Observational Data

Observed data is provided as point information at station locations. To provide continuous fields of climate information, station data can be interpolated spatially to produce gridded historical data. Such interpolated fields provide a portrait of climate beyond the station’s point locations but are impacted by potential interpolation errors, particularly in data sparce regions or in regions of complex topography (e.g., mountainous areas).

Dataset Descriptions

This Document provides guidance on:

  • Adjusted and Homogenized Canadian Climate Data - (AHCCD)
  • Canadian Gridded Climate Dataset - (NRCANmet)

Strengths and Limitations

Key Strengths of Observational Data

Strength Description
Based on real measurments Based on direct observations, observational datasets are easy to interpret.
Continuity Some stations provide many decades of weather observations.
Historical reference Grounded in physical measurements, observational datasets are widely accepted for defining historical baselines.

Key Limitations of Observational Data

Limitation Description
Potential discontinuities Changing observing systems and gaps in the data an create spurious variability/trends and discontinuous time series
Dependence on station coverage For direct observations, availability is limited to the locations of weather stations while for gridded products, data quality is highest where station density is high and decreases in remote, mountainous, or sparsely monitored regions.
Biases and errors Weather observation records may be influenced by various biases, from changing surroundings, changes in observers or instrumentation, changes of location etc. This limitation is mitigated in adjusted and homogenized datasets.
Limited amount of variables The products only provide information on individual variables at the land surface, and stations may provide different subsets of variables. Also, locations for different variables do not always correspond geographically.

Expert Guidance

Observational datasets are best understood as the record of past climate conditions and are most reliable where station coverage is dense. They are well suited to characterize the past climate and the assessment and validation of reanalysis datasets or climate simulations for historical periods. While some observational records in Canada date back to the 19th century, the number of stations in the Canadian network has been decreasing in recent decades, following a global trend, despite increasing automation of stations. In regions with low station density or for climate variables other than temperature and precipitation, reanalysis products might be preferable (or the only viable option).