The exchange of water and energy between a tropical peat forest and the atmosphere: Seasonal trends and comparison against other tropical rainforests. Authors: Angela C. I. Tang, Paul C. Stoy, Ryuichi Hirata, Kevin K. Musin, Edward B. Aeries, Joseph Wenceslaus, Mariko Shimizu, and Lulie Melling NOTICE: this is the author’s version of a work that was accepted for publication in Science of the Total Environment. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Science of the Total Environment, VOL# 683, (September 2019) DOI# 10.1016/j.scitotenv.2019.05.217. Tang, Angela C. I., Paul C. Stoy, Ryuichi Hirata, Kevin K. Musin, Edward B. Aeries, Joseph Wenceslaus, Mariko Shimizu, and Lulie Melling. "The exchange of water and energy between a tropical peat forest and the atmosphere: Seasonal trends and comparison against other tropical rainforests.." Science of the Total Environment 683 (30 May 2019): 166-174. DOI:10.1016/ j.scitotenv.2019.05.217. Made available through Montana State University’s ScholarWorks scholarworks.montana.edu 1 The exchange of water and energy between a tropical peat forest and the atmosphere: Seasonal trends and comparison against other tropical rainforests Angela C. I. Tang1,2, Paul C. Stoy1, Ryuichi Hirata3, Kevin K. Musin2, Edward B. Aeries2, Joseph Wenceslaus2, Mariko Shimizu4, and Lulie Melling2 1Department of Land Resources and Environmental Sciences, Montana State University, Bozeman, MT 59717, USA 2Sarawak Tropical Peat Research Institute, Lot 6035, Kuching-Kota Samarahan Expressway, 94300 Kota Samarahan, Sarawak, Malaysia 3Center for Global Environmental Research, National Institute for Environmental Studies, Tsukuba 305-8506, Japan 4Civil Engineering Research Institute for Cold Region, Sapporo 062-8602, Japan Correspondence to: Paul C. Stoy (paul.stoy@gmail.com) Abstract. Tropical rainforests control the exchange of water and energy between the land surface and the atmosphere near the equator and thus play an important role in the global climate system. Measurements of latent (LE) and sensible heat exchange (H) have not been synthesized across global tropical rainforests to date, which can help place observations from individual tropical forests in a global context. We measured LE and H for four years in a tropical peat forest ecosystem in Sarawak, Malaysian Borneo using eddy covariance, and 2 hypothesize that the study ecosystem will exhibit less seasonal variability in turbulent fluxes than other tropical ecosystems database as soil water is not expected to be limiting in a tropical forested wetland. LE and H show little variability across seasons in the study ecosystem, with LE values on the order of 11 MJ m-2 day and H on the order of 3 MJ m-2 day-1. Annual evapotranspiration (ET) did not differ among years and averaged 1579 ± 47 mm year-1. LE exceeded characteristic values from other tropical rainforest ecosystems in the FLUXNET2015 database with the exception of GF-Guy near coastal French Guyana, which averaged 8-11 MJ m-2 day-1. The Bowen ratio (Bo) in tropical rainforests in the FLUXNET2015 database either exhibited little seasonal trend, one seasonal peak, or two peaks. Volumetric water content (VWC) and VPD explained a trivial amount of the variability of LE and Bo in some of the tropical rainforests including the study ecosystem, but were strong controls in others, suggesting differences in stomatal regulation and/or the partitioning between evaporation and transpiration. Results demonstrate important differences in the seasonal patterns in water and energy exchange across different tropical rainforest ecosystems that need to be understood to quantify how ongoing changes in tropical rainforest extent will impact the global climate system. Keywords 3 Eddy covariance; FLUXNET; latent heat flux; sensible heat flux; tropical peat forest; tropical rainforest 1 Introduction Tropical ecosystems play a critical role in the global climate system by regulating the amount of heat and water that enters the atmosphere near the equator to drive deep convection. Ongoing changes to tropical forest extent (Kim et al., 2015) have impacted regional and global climate (Avissar and Werth, 2005; Bala et al., 2007; Medvigy et al., 2011; Werth, 2002), emphasizing the importance of