5:15 PM - 6:30 PM
[AHW20-P01] Contribution of understory vegetation to evapotranspiration in a tropical dry forest, Cambodia
Keywords:Tropical dry forest, Understory vegetation, Evapotranspiration, Leaf area index
Tropical dry forests cover about 20% of the global forested area, and their annual amounts of evapotranspiration are equivalent to more than 70% of annual rainfall, suggesting significant enhancements of water-vapor circulation between the land surface and the atmosphere over this land cover type. For example, Kanae et al. (2001) reported a decrease of rainfall around the Indochina peninsula engendered by forest decline. Although previous studies investigated evapotranspirational processes on the Indochina peninsula, Tanaka et al. (2008) pointed out the lack of research on dry deciduous forests. A dense understory vegetation generally covers the ground of dry deciduous forests, however, its contribution to evapotranspirational processes has not been comprehensively estimated. Here, we measured whole ecosystem evapotranspiration (ETW) and evapotranspiration from the understory vegetation (ETU) separately, and analyzed the factors controlling the understory contribution to evapotranspiration (ETU/ETW).
We measured evapotranspiration components at the dry deciduous forest located in Kratie province, Cambodia. The forest consists of deciduous trees, comprised of Dipterocarpus tuberculatus, Terminalia alata, Shorea obtusa and Xylia xylocarpa, and understory vegetation (Vietnamosasa pusilla). Both ETW and ETU were estimated by the band-pass eddy covariance method, applying the instruments installed above the sparse overstory canopy (leaf area index, LAI = 1.05) and understory vegetation, respectively. The overstory transpiration (TR) was estimated by sap flow technique (Granier, 1985) for all trees within the measurement plot (n=12). Canopy interception loss was calculated as the difference between gross rainfall and the sum of throughfall and stemflow. The period of analysis was from June 2010 to June 2011 (total available days was 354).
The total amount of ETW was 1064 mm corresponding to 74% of the annual rainfall. This percentage was typically similar with previous reports (e.g., Farrick and Branfireun, 2013). The dry season ETW (424 mm) was larger than rainfall (353 mm) indicating that this ecosystem uses water resources recharged during the wet season. The understory contribution to evapotranspiration (ETU/ETW) was 35%, and we concluded that understory vegetation is one of the most important components of evapotranspiration in this ecosystem. Seasonal changes of ETU/ETW were negatively correlated with LAI. We reviewed annual ETU/ETW values among the tropics, temperate, subhumid and boreal forests. We found that the annual ETU/ETW showed a negative correlation with LAI when LAI < 4. Thus, it is concluded that both seasonality and magnitude of ETU/ETW were controlled by LAI. On the other hand, overstory transpiration (TR) estimated by the original calibration coefficient showed high and positive correlation with the difference between ETW and the sum of ETU and interception loss, but we found clear underestimations of TR. When we applied recently published sets of calibration coefficients (Fuchs et al., 2017; Ouyang et al., 2018), the underestimation was improved. Applications of the coefficients of Ouyang et al. (2018) corrected almost all underestimations. We highly recommend the establishment of new calibration coefficients to evaluate TR precisely at dry deciduous forests in future studies.
This presentation is based on the following open access article: Iida, S., Shimizu, T., Tamai, K., Kabeya, N., Shimizu, A., Ito, E., Ohnuki, Y., Chann, S., Levia, D.F. (2020): Evapotranspiration from the understory of a tropical dry deciduous forest in Cambodia. Agricultural and Forest Meteorology, 295, 108170.
Cited papers:
Farrick and Branfireun (2013) Hydrol. Process. 27, 3254–3262.
Fuchs et al. (2017) Agric. For. Meteorol., 244, 151-161.
Granier (1985) Ann. Sci. For., 42, 193-200.
Kanae et al. (2001) J. Hydrometeorol., 2, 51-70.
Ouyang et al. (2018) Forests, 9, 162.
Tanaka et al. (2008) Agric. For. Meteorol., 148, 807-819.
We measured evapotranspiration components at the dry deciduous forest located in Kratie province, Cambodia. The forest consists of deciduous trees, comprised of Dipterocarpus tuberculatus, Terminalia alata, Shorea obtusa and Xylia xylocarpa, and understory vegetation (Vietnamosasa pusilla). Both ETW and ETU were estimated by the band-pass eddy covariance method, applying the instruments installed above the sparse overstory canopy (leaf area index, LAI = 1.05) and understory vegetation, respectively. The overstory transpiration (TR) was estimated by sap flow technique (Granier, 1985) for all trees within the measurement plot (n=12). Canopy interception loss was calculated as the difference between gross rainfall and the sum of throughfall and stemflow. The period of analysis was from June 2010 to June 2011 (total available days was 354).
The total amount of ETW was 1064 mm corresponding to 74% of the annual rainfall. This percentage was typically similar with previous reports (e.g., Farrick and Branfireun, 2013). The dry season ETW (424 mm) was larger than rainfall (353 mm) indicating that this ecosystem uses water resources recharged during the wet season. The understory contribution to evapotranspiration (ETU/ETW) was 35%, and we concluded that understory vegetation is one of the most important components of evapotranspiration in this ecosystem. Seasonal changes of ETU/ETW were negatively correlated with LAI. We reviewed annual ETU/ETW values among the tropics, temperate, subhumid and boreal forests. We found that the annual ETU/ETW showed a negative correlation with LAI when LAI < 4. Thus, it is concluded that both seasonality and magnitude of ETU/ETW were controlled by LAI. On the other hand, overstory transpiration (TR) estimated by the original calibration coefficient showed high and positive correlation with the difference between ETW and the sum of ETU and interception loss, but we found clear underestimations of TR. When we applied recently published sets of calibration coefficients (Fuchs et al., 2017; Ouyang et al., 2018), the underestimation was improved. Applications of the coefficients of Ouyang et al. (2018) corrected almost all underestimations. We highly recommend the establishment of new calibration coefficients to evaluate TR precisely at dry deciduous forests in future studies.
This presentation is based on the following open access article: Iida, S., Shimizu, T., Tamai, K., Kabeya, N., Shimizu, A., Ito, E., Ohnuki, Y., Chann, S., Levia, D.F. (2020): Evapotranspiration from the understory of a tropical dry deciduous forest in Cambodia. Agricultural and Forest Meteorology, 295, 108170.
Cited papers:
Farrick and Branfireun (2013) Hydrol. Process. 27, 3254–3262.
Fuchs et al. (2017) Agric. For. Meteorol., 244, 151-161.
Granier (1985) Ann. Sci. For., 42, 193-200.
Kanae et al. (2001) J. Hydrometeorol., 2, 51-70.
Ouyang et al. (2018) Forests, 9, 162.
Tanaka et al. (2008) Agric. For. Meteorol., 148, 807-819.