ArbNation Join free

Sensors on urban trees: sap-flow and soil moisture probes in research

Three studies fitted thermal dissipation sap-flow sensors and soil moisture probes to city trees in Maryland and in Dalian, China. Single maples over turf showed nearly three times the sap flux of maples in closed canopy.

Checked against the sources at the bottom of this page on October 21, 2026. Rules, fees and pay change: the source has the last word.

Researchers put two kinds of sensor on urban trees to measure water use: thermal dissipation probes in the sapwood, and soil moisture probes in the ground nearby. Three papers carried by the US Forest Service Treesearch library report what they recorded. Two come from the same Maryland research group, one in Scientific Reports in 2021 and one in Urban Forestry & Urban Greening in 2024. The third, in the Journal of Hydrology in 2011, comes from a park in Dalian, China, with a US Forest Service co-author.

The sap-flow sensor

Both Maryland papers use thermal dissipation sensors of the Granier type. Each sensor is two cylindrical probes 2 mm across, inserted 2 cm into the sapwood at about 1.4 m height and placed about 15 cm apart. The upper probe carries a heating wire on constant power, and the lower probe is an unheated reference. The voltage difference between the two is read every 30 seconds and stored as half-hour averages. A reading is converted to sap flux density, the flow of water per unit of sapwood area. The sensors are shielded with aluminum from rain and heat.

The Dalian study also used thermal dissipation probes, with 30 mm probes in most trees and 20 mm probes where the sapwood was shallow. Researchers bored the trunks of the sampled trees to measure sapwood area. A small square of bark was removed at 1.3 m height to expose the cambium for mounting, and the probes were sealed with foam and wrapped in foil.

The soil moisture probes

The 2024 Maryland paper installed soil moisture probes vertically to a depth of 30 cm at the dripline of each tree, recording every 30 minutes beside a weather station for rain, humidity and air temperature. The 2021 paper reports volumetric water content on the same basis. In Dalian, soil water monitors went in vertically to 100 cm at 25 cm intervals, in two sets: one under the crown and one in open ground. They also logged every 30 minutes.

What the Maryland studies found

The 2021 paper compared 18 mature red maples in three settings: single trees over turfgrass and a cluster of trees over turfgrass in Montgomery County, and closed-canopy forest with a leaf litter layer in Baltimore. Single trees had nearly three times the daily sum of sap flux density of closed-canopy trees. The averages over the measurement period were about 260, 195 and 91 g of water per square centimetre per day for single, cluster and closed-canopy trees. Single trees also responded more strongly to vapor pressure deficit.

The 2024 paper widened the sample to 41 trees of red maple, tulip poplar and sweetgum over the 2018 and 2019 growing seasons. For red maple the mean daily sum was 227, 163 and 87.5 g per square centimetre per day at the single, cluster and closed-canopy sites. The two papers print different figures for the same three settings, and the 2024 paper covers more trees and two growing seasons. In closed canopy the three species behaved much alike over the seasons, but tulip poplar was the most sensitive to drought on the 24-hour scale.

The authors also built whole-tree water use estimates from trunk diameter, species and setting. Water use rose with diameter. Plot-level transpiration was 9.1 percent of precipitation at the single-tree site, 71.2 percent at the cluster site and 52.4 percent at the closed-canopy site for April to October 2019. Both papers frame the findings around stormwater: they report that how urban forest is managed changes how much water trees return to the air.

What the Dalian study found

Three trees of each of four species were measured from June 25, 2008, to October 17, 2009, in a 350 square metre plot. Average monthly transpiration was 12.78 mm (standard error 0.73) for Cedrus deodara, 1.79 mm (0.16) for Zelkova schneideriana, 10.18 mm (0.55) for Euonymus bungeanus and 19.28 mm (2.24) for Metasequoia glyptostroboides. The authors flag the Metasequoia rates for further study because that species showed much higher sap flow and wider swings than the others.

Summer dry spells did not cut transpiration even when it exceeded rainfall, which the authors attribute to deeper soil moisture supply. Vapour pressure deficit was the dominant environmental factor, and the trees limited canopy conductance to hold back water loss under stress.

Gaps in the sources

The three papers describe sap-flow and soil moisture sensing only. No dendrometer study of urban trees turned up in the Treesearch records.

Sources