Data Upload

Upload Heat Pulse Data

Supported formats: ICT JSON, CSV, Legacy text files (.txt, .csv, .json, .dat)

Clock Drift Correction (Optional)

Correct clock drift assuming the first pulse time was correct (synced at data collection start).


Device Time (at collection end)

What time did the logger show?

Actual Time (at collection end)

What was the actual time?


Trim Incomplete Days (Optional)

Remove first and/or last day if they contain less than 23 hours of data. This is recommended for daily aggregation analyses.

Upload Weather Data (Optional)

CSV format with datetime, temperature, and relative humidity columns

VPD Calculation
to

Data Summary


                    

Probe & Wood Configuration

Probe Configuration


Wood Properties

Using default generic softwood properties.

Load Existing Configuration (Optional)

Upload an existing wood properties YAML file to edit.

Configuration Status

Wood Properties Configuration


Measure fresh weight immediately after sampling, then oven-dry until weight is stable.


Auto-calculated Densities:

These are calculated automatically from your weight and volume measurements.

Dual Density Measurements:

Measure BOTH densities on the same sample. Volume cancels out!


Physical constants for wood and sap. Defaults from Burgess et al. (2001).

Basic Properties:

Thermal Constants:

Initial Wound Configuration:

The wound around the probe affects the measured sapwood area.


Temporal Wound Tracking (Optional):

Dates auto-populate from loaded heat pulse data range (after any trimming). Adjust manually if needed to match original sensor installation dates.

Initial wound = drill bit + 2 × wound addition. Default: 2.0 + 2(0.3) = 2.6 mm


Calculate wood properties based on your measurements. Requires Method 1 or Method 2 measurements.


Optional - tree-specific measurements for scaling calculations.


Set acceptable ranges for quality control.

Export Configuration

Download your wood properties configuration as a YAML file. Derived properties will be automatically calculated if measurements are provided.


Probe Configuration Visualisation

Visual representation of probe placement relative to tree anatomy.


Heat Pulse Velocity Calculations

Select Calculation Methods

Select one or more heat pulse velocity calculation methods:

Quality Check Settings

Configure quality control checks applied after calculation:

Illogical Values
Statistical Outliers

These settings control outlier detection sensitivity and data validation. Lower thresholds are more strict, higher are more lenient.

Run Calculations



Calculation Results

Interactive Visualisation - Raw (Uncorrected) HPV

View raw heat pulse velocity calculations to identify outliers, missing data, and quality issues before applying corrections.

Plot Controls

Methods to Display

Sensor Position
Tip: Select only one sensor for faster plot rendering with large datasets

Quality Flags to Display

Time Range

Set the date/time range to display. Updates automatically when you use the range slider.


Display Options

Data Cleaning & Filtering

Interpolation Settings

Creates a straight line between valid points. For single missing points, this is the simple average of before/after values.

Uses weighted average of surrounding valid points. Window size is automatically determined based on gap size. Good for noisy data.

Warning: Gaps > 3 hours are not recommended

Quality Flags to Interpolate

Filter by Method/Sensor (optional)

Leave all unchecked to apply to all methods/sensors




Heat Pulse Velocity Time Series

Loading...

Plot Information & Quality Control

Plot Information
Quality Control Summary

Pulse Trace Controls

Selected Pulse

                        
Sensor Position

Calculation Windows

Show calculation windows used by each method. Heat pulse injection is at t=0.


Pulse Temperature Trace

Loading...

Spacing Correction

About Changepoint-Based Correction

Note: View Tab 4 (Visualise Raw HPV) first to identify data quality before correction.


This approach detects baseline shifts in daily minimum velocities caused by probe movement (tree swelling/shrinkage).

  • Step 1: Detect changepoints that divide data into segments
  • Step 2: Apply separate Burgess corrections per segment
  • Step 3: Review segment-specific results

Based on Burgess et al. (2001) with PELT changepoint detection

Define Zero-Flow Changepoints

Changepoints mark dates where probe alignment shifts, dividing data into segments for separate calibration.

Recommended Method
Detects changepoints when BOTH VPD AND sap flow are stable during pre-dawn periods. This dual-criterion approach eliminates false positives from stem refilling and provides higher confidence in baseline detection.
Weather data required
Please upload weather data in Tab 1 and calculate VPD before using this method.
VPD Stability Criteria

Sap Flow Stability Criteria

Detection Settings



Detected Dual-Stable Changepoints:
Changepoints where BOTH VPD and sap flow are stable. Shown as purple dotted lines on plot.


