data visualization

[lkuiucsb]

This is for the data visualization page:https://sbclter.msi.ucsb.edu/data/visuals/stearns-wharf/

• [x] latest data display in the figure: use the datatime when user click open the map. If there is missing data in the latest dates (the data server stop updating the data), it should show no data lines during the missing period. Currently, when data stop update, the figure stop at the last update time. 

• [x] Currently, there's a problem with the time axis. the tick mark space should be consistent and the datetime interval should be consistent. Now, the later datetime seem to compress.

• [x] labeling: -- At the top, add a label "Current Values" -- the date should be the date of the data. -- at the bottom, add some instructions "Click label to turn a display on/off. Y-axis scaling is automatic"

• [x] Control for y axis scale for different measurements. Set the y axis limits for each measurement at the beginning of the javascript ( in case we need to change later, it can be easily accessible).

a. temperature 0-30 b. chl a : 0-50 c. pressure : 1-5 d. salinity 30-35 if all four measurements are shown, select the widest range from above (e.g. 0-50).

[secabeen]

Perhaps there should also be some outlier filtering? It would be reasonable to filter out all water temperatures above 100C at a minimum.

[mobb]

This is for the data visualization page:https://sbclter.msi.ucsb.edu/data/visuals/stearns-wharf/

1. latest data display: use the datatime when user click open the map. If there is missing data in the latest dates (the data server stop updating the data), it should show no data lines during the missing period. Currently, when data stop update, the figure stop at the last update time.

Re 1 on this list: I think a better solution is to show the latest data with its correct date and time, and include the lag (e.g., " 15 minutes ago", "4 days ago"). The advantage is that it will be easier for us to see when the data stream is down, by looking at the lag. And for the community, if the lag is small, it will still be usable to some (e.g., all the surfers want to know is the current temperature - 15 minutes or an hour ago won't matter)

[mobb]

Perhaps there should also be some outlier filtering? It would be reasonable to filter out all water temperatures above 100C at a minimum.

filtering is not very practical for real time data. setting the y-axes to a reasonable range for each sensor will take care of cases where an outlier dominates the auto-scaling.

[lkuiucsb]

@lleeoochen Margaret revised the four points above. please go with the current description

[lleeoochen]

Done! Please double check if everything is fixed.