understanding how tropical rainforests exchange water and energy with the atmosphere and how differences in their response to environmental variability may determine these dynamics. Tropical rainforests exhibit seasonal patterns of energy and water exchange in response to wet and dry seasons; many studies report a doubling of the Bowen ratio (Bo, the sensible heat flux H divided by the latent heat flux LE) or more during the dry season (Da Rocha et al., 2004; Gerken et al., 2018). These findings are consistent with the notion that tropical rainforest tree species are isohydric, meaning that they strictly regulate canopy conductance in response to water deficits (Fisher et al., 2006; Konings and Gentine, 2016). At the same time, other studies using modeling approaches suggest that some tropical forests should exhibit anisohydric water regulation strategies (Inoue et al., 2017; Kumagai and Porporato, 2012) if water stress is less of a risk. Yet others argue that different tree species exhibit a range of 4 isohydric to anisohydric behavior (Klein, 2014), leaving it unclear if the conductance of diverse tropical forest canopies share similar responses to ongoing increases in the atmospheric vapor pressure deficit (VPD) (Novick et al., 2016; Sulman et al., 2016) and decreases in soil moisture (Jung et al., 2011). Gross primary productivity (GPP) varies considerably in response to increasing VPD across different tropical rainforests (Kiew et al., 2017; Wu et al., 2017); a recent synthesis of eddy covariance sites determined that VPD is significantly related to GPP across 12 tropical rainforerst sites, but to a widely varying degree (Fu et al., 2018). Namely, VPD explained less than 4% of the variability in GPP across three ecosystems (GF-Guy, GH- Ank, and VU-Coc) that happened to be near the ocean where water may be assumed to be more available than the other nine ecosystems where VPD explained 20% of the variability of GPP on average (Fu et al., 2018). If the control over GPP by VPD is due to stomatal closure, energy balance partitioning may also be strongly constrained by VPD in tropical rainforests that are characterized by a large transpiration fraction. The response of eddy covariance-measured H and LE to seasonal water and energy availability across different tropical ecosystems has not been studied to date. Here, we describe the seasonal variability in H and LE, and thereby the Bo, and their response to net radiation (Rn), VPD, and soil water availability (via volumetric soil water content, SWC) in a tropical peat rainforest ecosystem in Malaysian Borneo, hereafter abbreviated MY-MLM. We place these observations in the context of the seasonal variability 5 of surface-atmosphere water and energy exchange measured at other tropical rainforest ecosystems across all five continents on which tropical rainforests exist. We hypothesize that the study ecosystem will exhibit less seasonal variability in water and energy exchange compared to other tropical rainforests because soil water is not expected to be limiting. We also hypothesize that VPD will be of lesser importance to LE and H in the study ecosystem versus other tropical rainforests given the findings of Fu et al. (2018). We place particular emphasis on eddy covariance energy balance closure and uncertainty estimation in the study ecosystem, noting that whereas tropical rainforest ecosystems on average have higher energy balance closure than most other ecosystems (Gerken et al., 2018; Stoy et al., 2013); an analysis of energy balance closure cannot be excluded from a careful investigation of surface- atmosphere water and energy exchange using the eddy covariance technique. 