Detect extended periods of stable low VPD conditions suitable for spacing correction.
Weather data required
Please upload weather data in Tab 1 and calculate VPD before using this method.



Detected Stable VPD Changepoints:
Detected changepoints shown as green dotted lines on plot. Click [+] to add individually or use button below to add all.


Automatically detect baseline shifts in daily minimum velocities using PELT changepoint detection.



Detected Changepoints:
Detected changepoints shown as orange dotted lines on plot. Click [+] to add individually or use button below to add all.


Add a changepoint at a specific date/time where you know the baseline shifted.

Current Changepoints:
Confirmed changepoints shown as red dashed lines with baselines. Click [X] to remove.

Apply Spacing Correction

Baseline Correction Method
Segment Minimum: Traditional approach - uses minimum value in each segment between changepoints. Creates step-wise corrections.
Gradient Interpolation: Advanced - linearly interpolates between changepoint values for smooth, continuous correction. Eliminates artificial jumps. Use with Caution
Edge Period Limitation
Gradient interpolation requires empirical baseline values. Only data between the first and last changepoints can be scientifically corrected.

Data outside this range:
• Before first changepoint: No empirical support
• After last changepoint: No empirical support

Recommendation: Exclude edge periods or use Segment Minimum method for full dataset coverage.

Burgess Correction Method
Burgess: Uses Burgess et al. (2001) physics-based correction. Most accurate for HRM with offsets ≤ ±5 cm/hr.
Linear: Simple empirical offset subtraction. Works for all methods (HRM, MHR, Tmax) and large offsets.

Daily Minimum Velocities with Changepoints

Visualise daily minimum velocities and changepoints. Click on the plot to add a changepoint at that date.
Loading...

Tip: Click any date on the plot to populate the Manual tab for adding a changepoint.

Spacing Correction Results

No results yet. Define changepoints and run spacing correction.


Loading...

Each segment between changepoints gets separate Burgess correction coefficients.


                              

Wound Correction

Apply wound correction for probe reinstallations.

About Wound Correction

Wound correction accounts for the wound created by probe installation, which expands over time as wound tissue forms.

  • Initial Installation: First date of data with initial wound size (drill bit + wound tissue)
  • Reinstallations: Add dates when probe was removed and reinstalled with measured wound diameter
  • Temporal Tracking: Wound diameter is interpolated linearly between dates

Based on Burgess et al. (2001) and ICT International Appendix 23.1

Initial Installation

Initial installation date is automatically set to the first date of your data. Initial wound diameter is calculated from drill bit size and wound tissue addition.

Wood Properties Configuration:
Drill bit diameter and wound tissue addition are loaded from your wood properties YAML file.

                      

Manage Reinstallations

Add dates when the probe was removed and reinstalled. The wound diameter will reset to the initial size at each reinstallation.

Current Reinstallations:
Click [X] to remove a reinstallation.

Final Measurement (Growth Rate)

Define the final wound diameter at the end of the experiment (or a specific date). This determines the daily growth rate applied to all installation periods.

Apply Wound Correction

Apply wound correction to spacing-corrected velocity data. Correction uses temporal wound diameter tracking if reinstallations are defined.

Temporal Wound Diameter

Visualize how wound diameter changes over time based on initial installation and reinstallation dates.

No temporal wound tracking defined. Set final measurement date and diameter to see wound growth over time.

Loading...

Wound Correction Results

No results yet. Apply wound correction to see results.


Loading...

Wound correction coefficients (B) applied over time.

                              


                              

Active Wound Correction Status

No wound correction applied.

Wound correction is active and will be used in downstream analyses:


                          

Method Calibration

Calibrate secondary methods to a primary method using linear regression.

About Calibration

Review the diagnostic plots on the right to evaluate calibration quality for each method.

The segmented regression plot shows the breakpoint where methods diverge.

You can use auto-detected thresholds or set manual values per method.

Recommended: Apply AFTER spacing and wound corrections

Configuration

Primary method is HRM. All secondary methods will be calibrated against HRM.

Threshold Settings

Review the segmented regression plots, then set thresholds for each method below. The auto-detected value is the statistically identified breakpoint.

Apply Calibration


Calibration Applied


                        

Calibration Diagnostics - Review Before Applying

Review these plots to understand calibration quality and threshold selection.
LEFT: Segmented regression - shows breakpoint where methods diverge
RIGHT: Residuals plot - shows pattern and fit quality

Calibration Validation

Verify the quality of method calibration by comparing raw vs. calibrated velocities against the HRM baseline.

