Trend Analysis for a Whole Monitoring Network — Every Well, Every Analyte
ProDrop in a wide export from your laboratory database and get one figure: a matrix of wells against analytes, an arrow and a p-value in every cell, and a detail table underneath carrying n, the non-detect count, S, Z, p, Sen’s slope and its confidence interval for every series — inside the SVG, so the exported file is the summary table as well as the picture. Non-detects are handled exactly as the single-series test handles them, each well tied below its own reporting limit. Switch on the seasonal test and Q1 is compared only with Q1, so a water-table cycle cannot masquerade as a trend.
Your data
This tool is showing its sample dataset.
Everything in the figure is real output computed in your browser. Loading a file of your own is part of Pro.
Synthetic eleven-year quarterly record from a six-well groundwater monitoring network — benzene, TCE, chloride and sulfate, 23 series in one wide export. The source-area well is recovering while a downgradient well is getting worse, the upgradient and sentinel wells are non-detect at two different reporting limits, one chloride series has a seasonal cycle larger than its own trend, one analyte was added to the list in the final year, and one well has no sulfate column at all. Not measured data.
Saved projects
0 of 1Projects are stored in this browser only — never uploaded, and gone if you clear site data. Download one as a file to keep it, move it to another machine, or send it to a colleague.
SVG is vector — scale it to any size in Illustrator, Inkscape or LaTeX without losing quality, and edit the type. Use it unless a journal insists on raster.
Options
Layout
A column called “MW-04 Benzene” becomes the Benzene cell of well MW-04. This only moves cells around; it changes no statistic.
Locations are the rows of the matrix and analytes are its columns.
Non-detects
Applied per series, exactly as the single-series trend test applies it — each well’s own highest reporting limit, not the network’s.
Seasonality
The seasonal test compares Q1 only with Q1 and sums the seasons, so a cycle in the water table cannot masquerade as a trend. It costs power on a record that has no cycle.
Test
Two-sided is the honest default unless the direction was specified before the data were seen.
Which cells are drawn as a trend. Every p-value is printed whatever this is set to — and see the note on the figure about testing two dozen hypotheses at once.
A series shorter than this is reported as untested rather than given a p-value it has no power behind.
Slope
Two-sided limits by Gilbert (1987) §16.5, from the same variance the cell’s test used.
A year is 365.25 days and a month is a twelfth of one, so the three are exactly proportional.
Figure
All of that was worked out in this browser tab. Your file was not uploaded, and no request goes out while you work — open the network panel and watch, if you like.
The figure above is this tool's sample dataset, and everything in it is real output — every option works and every number is computed here in your browser. Running Monitoring network trends on a file of your own is part of Pro.