2 Methods 2.1 Micrometeorological Measurements at Maludam National Park, Malaysia The primary study ecosystem, MY-MLM, is a tropical peat forest ecosystem in Maludam National Park in the Betong Division of Sarawak, Malaysia. Dominant vegetation in the overstory includes Shorea albida, Gonystylus bancanus and Stemonurus spp (Anderson, 1972) with an average canopy height of 35 m. Peat thickness is 8 m in the vicinity of the eddy 6 covariance tower at 1º27'13” N, 111º8'58” E. Mean annual precipitation during the measurement period was 2798 mm year-1 and mean annual temperature was 26.9 °C. Rn and global radiation (Rg) were measured at 41 m using a CNR4 net radiometer (Kipp & Zonen, Delft, The Netherlands). Two LI-190SB quantum sensors (LI-COR) were likewise mounted at 41 m and pointed downward and upward to measure incoming and outgoing photosynthetic photon flux densities (PPFD). Air temperature (Tair) relative humidity (RH) and thereby VPD were measured at 11 and 41 m using CS215 temperature and relative humidity probes (Campbell Scientific). The tower was also equipped with a three-cup anemometer and wind vane (01003-5 R.M. Young Co., Traverse, MI, USA) at 41 m to measure wind speed (WS) and wind direction in addition to the sonic anemometer. P was measured by a TE525MM tipping-bucket rain gauge (Texas Electronics, Dallas, Texas, USA) 1 m above the ground surface in an open area located ca. 5 m from the tower. Soil temperature was measured with platinum resistance thermocouples at 5 and 10 cm below the ground surface. Volumetric soil water content was measured at a depth of 30 cm using CS616 time domain reflectometry (Campbell Scientific). All meteorological variables were continuously recorded using CR3000 and CR1000 dataloggers at a sampling frequency of 5 min and averaged over each 30 min period except WT, which was monitored on a half-hourly basis using a water level logger (DL/N 70 STS Sensor Technik Sirnach AG, Sirmarch, Switzerland). 7 2.2 Eddy Covariance Measurements at MY-MLM H and LE were measured at MY-MLM using the eddy covariance technique from 2011 to 2014 in the roughness sublayer at 41 m. The eddy covariance system was comprised of a LI-7500A open-path CO2/H2O analyzer (LI-COR Inc., Lincoln, NE, USA) coupled to a CSAT3 three- dimensional sonic anemometer (Campbell Scientific). 10 Hz observations were stored on a CR3000 Datalogger (Campbell Scientific) and half-hourly flux measurements were calculated using EddyPro (LI-COR) with WPL correction (Webb et al., 1980), block averaging, double rotation to align the sonic anemometer with the mean flow (Wilczak et al., 2001), and time-lag and frequency response corrections (Massman, 2000). 2.3 Soil Heat Flux Soil heat flux (G) was not measured at MY-MLM, so we estimated it using a model. Assuming that peat soils have porosity of 80% and water content of 50% (Monteith & Unsworth, 1990) , G can be calculated as the sum of the flux at depth z and the rate of change heat storage above the depth z: G = −𝜆 %&'()*%+ + 𝑐. /%&'()*%0 1 ∆𝑧 (1) where 𝜆 is thermal conductivity (0.29 𝑊 𝑚<= ℃ 10 hPa, the value at which VPD often constrains canopy conductance to water vapor flux (Lasslop et al., 2010). We also investigate the relationship between VWC, VPD, and Bo for positive Bo values across the different tropical forest ecosystems to explore how variability in soil supply of water and atmospheric demand for water impact energy balance partitioning. 10 3 Results 3.1 Water and Energy Fluxes from MY-MLM LE at MY-MLM decreased from the wet season (with characteristic values on the order of 11 MJ m-2 day-1), to the dry season (with characteristic values on the order of 10 MJ m-2 day- 1) (Figure 2) and largely followed seasonal patterns in Rn, which varied from 13-14 MJ m-2 day- 1 on average. H exhibited less seasonal variability with average daily sums on the order of 3 MJ m-2 day-1. Evapotranspiration (ET) calculated from LE measurements did not significantly differ among the study years and averaged 1579 ± 47 mm year-1 (mean ± standard deviation) (Figure 3). As a consequence, the evaporative fraction (EF = Rn/LE) and Bo exhibited little interannual differences (data not shown) nor differences during the daytime regardless of dry or wet season (Figure 4). The EF was close to 0.74 and the Bo near 0.33 during daytime, and the Bo was more negative (and the EF more positive) during nighttime periods of the dry season across all years (Figure 4). The Bo increased in response to Rn across both wet and dry seasons, indicating that additional energy is disproportionally partitioned to H as Rn increases (Figure 5). Whereas ET did not exhibit annual differences as noted (Figure 3), there was a decrease in shortwave albedo during the later years of measurement (2013-2014) from 8.5% during mid-day earlier in the measurement period to 8% later in the measurement period (Figure 6). 