Standardised protocol for sap flow measurements



Anthony O’Grady

Additional Authors

Patrick J. Mitchell, Stephen S.O. Burgess, Nathan Phillips


The development of thermometric approaches to quantifying plant water use has made a significant contribution to our understanding of the environmental and physiological controllers of the movement of water through the soil-plant-atmosphere continuum. This protocol outlines the common procedures to be used for making heat tracer based estimates of sap velocity in trees.


There are several methods for making measurements of sap velocity in trees, for example the compensation heat pulse method (Marshall, 1958), heat dissipation method (Granier, 1985), the heat ratio method (Burgess et al., 2001) are common but not the only approaches for estimating sap velocity. Although procedures and analyses protocols will vary among each of these procedures, this protocol outlines some of the common procedures for making measurements of sap velocity and flow in trees using these heat tracer based approaches.


  • Sap flow equipment/data loggers
  • Cordless drill
  • Drill guide, drill bits, small wire brush
  • Petroleum jelly
  • Diameter tape
  • Increment corer
  • Bark gauge
  • Methyl orange dye
  • Reflective insulation
  • Chisel and mallet

Units, terms, definitions

A suggested unifying nomenclature for sapflow measurements is given in (Edwards et al., 1996) and is followed here.

Nomenclature for velocities

Heat pulse velocity (vh, cm hr-1)

Heat pulse velocity corrected for wound effects (vc, cm hr-1)

Sap velocity within lumens (vl, cm hr-1)

Sap velocity on a sapwood area basis (vs cm hr-1)

Nomenclature for sap flows

Sap flow per tree (Q) (m3 per day per tree)

Sap flow per unit leaf area (Ql) (m3 m2 h-1)

Crown related sap flow (Qc) (can this be defined I am not sure what this means; probably reflects my ignorance)

Sap flow per unit conducting sapwood area (At the measurement point-Qs; m3 m-2 h-1)

Sap flow per unit basal area (at the measurement point-Qb; m3 m-2 h-1))

SI units should be used at all times. The appropriate units for each dimension are:

Depth: mm or m

Volume; m3or mm3

Time: s-1, hr-1or day-1 (not sure the minus should be here)

Area: mm-2or m-2 (not the – (ie minus) should be in the units here as area is area, not per unit area)


Tree Selection

This aspect will depend largely on the question being addressed, however, care should be taken to avoid trees that are obviously damaged or have been attacked by defoliating insects or stem borers (unless of course this was the purpose of the study!).

After selecting the tree to be instrumented some mensuration is required. Typical tree measures include; Diameter at 1.3 m (DBH), tree height, Diameter at crown break (Dcb), canopy cover, leaf area, diameter at the site of instrumentation (if this is different to DBH), bark thickness, sapwood area, heartwood area, sapwood density.

A range of tree sizes should be selected that spans the majority of the range of sizes present at the site for that species.

Tree preparation

  • Remove loose bark and prepare the site of instrumentation
  • Measure the diameter or circumference of the tree at the instrumentation site
  • Using the bark gauge determine the bark thickness. This can be achieved by pushing the bark gauge through the bark until it is resting in or preferably at the surface of the sapwood

Accurate determination of sapwood thickness is critical for determining where to position probes and for scaling sap velocities to tree and stand water use. Sapwood depth is usually determined using one of two methods, from wood cores or from sap velocity profiling.

Determining sapwood area using wood coring

  • Remove a small core (D ~ 5mm) from the tree using an increment corer
  • Sapwood area can be distinguished in some cases on the basis of a distinct colour change between the sapwood and heartwood. If not, sapwood staining may be required.
  • Using methyl orange (see standard operating procedure for stain) stain the wood core, heartwood will stain a red/purple colour (it may make a minute or two for the colour to develop)
  • Measure the thickness of the sapwood band

Determining sapwood area using sap velocity profiling(see notes on installing probes below)

  • Install a reference probe within the sapwood (for example at a depth of 10 mm under bark)
  • Install a second mobile probe nearby at a depth of 2 mm under bark. Throughout the period of peak sapflow move the mobile sensor deeper at increments of say 2 mm across the sapwood thickness to determine the sapwood-heartwood boundary
  • The simultaneous measurements of the fixed and mobile probes in conjunction with knowledge of the limit of sensitivity of the sapflow equipment will allow discrimination of the “no flow” boundary at the sapwood heartwood interface. It should be noted however, that the boundary can change across the year and is not a fixed point in many species.
  • See (Hatton, 1995) for more detail

Installing probes

Care must be taken when installing sapflow probes to minimize damage to the surrounding sapwood tissue and to ensure that upstream and downstream thermistors or thermocouples are correctly aligned. The positioning of the thermistors or thermocouples within the sapwood will depend normally on the questions being asked and the configuration of the probe sets. Commonly the measurement points will be stratified with depth to ensure representative sampling of sap velocity profiles.

  • A drill guide should be used to ensure that probes are accurately positioned and parallel to each other.
  • Mark the drill bits to the required depth (using correction fluid or marking pens or a strip of electrical tape attached to the drill bit).
  • Carefully drill holes to the required depth.
  • Minimize damage to surrounding sapwood tissue by ensuring drill bits are sharp and clean. Drill slowly and carefully. A wire brush is useful for removing wood shavings that accumulate in the drill flutes. Note: electric drills develop considerable RPM and may result in burning of the sapwood tissue. We suggest using a battery operated cordless drill.
  • Check alignment of holes by inserting oversized (overlong ) probes and measure the distance between each hole to ensure that probes are parallel.
  • Mark required depth on the sapwood probes and smear with petroleum jelly.
  • Insert probes into the tree.
  • If probes are not fully inserted, the exposed parts can be embedded in small pieces of polystyrene foam.
  • Cover the whole instrumentation site with reflective insulating material such as heavy duty insulating foil to reduce thermal disturbance of the instrumentation site. Alternatively, use “bubble wrap” that is then covered with aluminium foil.

