Four Criteria for Building a Reliable, Robust Climate Dataset

Climate datasets are used in a variety of projects across multiple disciplines, supporting various types of assessments and decision making at all stages of a project life cycle. These climate datasets may represent current conditions, future projections, or both. Climate data is readily available from a variety of sources; however, it is important to understand how the climate data will be used before selecting the source.

While many climate products are produced by research centres and scientists, the sources may be intended for specific purposes or have different quality assessment and control procedures. For example, one source may provide high-level information on maps that is used to support planning decisions while another source may provide detailed datasets used to support infrastructure design. It is important to not only understand what climate information is needed, but also the source being accessed in order to help solve the challenge of finding reliable and defensible data consistent with its application.

This is further illustrated by considering the different climate data needs between air quality and hydrology. Air quality generally focuses on current climate conditions over shorter time frames (approximately 5 years) across multiple variables to produce a large range of meteorological conditions to understand how air quality concentrations move through a region of interest. A flow-frequency analysis in hydrology requires longer historical climate records (more than 50 years in some cases) to understand the range and frequency of storm events observed in a region. This helps to identify low frequency events with high impacts (like a 100-year storm) and understand how water moves through a catchment.

Climate change assesses how climate has been changing and how it’s projected to change in the future by looking at changes in long-term averages and trends (approximately 20 to 30 years). Climate change acts as a modifier of an air quality or hydrological assessment, and the needs of a climate change dataset developed for air quality are very likely to be different from a dataset developed for hydrology. A climate change dataset can be used to look at future changes in both assessments using long-term averages and trends for the variables of interest (i.e., long-term changes in wind speed would be important for an air quality assessment).

The needs of each one of these disciplines is different, and for good reason. In order to build a reliable and robust dataset it’s important to have a firm understanding of these climate data needs.

There are four main criteria to evaluate when developing a reliable climate dataset to satisfy unique project needs and aid in decision making.

Keeping in mind that there is a certain level of uncertainty and some limitations associated with climate data, it is important to clearly document where this uncertainty comes from and communicate this along with the use or analysis of the climate data. The proper use and application of climate data is important for a successful project. Following a well-thought-out procedure with clear and transparent communication will help support assessments and decision making at all stages of a project.

 

LEARN MORE ABOUT GOLDER DIGITAL