11 LE did not decrease in response to increasing VPD during any year or season of measurement at MY-MLM (Figure 7). Bo decreased during the dry season as VPD approached 1 kPa during the dry seasons of the first two years of the measurement record (Figure 8), demonstrating that energy was partitioned even more toward LE during periods when VPD was high. 3.2 Energy Balance Closure at MY-MLM The mean of the daily sum for G during the four years of measurement period was -0.26 ± 0.58 MJ m-2 d-1, indicating a net flow of heat from the soil to the atmosphere. Energy balance closure (calculated as the relationship between the daily sum of H plus LE versus Rn minus the model for G) increased from 67% during the wet season to 74% during the dry season (Figure 9) and averaged 70% across the entire measurement period, somewhat less than globally-distributed eddy covariance sites from the FLUXNET La Thuile database (84 ± 20%) and tropical sites from the La Thuile database (94% ± 16%) (Stoy et al., 2013). 3.3 Water and Energy Fluxes from Global Tropical Rainforest Ecosystems The magnitude of average daily LE across the calendar year tended to be higher at MY-MLM (11 MJ m-2 day-1) than most other tropical rainforest ecosystems in the FLUXNET2015 database with the exception of GF-Guy in French Guyana (Figure 10a). Mean daily LE across 12 the calendar year exhibited diverse patterns across the different study ecosystems, with average daily fluxes differing by four times (ca. 2 MJ m-2 day-1) from dry to wet season at GH-Ank in Ghana to a seasonally invariant 8 ± 1 MJ m-2 day-1 at a tropical rainforest on Peninsular Malaysia (MY-PSO). H at MY-MLM was less variable than most other study ecosystems despite large differences in wet and dry season precipitation that differed for example from more than 550 mm month-1 in December 2013 to less than 50 mm month-1 in July 2013. Daily average H reached as high as ca 8 MJ m-2 day at a tropical rainforest in Australia (AU-Rob), where net negative daily H values were also observed, similar to GF-Guy (Figure 10b). As a consequence of the seasonal patterns in LE and H, the seasonal pattern of Bo can be characterized into sites that have two calendar year peaks (MY-PSO and PA-SPn), sites with little seasonal pattern in Bo (BR-Sa1 and BR-Sa3), and sites with characteristically low Bo values less than 0.25 early in the calendar year and larger values later (Au-Rob, GF-Guy, GH-Ank, and the study ecosystem, MY-MLM) (Figure 10c). 3.3 The Response of Turbulent Fluxes to Variability in Water Supply and Demand The relationship between VPD (for VPD values greater than 1 kPa) and the residual of the relationship between Rn and LE was significant at the common P < 0.05 level but rarely explained more than 2% of its variance with the exception of GF-Guy and PA-SPn at which 13 nearly 5% and over 11% of the variance of the model residual was explained by VPD, respectively (Fig. 11). More than 10% of the variability in the residual of the linear relationship between Rn and H is explained by VPD in these two ecosystems (data not shown). As a consequence, VPD explained 7% of the variability of Bo at GF-Guy and 20% of its variability at PA-SPn (Table 2). VPD also explained 19% of the variability of Bo at AU-Rob, which like MY-MLM demonstrated a positive relationship between VPD and Bo. VPD and VWC explained less than 1% of the variability in Bo at MY-MLM (Table 2), and VWC explained at most 6% of the variability in the residual of the relationship between Rn and LE across all sites. 4 Discussion 4.1 Water and Energy Fluxes from MY-MLM: Energy Balance Closure and Seasonal Patterns Annual rainfall decreased during the measurement period at MY-MLM (3290, 2941, 2688 and 2272 mm for 2011, 2012, 2013 and 2014, respectively), but annual ET was comparable between years; 1568, 1616, 1516 and 1614 mm for 2011, 2012, 2013 and 2014, respectively (Figure 3). These observations suggest that ET was insensitive to observed variability in P, and also that water available for groundwater recharge of surface flow decreased across the measurement period from ca. 1700 mm in 2011 to ca. 660 mm in 2014. These observations are 14 consistent with the notion that ET is a conserved