Understanding "The Temperature"

The Different Types of Temperature Data

It's important for weather market participants to know that the temperature reported publicly might not tell the whole story. Different types of temperature data come with varying levels of precision. Here’s a breakdown:

• High-Frequency Data: Reported publicly at five minute intervals, this is the most frequent data but often the least precise. Temperatures are rounded at multiple stages, introducing potential inaccuracies.
• Hourly Data: Generally reported near the top of each hour, this data strikes a balance between frequency and precision, offering a Celsius value rounded to the fist decimal.
• SPECI: A special weather report, issued when there are significant changes in the weather conditions, offering a high level of precision.
• DSM (Daily Summary Message): Intra-day reports that provide a very high level of precision.
• CLI (Climate Report): The gold standard. These reports include official highs, lows, and averages, calculated from raw sensor data without additional rounding layers. This is the report that is most commonly used to resolve daily weather markets.

How Rounding Affects Temperature Data

Public temperature data often undergoes multiple layers of rounding, especially in high-frequency reports. Here’s an example to demonstrate how rounding can create errors:

• The temperature sensor records 77.6°F and the system rounds this to the nearest whole number: 78°F.
• This rounded value is converted to Celsius: 25.5556°C, which is then rounded to the nearest whole number: 26°C.
• When converted back to Fahrenheit, 26°C becomes 78.8°F, which is what is likely to show on most weather apps/websites.
• Because the daily summary uses whole numbers in Fahrenheit, you might expect it to be 79°F because it reached 78.8°F on the time series, but it was a mirage caused by rounding.

Due to the multiple layers of rounding, the public data could inaccurately reflect a different value than the original sensor measurement.

A Real-World Example

Imagine this scenario: A public weather app shows a temperature of 78.8°F throughout the day. Later, the official daily summary reports the high temperature as 78°F. What happened?

The discrepancy occurs because the high-frequency data is reported in Celsius that has been rounded to the nearest whole number and converted back to Fahrenheit for display, while the daily summary uses the raw sensor data rounded to the nearest whole number (78°F), avoiding additional rounding layers.

Why Wethr.net Charts Show a Range

At Wethr.net, we account for these nuances by displaying a range for high-frequency data. This approach reflects the inherent margin of error and gives you a more realistic picture of the temperature trends. When the data is not precise enough for us to provide an exact value, we will show you the range of possibilities. 77°F - 78°F.

This transparency helps weather market participants make more informed decisions when it comes to managing their positions.