Validation Controls

Calibration State
Compare raw vs. calibrated velocities

Methods to Display
Select methods to compare against HRM baseline

Sensor Position

Time Range

Select date range to display


Display Options

Calibration Validation Plot

This plot shows how well the calibrated secondary methods align with HRM (the primary/baseline method). Good calibration results in similar patterns between HRM and secondary methods.
Loading...

Calibration Statistics

Selectable DMA (sDMA) Method Switching

Apply Selectable Dual Method Approach (sDMA) to switch between calibrated methods based on recalculated Peclet numbers and flow conditions.

About sDMA

sDMA (Selectable Dual Method Approach) automatically switches between methods based on flow conditions.

  • Step 1: Recalculate Peclet numbers using calibrated HRM velocities
  • Step 2: Select secondary method and Peclet threshold
  • Step 3: Apply sDMA switching logic

Peclet number determines the theoretical validity limit of HRM. When Pe ≥ threshold, sDMA switches to the secondary method.

Step 1: Recalculate Peclet Number

Recalculate Peclet numbers using the latest calibrated HRM velocities.

This ensures switching thresholds reflect wound and spacing corrections.
Peclet Recalculated

                          

Step 2 & 3: sDMA Configuration

Sensor Positions:

Secondary Methods:
Select one or more methods to apply sDMA switching.

When Pe ≥ threshold, use secondary method. Otherwise use HRM.
sDMA will be applied to all selected sensor/method combinations.

sDMA Results


Loading...


                              

Visualize switching points: HRM below threshold, secondary method above.

Loading...

No sDMA results yet. Follow the steps on the left to apply sDMA switching.

sDMA Validation - Interactive Time Series

Compare HRM baseline, calibrated secondary methods, and sDMA results in an interactive time series plot.

Plot Controls

Methods to Display
Select specific methods to show

Sensor Position

Time Range

Select date range to display


Display Options

sDMA Validation Time Series

Compare HRM baseline, calibrated secondary methods, and sDMA results. Click-drag to zoom, double-click to reset zoom.
Loading...

Sap Flux Density Conversion & Tree Water Use

Convert corrected heat pulse velocity (Vh) to sap flux density (Jv) and integrate across sapwood area for whole-tree water use.

About Sap Flux Density Conversion

Sap flux density (Jv) represents the volume of sap flowing through a unit area of sapwood per unit time.

  • Formula: Jv = Z × Vh
  • Z factor: Wood-specific conversion factor calculated from wood properties
  • Units: cm³/cm²/hr (equivalent to cm/hr sap velocity)
  • Sensors: Conversion applied to all sensor positions (inner + outer) for radial integration

Based on Burgess et al. (2001) after Barrett et al. (1995)


Recommended Workflow:
  1. Calculate raw heat pulse velocity (Vh)
  2. Apply spacing correction
  3. Apply wound correction (optional)
  4. Convert to sap flux density (Jv) - all sensors
  5. Integrate across sapwood area → tree-level water use

Conversion Settings

Select which method(s) to convert to flux density. sDMA methods combine multiple measurement approaches.
Select Methods:

Flux density will be calculated for all available sensor positions (inner and outer). Both sensors are required for radial integration across the sapwood.

Wood Properties:
Z factor (sap flux conversion factor) is calculated from your wood properties.

                        

Tree Water Use Integration

Integrate flux density across sapwood area to calculate whole-tree water use (Q).

Convert to flux density first.

Tree Dimensions:

Weighted average accounts for radial variation in sap flux density.

Export Flux Density Data

No flux density data available yet.

Export sap flux density data to CSV for further analysis.
Download CSV

1. Flux Density Conversion Summary

No flux density data yet. Click 'Convert to Sap Flux Density (Jv)' to begin.


                          

Flux density conversion is complete. Proceed to calculate tree water use (Q) below, or go to Tab 8b: Flux Density Validation to explore interactive plots.


2. Tree Water Use Summary

No tree water use data yet. Convert flux density first, then click 'Calculate Tree Water Use (Q)'.


                          

Tree water use calculation is complete. Go to Tab 8b: Flux Density Validation to view interactive plots, or proceed to Tab 9: Aggregation for temporal summaries.

Active Conversion Status

No conversions applied yet.


                        

Flux Density & Water Use Validation

Interactive visualization of flux density and tree water use results with filtering and time range controls.

Plot Controls

Methods to Display
Select specific methods to show

Sensor Position

Time Range

Select date range to display


Display Options

Sap Flux Density Time Series (Jv)

Sap flux density over time for all selected methods. Click-drag to zoom, double-click to reset zoom.
Loading...

Tree Water Use - Hourly (L/hr)

Whole-tree water use integrated across sapwood area. Click-drag to zoom, double-click to reset zoom.
Loading...