quantity compared to other terms in the water balance in energy-limited ecosystems (Oishi et al., 2010). Turbulent fluxes at MY-MLM were poorly related to water limitation across the observed range of variability of VWC and VPD. Daytime EF and Bo were insensitive to dry or wet season at MY-MLM (Figure 4), but the surface cooled more quickly at night during the dry season, resulting in Bo values that were more negative. These observations are consistent with a more rapid nighttime cooling during the dry season when WT heights were characteristically beneath the soil surface at MY-MLM (Tang et al., 2018). In other words, the largest differences between the dry and wet season in LE and H occurred at night and are consistent with the influence of the heat capacity standing water on ecosystem energy fluxes rather than the influence of this water on daytime energy partitioning. The non-closure of surface energy balance is common in eddy covariance measurements (Stoy et al., 2013; Wilson et al., 2002) and is often attributed to a number of causes such as inadequacy of instrument system, mismatch in footprint between Rn and eddy fluxes, neglected energy sinks, advective flux divergence, low and high frequency loss of turbulent fluxes at individual sites (Foken, 2008; Wilson et al., 2002). We note that the present analysis of energy balance closure excludes unmeasured ecosystem heat fluxes a model for G (equation 1), which are minor terms in the energy balance at the diurnal time scales analyzed here (Leuning et al., 2012). Because energy balance closure was lower during the wet season 15 when the water table characteristically exceeded the soil surface (Tang et al., 2018), our observations are consistent with the notion that the advective exchange of heat due to flowing water is a nontrivial term in the energy balance, as has been found in other wetland ecosystems (Barr et al., 2013). Energy balance closure values from the present analysis are similar to wetland ecosystems in the La Thuile FLUXNET database (0.76), which had the lowest average energy balance closure of any La Thuile database ecosystem (Stoy et al., 2013). Results of this study and others suggest that additional instrumentation is needed to capture advective energy flux in ecosystems characterized by flowing water. 4.2 Surface-Atmosphere Energy Flux in Tropical Rainforest Ecosystems The mean annual ET at MY-MLM (1579 mm yr-1) was similar to a hydrologically disturbed peat forest in Kalimantan, Indonesia (ID-Pag, 1636 mm yr-1, Hirano et al., 2015) and a rainforest at Lambir Hills National Park in Sarawak, Malaysia (1545 mm yr-1, Kumagai et al., 2005), but higher than a rainforest in Peninsular Malaysia (MY-PSO, 1287 mm yr-1, Kosugi et al., 2012), central Amazonian forests (BR-Ma1, 1123 mm yr-1, Malhi et al., 2002; Hutyra et al., 2007), and an afforested site (PA-SPn, 1114 mm yr-1) and a pasture site (1034 mm yr-1) in Panama (Wolf et al., 2011). These observations are consistent with experimental hypothesis that LE (and by conversion ET) would be higher at MY-MLM than other tropical rainforest sites due to the consistent availability of water in a tropical peat forest wetland ecosystem and 16 relative lack of seasonality, as also demonstrated by the insensitivity of daytime EF and Bo to wet and dry season. Tropical rainforests near the tropical/subtropical boundary, namely AU- Rob, exhibited daily average Bo approaching 1.5, and Amazonian terra firme rainforests exhibited sharp increases in Bo to ca. 0.7 during conditions of high Rn (exceeding 900 W m-2) (Gerken et al., 2018), suggesting that reduction in LE in response to dry conditions is an important feature of these tropical rainforests, recalling that LE continued to increase in response to VPD at MY-MLM (Fig. 7). Bo rarely exceeded 0.3 at MY-MLM, even during the dry season, suggesting that water is consistently available for ET, and seasonal patterns of Bo in other tropical forested ecosystems indicates large differences in their hydrologic function with important implications for global heat and moisture transport (Fig. 8). The residual of the relationship between LE and Rn as well as energy balance partitioning was poorly described by VPD and VWC across the range of forests and