Statistical, spatial and time series analysis techniques
Summary of key issues:
|Description||Data analysis techniques applicable to estuarine modelling.|
|Temporal applicability||From the short to long term, including tidal to geological/millennial.|
|Links with other tools||These techniques can be used in conjunction with most other modelling tools in the analysis of estuarine processes and systems, e.g. the use of spatial analysis within GIS for Historical Trends Analysis and Expert Geomorphological Analysis.|
|Necessary software tools / skills||Statistical analysis requires knowledge of statistics and there are a variety of software packages available.
Spatial analyses require knowledge of GIS and use of a dedicated GIS software package such as ArcMap, or MapInfo.
|Typical analyses||Use of time series data for derivation of trends (Figure 1),|
|Limitations||Data availability, in terms of sufficient data to perform adequate statistical analyses, or sufficient data coverage for spatial analyses.|
The statistical analysis of data is a huge subject, which is covered extensively by a wide range of standard texts in the literature (Benjamin & Cornell, 1970; Coolican, 1996; Gumbel, 1958; Leadbetter et al., 1982). In addition, many statistical software packages now include their own handbooks or supporting explanatory notes (MATLAB, NAG, SAS, SPSS, StatSoft and many others: see Statistical software).
The need for studies of estuaries often relates to the analysis of time-series data, the manipulation of geographic information, or a combination of the two. Again time-series analysis is an extensive subject that is well covered in the literature (Williams, 1997; Young, 1999) and on the web (e.g. Time series analysis). The development of spatial analysis is less well developed but with the rapid proliferation of GIS for environmental studies there is now a growing literature (Burrough & McDonnell, 1998; Raper, 2001).