why you should avoid sapflow+
There are several methods to estimate sap flow, or plant water use, via heat pulse velocity (HPV) methods. However, all known methods have at least one disadvantage that limits the use of that method to estimate sap flow in all plants and under all conditions. To address these shortcomings, Vandegehuchte and Steppe (2012) proposed the Sapflow+ method. According to that publication, the Sapflow+ method is more theoretically correct than other HPV methods and it has the potential to measure reverse, slow and high rates of sap flow. Vandegehuchte and Steppe (2012) also advanced the Sapflow+ method as it can measure stem water content as well as sap flow. However, on closer inspection the Sapflow+ method is not the golden or universal method many sap flow scientists hoped it would be. Here, we outline some reasons why sap flow scientists should avoid using the Sapflow+ method to estimate sap flow in plants.
The accuracy of the Sapflow+ is extremely poor. Vandegehuchte and Steppe (2012) demonstrated that, on average, the Sapflow+ method has a 38% error in estimating true sap flow. To test the accuracy of sap flow sensors, a robust method is to measure plant water use via the sap flow method and test it, via a linear regression, against an independent measure of plant water use such as a weighing lysimeter. If the method is accurate, the linear regression equation will have a slope of one. Vandegehuchte and Steppe (2012) tested the Sapflow+ method against an independent measure of plant water use and found the slope to range between 0.46 and 0.82, with an average of 0.62. Therefore the Sapflow+ method underestimates true sap flow with an average error of 38%.
2. probe configuration
The Sapflow+ method requires at least four probes to be inserted into the plant’s stem: the heater probe, plus a downstream, upstream and tangential temperature probe. The four probe configuration means that the Sapflow+ method can only be used on larger stems or large trees. More importantly, with four probes the probability of probe misalignment, and associated errors, increases significantly. Probe misalignment is the primary cause or errors in HPV methods (Burgess et al 2001) with a 2mm probe misalignment leading to a 100% error in sap flow estimation (Bleby et al 2004). Probe misalignment not only causes error in the HPV equation, but associated equations such as the accurate calculation of thermal diffusivity or conductivity (Ren et al 2017). It is extremely difficult to accurately install two or three probes, let alone four probes. Furthermore, Vandegehuchte and Steppe (2012) did not offer any corrections or advice on probe misalignment and noted that a 5% error in probe misalignment will lead to a 16% error in heat velocity calculations.
3. sample size and cost
Aside from accuracy, scientists require large sample size and replication in their studies. Unfortunately, due to the probe configuration and complexity in computing algorithm, the Sapflow+ method requires a significant investment in data logger capacity in order to reach any meaningful sample size. Vandegehuchte and Steppe (2012) were limited to one sensor on their data logger where with other methods up to eight sensors can be supported. Therefore, for every one data logger other HPV methods require, the Sapflow+ requires eight data loggers which adds significantly to costs, reduces sample size and replication, and lowers the robustness of the study or measurement campaign.
4. sensor calibration
With the Sapflow+ method, every sensor must be individually calibrated. The sensor must be calibrated for an accurate measurement of q or heat input. The calibration method is tedious and labour intensive and adds to the cost of the project.
The Sapflow+ method is highly complex and difficult to understand. Other HPV methods, such as T-max and compensation heat pulse method, are intuitive and based on simple algebraic equations. The Sapflow+ method, in contrast, is based on complex equations and even requires the purchasing of commercial software in order to estimate several parameters.
6. stem water content
One advantage the Sapflow+ method seemed to have over all other HPV methods was its ability to measure sap flow and stem water content. However, the method proposed by Vandegehuchte and Steppe (2012) to measure stem water content is well known and is directly related to the conduction/convection theory of heat movement in materials. Soil scientists have long known that the conduction/convection theory can measure water content and they have been using the method to measure soil water content for many decades (e.g. Ochsner et al, 2003). The method to measure stem water content outlined by Vandegehuchte and Steppe (2012) can also be used in other HPV methods by a simple alteration to the data logger algorithm. Other scientists have used such methods to measure sap flow and stem water content (e.g. Lopez-Bernal et al, 2012) and the method is not unique to Sapflow+.
Sapflow+ may be theoretically more correct than other HPV methods in measuring the heat conduction and convection in plant stems. However, even Vandegehuchte and Steppe (2013) argued that the traditional HPV methods still yield correct results. Furthermore, the Sapflow+ method has a large error, requires numerous probes and large data logging capacity, and is highly complex and requires commercial software to estimate heat velocity. Given these significant disadvantages and limitations, the Sapflow+ method should be avoided.
Bleby et al (2004). Functional Plant Biology, 31, 645-658.
Burgess et al (2001). Tree Physiology, 21, 589-598.
Lopez-Bernal et al (2012). Tree Physiology, 32, 1420-1429.
Ochsner et al (2003). Vadose Zone Journal, 2, 572-579.
Ren et al (2017). Agricultural and Forest Meteorology, 232, 176-185.
Vandegehuchte and Steppe (2012). New Phytologist, 196, 306-317.
Vandegehuchte and Steppe (2013). Functional Plant Biology, 40, 213-223.