environmental variability explored here, except at AU-Rob and PA-SPn where they explained 19-30% of the variability in these terms and to a lesser degree GF-Guy, where they explained 4-7% (Table 1; Fig. 11). AU-Rob is near the tropical/subtropical ecotone and PA-SPn is a re- establishing forest with a developing root system (Wolf et al., 2001a,b,c), and VPD explained 11-32% of the variability of GPP in these ecosystems (Fu et al., 2018) suggesting a strong coupling between carbon dioxide, turbulent fluxes, and hydrologic variability. Turbulent fluxes at most other tropical forests studied here, including MY-MLM, were poorly related to VPD 17 and VWC, and there is little evidence that the tropical peat forest MY-MLM is any less sensitive to water availability and demand given the range of environmental variability of the available data, and even though annual P was nearly 1/3 lower during the last year of measurement at MY-MLM than the first. The diversity of responses of turbulent fluxes to water availability across tropical forests could be due to a number of factors that require further investigation. Trees may have been insufficiently water stressed to elicit a substantial response in energy balance partitioning, but it is reasonably well-established that canopy conductance to water flux decreases in response to VPD, especially above 1 kPa (Lasslop et al., 2010), which was frequently exceeded in the observations (Fig. 11). Greater evaporation under elevated VPD may therefore be responsible, and the partitioning of evaporation and transpiration at the ecosystem scale remains a critical area of research for understanding the global water cycle (Stoy et al., 2019). As noted above, variability in tree isohydricity (Klein, 2014) may be responsible for the observations as well as the physical and biological environment in which plants compete for water, which impacts the response of canopy conductance to water availability (Mrad et al., 2019). 18 5 Conclusion The experimental hypothesis that H and LE would be less variable and less related to water availability at MY-MLM than other tropical rainforest ecosystems could not be falsified using observations and approaches used here. ET was relatively constant at MY-MLM despite decreasing rainfall over the four-year study period, consistent with the notion that ET is a conserved quantity in radiation-limited ecosystems. Results demonstrate important differences in the seasonal patterns in water and energy exchange in tropical rainforest ecosystems that exhibit large differences in the response of canopy processes to atmospheric moisture stress via VPD. These diverse response of turbulent fluxes to hydrologic variability suggest considerable differences in ecosystem hydrology among tropical rainforests that need to be understood to quantify how ongoing changes in tropical rainforest extent will impact the global climate system. Acknowledgements This work is supported by both the Sarawak State Government and the Federal Government of Malaysia. PCS acknowledges support from the U.S. National Science Foundation Department of Environmental Biology grant #1552976 and the U.S. Department of Energy as part of the GoAmazon project (Grant SC0011097). This work used eddy covariance data acquired and shared by the FLUXNET community, which includes AmeriFlux, AfriFlux, AsiaFlux, 19 CarboAfrica, CarboEuropeIP, CarboItaly, CarboMont, ChinaFlux, Fluxnet-Canada, GreenGrass, ICOS, KoFlux, LBA, NECC, OzFlux-TERN, TCOS-Siberia, and USCCC networks. The FLUXNET eddy covariance data processing and harmonization was carried out by the European Fluxes Database Cluster, AmeriFlux Management Project, and Fluxdata project of FLUXNET, with the support of CDIAC and ICOS Ecosystem Thematic Center, and the OzFlux, ChinaFlux and AsiaFlux offices. We would also like to thank Takashi Hirano for his advice and assistance in this study, and Tobias Gerken, Rob Payn, Mark Greenwood and Benjamin Poulter for valuable comments on the manuscript. References Albinet, C., Koleck, T., Le Toan, T., Borderies, P., Villard, L., Hamadi, A., Laurin, G. V., Nicolini, G. and Valentini, R.: First results of AfriScat, a tower-based radar experiment in African forest, in Geoscience and Remote Sensing Symposium (IGARSS), 