Tree Water Use - Daily (L/day)

Daily totals of tree water use. Grouped bars show multiple methods side-by-side.
Loading...

Velocity vs Flux Density Comparison

Visualize the conversion from heat pulse velocity (Vh) to sap flux density (Jv). Dashed line shows the Z factor relationship (Jv = Z × Vh).
Loading...

Daily Sap Flux Totals

Daily sap flux totals integrated over 24-hour periods.
Loading...

Temporal Aggregation

Aggregate flux density data to daily/weekly/hourly summaries for temporal analysis.

Aggregation Settings

Data Type

Temporal Aggregation

Aggregation Function

Plot Type

Summary Statistics

Aggregated Data

Loading...

Aggregated Data

Interactive Visualisation - Corrected HPV

View corrected heat pulse velocity data after applying spacing corrections and/or wound corrections.

Plot Controls

Methods to Display

Sensor Position
Tip: Select only one sensor for faster plot rendering with large datasets

Quality Flags to Display

Time Range

Set the date/time range to display. Updates automatically when you use the range slider.


Display Options

Data Cleaning & Filtering

Interpolation Settings

Creates a straight line between valid points. For single missing points, this is the simple average of before/after values.

Uses weighted average of surrounding valid points. Window size is automatically determined based on gap size. Good for noisy data.

Warning: Gaps > 3 hours are not recommended

Quality Flags to Interpolate

Filter by Method/Sensor (optional)

Leave all unchecked to apply to all methods/sensors




Heat Pulse Velocity Time Series

Loading...

Plot Information & Quality Control

Plot Information
Quality Control Summary

Pulse Trace Controls

Selected Pulse

                        
Sensor Position

Calculation Windows

Show calculation windows used by each method. Heat pulse injection is at t=0.


Pulse Temperature Trace

Loading...

Probe Configuration Builder

Create or edit probe configuration YAML files for use in sap flow analysis workflows.

Probe Configuration Builder

Create or edit probe configuration YAML files for use in sap flow analysis workflows. Upload an existing YAML to edit, or create a new configuration from scratch.

Load Existing Configuration (Optional)

Upload an existing probe configuration YAML file to edit.

Configuration Status

Probe Configuration



Define the spatial arrangement of thermistor sensors relative to the heater.


Probe Insertion

If probe is not fully inserted (e.g., due to external spacer for thin bark/sapwood), enter spacer thickness.


Physical dimensions and sensor positions.


Select calculation methods compatible with this probe configuration.

Compatible Methods

Recommended Methods

Select the subset of methods recommended for this configuration.

Export Configuration

Download your probe configuration as a YAML file.

Wood Properties Builder

Create or edit wood properties YAML files for use in sap flow analysis workflows.

Load Existing Configuration (Optional)

Upload an existing wood properties YAML file to edit.

Configuration Status

Wood Properties Configuration


Measure fresh weight immediately after sampling, then oven-dry until weight is stable.


Auto-calculated Densities:

These are calculated automatically from your weight and volume measurements.

Dual Density Measurements:

Measure BOTH densities on the same sample. Volume cancels out!


Physical constants for wood and sap. Defaults from Burgess et al. (2001).

Basic Properties:

Thermal Constants:

Initial Wound Configuration:

The wound around the probe affects the measured sapwood area.


Temporal Wound Tracking (Optional):

Dates auto-populate from loaded heat pulse data range (after any trimming). Adjust manually if needed to match original sensor installation dates.

Initial wound = drill bit + 2 × wound addition. Default: 2.0 + 2(0.3) = 2.6 mm


Calculate wood properties based on your measurements. Requires Method 1 or Method 2 measurements.


Optional - tree-specific measurements for scaling calculations.


Set acceptable ranges for quality control.

Export Configuration

Download your wood properties configuration as a YAML file. Derived properties will be automatically calculated if measurements are provided.

Reproducible Code Generation

Generate executable R scripts that reproduce your Shiny analysis workflow using sapfluxr functions.

About Code Generation

This tool generates executable R code that reproduces your entire Shiny analysis workflow.

  • Purpose: Scientific reproducibility
  • Output: Self-contained R script
  • Usage: Run in R console or RStudio
  • Benefits: Publication-ready, auditable, reusable

Tracked Steps:

                        

Export Options

No analysis steps recorded yet. Perform analysis in other tabs.


Download R Script

Generated R Script

Code will appear here after you perform analysis steps in the app.

This script reproduces your analysis. Copy or download to run in R.
                              

Analysis Steps Breakdown

Detailed breakdown of each analysis step: