|
 |
Climatology, interpolation, and forecasting are all fundamentally based on estimates. That is, the data that is available is used to derive values for unknown points. Climatology is based on historic events, measured values, and patterns. Interpolation creates values for points where no values have been measured. Forecasting tries to predict the future. All of these processes involve incomplete, and sometimes inaccurate, data. There are many errors in this project that can be associated with the quality, accuracy, and amount of data.
An examination of the data errors is initiated by looking at the source of the data. All of the precipitation values came from the Environment Canada website, but beyond that it is not known how the data was collected or monitored at each station. An assumption can be made that data from stations such as UBC and YVR are more accurate and data from smaller stations may not be as reliable. This also relates to the errors associated with the collection of the data. Common techniques for measuring data in rain gauges are very inaccurate (Ward and Robinson, 2000). Wind, aspect, surrounding features (trees, buildings, etc), and slope are all factors in the effectiveness of the rain gauge.
Another source of error for the data is human error. Because this data has gone through many people and many formats to get to the point where it could be use in the modeling, there were many opportunities for data to be entered/recorded erroneously. These mistakes could have been made by the original person taking the measurement, by the data entry people at Environment Canada, or by anyone working on this project.
Other factors are associated with the accuracy of the data. Some measurements were made to a single decimal (i.e. one tenth of a mm) while others were rounded to the nearest whole number. Problems with accuracy also apply to the locational data for the stations. Environment Canada provided latitude and longitude values to minutes only. Acquisition of the locations to seconds for most of the locations was possible, but not for all of them. This created problems for points such as Delta Pebble Hill, which is now located outside of the GVRD because of inaccurate locational data. The points with latitude and longitude to only minutes may be off by a maximum of approximately 1.8 km.
Other problems associated with the data that are not actual errors in the data are outlined in Systematic Errors.
 |
back to
top |
 |