The CHASE tool integrates data on hazardous substances in water, sediments and biota as well as bio-effect indicators and is based on a substance- or bio-effect-specific calculation of a ‘contamination ratio’, being the ratio between an observed concentration and a threshold value. Values <1.0 indicate areas ‘unaffected’, while values >1.0 indicate areas potentially ‘affected’. These ratios are combined within the matrices, water, sediment and biota and for biological effects and the overall assessment for each assessment unit is determined by use of a ‘one out, all out principle’ with regard to each matrix. That is, the ‘worst’ matrix determines the status for the assessment unit in question.
In addition to this primary status assessment, the CHASE tool makes a secondary assessment, estimating the confidence if the primary assessment result. This is done by combining a number of qualitative and semi-quantitative confidence parameters for each indicator.
The CHASE assessment is carried out by calling the function
CHASEassessment. The source code for this function is contained in
CHASE.R and auxiliary functions are contained in
CHASE_functions.R. These two files contain the
code required for making the assessment. Other files in this repository
are used only for testing.
source('src/CHASE.R')
result <- CHASEassessment(indicator_dataframe)A required argument for the function is a dataframe containing the assessment indicator data. Required column names are seen in the table here:
| Column | Required | Description |
|---|---|---|
| AU | NO1 | assessment unit name or identifier |
| Matrix | YES | Water, Sediment, Biota or Biological effects |
| Substance | YES | name of indicator |
| Type | YES | HM / Org |
| Threshold | NO | concentration limit |
| Status | NO | observed concentration |
| CR | YES2 | ratio of status to threshold |
| ConfTemp | NO3 | temporal confidence: H[igh], M[oderate] or L[ow] |
| ConfSpatial | NO3 | temporal confidence: H[igh], M[oderate] or L[ow] |
| ConfAcc | NO3 | accuracy confidence: H[igh], M[oderate] or L[ow] |
| ConfMethod | NO3 | methodological confidence: H[igh], M[oderate] or L[ow] |
| ConfThresh | NO3 | threshold confidence: H[igh], M[oderate] or L[ow] |
1If no assessment units are provided, the indicators will be
aggregated for a single assessment.
2If CR is not provided, then Threshold and Status are
required columns.
3Confidence components are not required in order to give a
status assessment. If any single confidence component is not provided,
it will be assumed to be Low.
By default, the function returns a list of 4 dataframes:
- ‘Indicators’ is essentially the same as the input dataframe, with the addition of overall confidence score for each indicator, as well as results for the AU/matrix combination to which the indicator belongs.
- ‘Matrix by row’ gives an aggregated result with each combination of AU/matrix on a separate row.
- ‘Matrix by column’ gives an aggregated result with each AU on a separate row and Matrices in columns.
- ‘AssessmentUnits’ gives an aggregated overall result for each AU.
The CHASE assessment is tested here: CHASE_test.md