Sapflow

Exploratory notebook for processing raw data from sapflow units. Source code.

Attach data

Attach one or more datalog.csv files from the sapflow datalogger. Note: datalog.csv files should contain at least the following columns:

• temp1
• temp2
• millisSinceHeatPulse
• rtcUnixTimestamp
• millisSinceReferenceTemp
• nodeId

Exploring the raw data

First, let's take a look at the raw data:

Data table

Total rows in attached data file(s): .

Let's limit the number of data points rendered to zoom in on a single heat pulse:

Modify the inputs above to change start/end indices.

Currently showing data rows at index through . Our goal is to identify when both temperature sensors are declining in temperature.

Example (click to expand)

Below, you can see that the temperature readings for both temperature sensors begin to decline at around 75000 milliseconds after the heat pulse has fired.

Converting raw temperature data to mean heat ratios

Example (click to expand)

Given a few days' worth of data, you should be able to see a diurnal pattern where the mean heat ratio values are:

• close to 1.0 during nighttime, when little/no transpiration takes place;
• increase after sunrise, peaking at around mid-day; and
• fall after sunset.

You might need to invert the values in case the sapflow sensors were installed such that the downstream and upstream sensors were flipped.

Data table

Filtering

You can filter the data by date range and mean heat ratio values.