2015 IEEE International, pp. 5356–5358, IEEE., 2015. Anderson, J. A. R.: Synoptical Key for the Identification of the Trees of the Peat Swamp Forests of Sarawak, Forest Department, Sarawak., 1972. Avissar, R. and Werth, D.: Global hydroclimatological teleconnections resulting from tropical deforestation, J. Hydrometeorol., 6(2), 134–145, 2005. Bala, G., Caldeira, K., Wickett, M., Phillips, T. J., Lobell, D. B., Delire, C. and Mirin, A.: 20 Combined climate and carbon-cycle effects of large-scale deforestation, Proc. Natl. Acad. Sci. USA, 104, 6550–6555, 2007. Barr, J. G., Fuentes, J. D., Delonge, M. S., O’Halloran, T. L., Barr, D. and Zieman, J. C.: Summertime influences of tidal energy advection on the surface energy balance in a mangrove forest, Biogeosciences, 10(1), 501–511, doi:10.5194/bg-10-501-2013, 2013. Bonal, D., Bosc, A., Ponton, S., Goret, J.-Y., Burban, B., Gross, P., Bonnefond, J.-M., Elbers, J., Longdoz, B., Epron, D., Guehl, J.-M. and Granier, A.: Impact of severe dry season on net ecosystem exchange in the Neotropical rainforest of French Guiana, Glob. Chang. Biol., 14(8), 1917–1933, doi:10.1111/j.1365-2486.2008.01610.x, 2008. Bradford, M. G., Metcalfe, D. J., Ford, A., Liddell, M. J. and McKeown, A.: Floristics, stand structure and aboveground biomass of a 25-ha rainforest plot in the Wet Tropics of Australia, J. Trop. For. Sci., 543–553, 2014. Canadell, J., Jackson, R., Ehleringer, J., Mooney, H. A., Sala, O. E. and Schulze, E.-D.: Maximum rooting depth of vegetation types at the global scale, Oecologia, 108, 583–595, doi:10.1007/BF00329030, 1996. Fisher, R. A., Williams, M., Do Vale, R. L., Da Costa, A. L. and Meir, P.: Evidence from Amazonian forests is consistent with isohydric control of leaf water potential, Plant. Cell Environ., 29(2), 151–165, doi:10.1111/j.1365-3040.2005.01407.x, 2006. Foken, T.: The energy balance closure problem: An overview, Ecol. Appl., 18(6), 1351– 21 1367, doi:10.1890/06-0922.1, 2008. Gerken, T., Ruddell, B. L., Fuentes, J. D., Araújo, A., Brunsell, N. A., Maia, J., Manzi, A., Mercer, J., dos Santos, R. N. and von Randow, C.: Investigating the mechanisms responsible for the lack of surface energy balance closure in a central Amazonian tropical rainforest, Agric. For. Meteorol., 255, 92-103, doi: 10.1016/j.agrformet.2017.03.023, 2018. Goulden, M. L., Miller, S. D. and Da Rocha, H. R.: Nocturnal cold air drainage and pooling in a tropical forest, J. Geophys. Res. Atmos., 111(D8), 2006. Hirano, T., Kusin, K., Limin, S. and Osaki, M.: Evapotranspiration of tropical peat swamp forests, Glob. Chang. Biol., 21(5), 1914–1927, doi:10.1111/gcb.12653, 2015. Hollinger, D. Y. and Richardson, A. D.: Uncertainty in eddy covariance measurements and its application to physiological models, Tree Physiol., 25, 873–885, 2005. Hutyra, L. R., Munger, J. W., Saleska, S. R., Gottlieb, E., Daube, B. C., Dunn, A. L., Amaral, D. F., de Camargo, P. B. and Wofsy, S. C.: Seasonal controls on the exchange of carbon and water in an Amazonian rain forest, J. Geophys. Res. Biogeosciences, 112, doi:10.1029/2006JG000365, 2007. Inoue, Y., Ichie, T., Kenzo, T., Yoneyama, A., Kumagai, T., Nakashizuka, T.: Effects of rainfall exclusion on leaf gas exchange traits and osmotic adjustment in mature canpopy trees of Dryobalanops aromatica (Sipterocarpaceae) in a Malaysian tropical rain forest. Tree Physiol., 37(10), 1301–1311, doi: 10.1093/treephys/tpx053, 2017. 22 Jackson, R. B., Canadell, J., Ehleringer, J. R., Mooney, H. A., Sala, O. E. and Schulze, E. D.: A global analysis of root distributions for terrestrial biomes, Oecologia, 108(3), 389–411, doi:10.1007/BF00333714, 1996. Jung, M., Reichstein, M., Margolis, H. A., Cescatti, A., Richardson, A. D., Arain, M. A., Arneth, A., Bernhofer, C., Bonal, D., Chen, J., Gianelle, D., Gobron, N., Kiely, G., Kutsch, W., Lasslop, G., Law, B. E., Lindroth, A., Merbold, L., Montagnani, L., Moors, E. J., Papale, D., Sottocornola, M., Vaccari, F. and Williams, C.: Global patterns of land- atmosphere fluxes of carbon dioxide, latent heat, and sensible heat derived from eddy covariance, satellite, and meteorological observations, J. Geophys. Res. Biogeosciences, 116, G00J07, doi:10.1029/2010JG001566, 2011. Kiew, F., Hirata, R., Hirano, T., Wong, G.X., