15°C)' WHEN MaxTemp >= 15 AND MaxTemp < 25 THEN 'Mild (15-25°C)' WHEN MaxTemp >= 25 AND MaxTemp < 30 THEN 'Warm (25-30°C)' ELSE 'Hot (>30°C)' END AS temp_category, COUNT(*) AS count FROM weather.csv GROUP BY temp_category ORDER BY count DESC

Data

Temperature Category	Days
Mild (15-25°C)	178
Cold (<15°C)	90
Warm (25-30°C)	61
Hot (>30°C)	37
4 rows

The temperature distribution shows 49% of days fall in the mild range (15-25°C), while extreme temperatures (below 15°C or above 30°C) account for 35% of observations. This distribution is typical of temperate Australian regions.

View Source

SQL

SELECT CASE WHEN Rainfall = 0 THEN 'No Rain (0mm)' WHEN Rainfall > 0 AND Rainfall <= 5 THEN 'Light (0.1-5mm)' WHEN Rainfall > 5 AND Rainfall <= 20 THEN 'Moderate (5-20mm)' ELSE 'Heavy (>20mm)' END AS rainfall_category, COUNT(*) AS count FROM weather.csv GROUP BY rainfall_category ORDER BY count DESC

Data

Rainfall Category	Days
No Rain (0mm)	263
Light (0.1-5mm)	70
Moderate (5-20mm)	30
Heavy (>20mm)	3
4 rows

Heavy rainfall events (>20mm) are rare, occurring only 3 days throughout the year, while 72% of days recorded no measurable precipitation.

View Source

SQL

SELECT SUM( CASE WHEN Rainfall > 20 THEN 1 ELSE 0 END ) AS heavy_rain_days, ROUND( 100.0 * SUM( CASE WHEN Rainfall = 0 THEN 1 ELSE 0 END ) / COUNT(*), 0 ) AS no_rain_pct FROM weather.csv

Data

Heavy Rain Days	No Rain Pct
3	72
1 row

Wind Pattern Analysis

Wind patterns provide crucial predictive signals for weather forecasting. The dataset captures 16 compass directions for wind gusts, with northwesterly winds being the most prevalent.

View Source

SQL

SELECT WindGustDir, COUNT(*) AS count FROM weather.csv WHERE WindGustDir IS NOT NULL GROUP BY WindGustDir ORDER BY count DESC LIMIT 10

Data

Direction	Occurrences
NW	73
NNW	44
E	37
WNW	35
ENE	30
ESE	23
S	22
N	21
W	20
NE	16
10 rows

Average wind gust speed is 39.8 km/h, with maximum recorded gusts reaching 98 km/h. Northwesterly winds (NW, NNW, WNW) account for 42% of all observations.

View Source

SQL

SELECT ROUND(AVG(CAST(WindGustSpeed AS FLOAT)), 1) AS avg_gust, MAX(CAST(WindGustSpeed AS FLOAT)) AS max_gust FROM weather.csv WHERE WindGustSpeed IS NOT NULL AND WindGustSpeed != 'NA'

Data

Avg Gust	Max Gust
39.8	98
1 row

Rain Prediction Target Analysis

The primary ML target variable RainTomorrow indicates whether rain occurred the following day. The dataset also includes RISK_MM showing the expected rainfall amount, enabling both classification and regression tasks.

View Source

SQL

SELECT RainToday, RainTomorrow, COUNT(*) AS count FROM weather.csv GROUP BY RainToday, RainTomorrow ORDER BY RainToday, RainTomorrow

Data

Condition	No Rain Tomorrow	Rain Tomorrow
No Rain Today	255	45
Rain Today	45	21
2 rows

The correlation analysis reveals important predictive patterns:

When it rains today: 32% probability of rain tomorrow (21 out of 66 rainy days)
When it doesn't rain today: 15% probability of rain tomorrow (45 out of 300 dry days)
Overall class distribution: 18% positive (rain) vs 82% negative (no rain) - moderate imbalance

Atmospheric Measurements

The dataset captures atmospheric conditions at two standardized times daily (9am and 3pm), following Bureau of Meteorology observation protocols. This dual-reading approach reveals diurnal patterns critical for accurate forecasting.

View Source

SQL

SELECT ROUND(AVG(Humidity9am), 0) AS avg_humidity_9am, ROUND(AVG(Humidity3pm), 0) AS avg_humidity_3pm FROM weather.csv

Data

Time Of Day	Average Humidity (%)
9am	72
3pm	44
2 rows

Morning humidity averages 72% while afternoon readings drop to 44%, a 28 percentage point decrease reflecting typical evaporation patterns. Atmospheric pressure remains stable: 1019.7 hPa (9am) to 1016.8 hPa (3pm).

View Source

SQL

SELECT ROUND(AVG(Humidity9am), 0) AS h9, ROUND(AVG(Humidity3pm), 0) AS h3, ROUND(AVG(Pressure9am), 1) AS p9, ROUND(AVG(Pressure3pm), 1) AS p3 FROM weather.csv

Data

H9	H3	P9	P3
72	44	1,019.7	1,016.8
1 row

Data Quality Summary

Understanding data quality is essential for building reliable models. This dataset is generally complete with a few documented exceptions:

98.4% Complete Records: The Sunshine column contains 6 missing values (1.6%) represented as 'NA' strings. All other columns have complete data for all 366 observations.

Type Conversion Required: WindGustSpeed, WindSpeed9am, WindSpeed3pm, and Sunshine are stored as VARCHAR. Convert to numeric types before mathematical operations: CAST(WindGustSpeed AS FLOAT)

Comparison with Alternative Datasets

How does this dataset compare to other weather data sources?

Methodology and Data Collection

This dataset originates from the Australian Bureau of Meteorology's network of weather stations. Observations follow World Meteorological Organization (WMO) standards for consistency and reliability. Key methodological notes:

RainToday/RainTomorrow: Binary flag set to TRUE when rainfall exceeds 1mm threshold
RISK_MM: Continuous target variable representing actual next-day rainfall amount
Dual observations: 9am and 3pm readings capture diurnal variation patterns
Wind directions: 16-point compass rose (N, NNE, NE, ENE, E, etc.)

For extended analysis, consider combining this weather data with complementary datasets:

Air Quality Data: Correlate weather patterns with pollution levels (PM2.5, ozone)
Solar Irradiance Data: Combine sunshine hours with solar radiation measurements
Historical Climate Records: Extend temporal coverage for trend analysis
Agricultural Yield Data: Model weather impact on crop production

Table Overview

weather

Contains 366 rows and 22 columns. Column types: 14 numeric, 6 text, 2 boolean.

366 rows22 columns

weather

366

rows

columns

Data Preview

Scroll to see more

MinTemp	MaxTemp	Rainfall	Evaporation	Sunshine	WindGustDir	WindGustSpeed	WindDir9am	WindDir3pm	WindSpeed9am	WindSpeed3pm	Humidity9am	Humidity3pm	Pressure9am	Pressure3pm	Cloud9am	Cloud3pm	Temp9am	Temp3pm	RainToday	RISK_MM	RainTomorrow
8	24.3	0	3.4	6.3	NW	30	SW	NW	6	20	68	29	1,019.7	1,015	7	7	14.4	23.6	false	3.6	true
14	26.9	3.6	4.4	9.7	ENE	39	E	W	4	17	80	36	1,012.4	1,008.4	5	3	17.5	25.7	true	3.6	true
13.7	23.4	3.6	5.8	3.3	NW	85	N	NNE	6	6	82	69	1,009.5	1,007.2	8	7	15.4	20.2	true	39.8	true
13.3	15.5	39.8	7.2	9.1	NW	54	WNW	W	30	24	62	56	1,005.5	1,007	2	7	13.5	14.1	true	2.8	true
7.6	16.1	2.8	5.6	10.6	SSE	50	SSE	ESE	20	28	68	49	1,018.3	1,018.5	7	7	11.1	15.4	true	0	false

Row 1

MinTemp8

MaxTemp24.3

Rainfall0

Evaporation3.4

Sunshine6.3

+17 more columns

Row 2

MinTemp14

MaxTemp26.9

Rainfall3.6

Evaporation4.4

Sunshine9.7

+17 more columns

Row 3

MinTemp13.7

MaxTemp23.4

Rainfall3.6

Evaporation5.8

Sunshine3.3

+17 more columns

Showing 5 of 366 rows

Data Profile

366

rows

columns

100%

complete

393.2 KB

estimated size

Column Types

14 Numeric6 Text2 Boolean

High-Cardinality Columns

Columns with many unique values (suitable for identifiers or categorical features)

Temp3pm(200 unique values)
Pressure3pm(193 unique values)
Pressure9am(190 unique values)
MaxTemp(187 unique values)

Data Dictionary

weather

Column	Type	Description	Example
`MinTemp`	numeric	Numeric value (range: 7.6 - 14)	8, 14
`MaxTemp`	numeric	Numeric value (range: 15.5 - 26.9)	24.3, 26.9
`Rainfall`	numeric	Numeric value (range: 0 - 39.8)	0, 3.6
`Evaporation`	numeric	Numeric value (range: 3.4 - 7.2)	3.4, 4.4
`Sunshine`	string	Sunshine	"6.3", "9.7"
`WindGustDir`	string	Wind Gust Dir	"NW", "ENE"
`WindGustSpeed`	string	Wind Gust Speed	"30", "39"
`WindDir9am`	string	Wind Dir9am	"SW", "E"
`WindDir3pm`	string	Wind Dir3pm	"NW", "W"
`WindSpeed9am`	string	Wind Speed9am	"6", "4"
`WindSpeed3pm`	numeric	Numeric value (range: 6 - 28)	20, 17
`Humidity9am`	numeric	Numeric value (range: 62 - 82)	68, 80
`Humidity3pm`	numeric	Numeric value (range: 29 - 69)	29, 36
`Pressure9am`	numeric	Numeric value (range: 1,005.5 - 1,019.7)	1019.7, 1012.4
`Pressure3pm`	numeric	Numeric value (range: 1,007 - 1,018.5)	1015, 1008.4
`Cloud9am`	numeric	Numeric value (range: 2 - 8)	7, 5
`Cloud3pm`	numeric	Numeric value (range: 3 - 7)	7, 3
`Temp9am`	numeric	Numeric value (range: 11.1 - 17.5)	14.4, 17.5
`Temp3pm`	numeric	Numeric value (range: 14.1 - 25.7)	23.6, 25.7
`RainToday`	boolean	Boolean flag	false, true
`RISK_MM`	numeric	Numeric value (range: 0 - 39.8)	3.6, 3.6
`RainTomorrow`	boolean	Boolean flag	true, true

Australian Weather Dataset

Dataset Overview

Data Preview

Climate Patterns and Seasonal Insights

Wind Pattern Analysis

Rain Prediction Target Analysis

Atmospheric Measurements

Data Quality Summary

Comparison with Alternative Datasets

Methodology and Data Collection

Related Datasets

Table Overview

weather

weather

Data Preview

Data Profile

Column Types

High-Cardinality Columns

Data Dictionary

weather