Treatment of wood cores

Sapwood samples will be required for determining wood properties that are required for the analysis of the sap velocity data. The specific parameters required will vary among sapflow techniques, below are considered standard requirements

  • Wood cores should be taken for analysis of wood properties (do not use stained cores)
  • Store wood cores in waterproof plastic bags or vials away from direct sunlight
  • Separate heartwood and sapwood for analysis
  • Measure fresh weight of the sapwood
  • Measure the specific gravity of the sapwood sample
    • Set up a beaker of pure water at 20C on a tared balance
    • Set up a retort stand next to the balance with a dissecting needle attached that can be slid up and down
    • Attach the sapwood sample to the dissecting needle and carefully immerse the sapwood sample in the distilled water and record the weight (g)
  • Oven dry the wood sample at 80-90 C for ~ 48 hrs

Determination of wounding

Drilling into the sapwood creates a wound around the probe that must be accounted for to accurately calculate sap flow. The width of the wound on either side of the hole can be measured after the probes have been installed and the wound has stabilised. It is often necessary to implant -dummy’ probes (non-functioning probes) into several nearby trees and monitor the change in wound size with time. In general, the wound response stabilises within a couple of weeks of installing the probes. To measure the wound widths on the dummy probes:

  • Remove bark and cambial layers from an area > 20 mm around the probe hole.
  • Take a flat sapwood sample around the probe using a chisel and mallet to a depth of 10 – 20 mm.
  • Using a magnifying glass or dissecting microscope, record the average width of discolouration on either side of the hole.
  • Repeat on several trees, preferably avoiding trees already being sampled for water use and calculate a species average for the stable wound response.

Another method for assesing the extent of the probe implantation wound is to use the ‘closest intact conduit’ rule. Basic fuchsin dye must first be perfused through the xylem to stain conductive vessels. One technique is to drill a wide hole (e.g. 10 mm) approximately 20 mm directly beneath the site of probe installation and then create a small ‘skirt’ around the hole using flexible putty. Fill this receptacle until the hole becomes full of dye. This should be performed at dawn to prevent air being drawn into conduits severed by the drill. Allow some hours of transpiration, until dye has perfused past the probes. Then excise the site of probe implant using a saw or chisel. If the probe installation contains functioning probes, exchange these for expendable ‘dummy’ probes (e.g. correct sized stainless steel hypodermic needles). Sawing gently along the length of a probe creates a good cross-section along the plane of probe insertion. Once sectioned in this way, probes should be removed and further shaving should be made with a razor blade until the drill hole is perfectly bisected. Using a hand lens, look for pink-stained conduits. The absense of stain proves nothing (since conduits may have been damaged during drilling for dye injection), but the presence of some conductive conduits close to the drill hole can be interpreted as functional. The closest functional conduits can be considered to indicate the outer bounds of the wound width.

Maintenance and monitoring data quality

The appropriate duration of continuous sapflow monitoring from any one probe set depends on many factors including; the study objectives, plant growth rate and wounding effects. After each download the sapflow data needs to be checked closely for any signs of heater, thermocouple or thermistor malfunction, poor probe placement, degradation in signal quality from wounding or displacement of sensors due to rapid growth. Site maintenance requires ongoing vigilance of these and other potential problems to ensure data quality.

Additional measurements that improve the usefulness of sapflow data in the literature

Whenever possible, measurement of micrometeorological vairables (vapour pressure deficit, solar radiation and air temperature and soil moisture content at two depths (typically 10 cm and 50 cm) make the interpretation of sapflow data more comprehensive and useful to other users. It is best to not measure the micro-met variables in the understorey, it is better to measure these either in a large clearing (gap) at the study site, or on a tower with instruments located just above the canopy of the trees.


Scaling-up from tree-based measurements to stand-scale values is a valuable exercise and requires plot-scale estimates of sapwood area (m2 of sapwood per ha of ground). In monocultures (plantations) this requires measurement of sapflow for only one species. In natural woodlands and forests sapflow needs to be measured across as many species as are practicable. Often the inclusion (ie measuremetn of sapflow) of the dominant three tree species present at a site can account for 75 % or more of the total basal area of all the tree species present (this is not true at all sites).

Links to resources and suppliers

Hatton T J, Wu H-I (1995) Scaling theory to extrapolate individual tree water use to stand water use. Hydrological Processes 9, 527-540.

Kostner B, Granier A, Cermak J (1998) Sapflow measurement in forest stands: methods and uncertainties. Annals of Forest Science 55, 13-27.

Working group on sapflow:

Literature references

Burgess SSO, Adams MA, Turner NC, Beverley CR, Ong CK, Khan AAH, Belby TM (2001) An improved heat pulse method to measure low and reverse rates of sapflow in woody plants. Tree Physiology 21, 589-598.

Edwards WRN, Becker P, Cermak J (1996) A unified nomenclature for sapflow measurements. Tree Physiology 17, 65-67.

Granier A (1985) Une nouvelle methode pour la measure de seve brute le tronc des arbes. Annales Des Sciences Forestieres 42, 193-200.

Hatton TJ, Moore SJ, Reece PH (1995) Estimating stand transpiration in a Eucalyptus populnea woodland with the heat pulse method: measurement errors and sampling strategy. Tree Physiology 15, 219-227.

Marshall DC (1958) Measurement of sapflow in conifers by heat transport. Plant Physiology 33, 385-396.

Health, safety & hazardous waste disposal considerations

Download: Dimethyl yellow MSDS

Download: Standard Operating Procedure-dimethyl yellow

Leave a Reply