Week 3: Tools for Analyzing and Describing Observations

Why is Observational data alone is not enough

Raw observational data, without analysis, is meaningless.

Statistical Analysis

This is used to make sense of the vast amounts of climate data through analysis, interpretation, and presentation.

Descriptive statistics

Describes data quantitatively

Inferential statistics

Used for testing hypotheses, making predictions, and drawing conclusions. It also deals with uncertainty.

Frequency distribution

Shows how many of each value exists within a one-dimensional set of numbers.

Scatter plot

Used for two-dimensional data, where each point has two associated values. It helps in finding a general relationship between the values by observing the clustering of points.

Axis Scales

Axes can be either linear or logarithmic, depending on the data being displayed, such as in stream discharge graphs.

Rose diagram

A radial graph used to display information that has a directional component, such as wind speed and direction

Pie graph

Used to display a part-to-whole relationship in a data set. An example is the percentage of total Accumulated Cyclone Energy (ACE) contributed by individual storms

Shape/Distribution

The shape of a frequency graph that is repeatable, most commonly the normal distribution or "bell curve". Peak values are the most common, while tail values are the least common.

Time Series Analysis

An analysis of climate observations taken over time, accounting for seasonal variations or other internal patterns.

What is the main objective

the two main objectives are to understand variations in an ECV and to use that understanding to forecast future changes.

Trends

The general tendency of values to increase or decrease over time.

Cycles

Regular fluctuations of similar magnitude and time span. An example is the seasonal cycle of CO2 concentration observed at Mauna Loa, Hawaii, where photosynthesis causes a decrease in CO2 during the spring bloom.

Climate Normals

Climate is typically described in terms of "normals" (30-year averages), "central tendencies," and "extremes" and "deviations" in ECVs.

Ice-core data:

Provides a record of CO_2 concentration before 1958, which is combined with Mauna Loa data for a more extended time series.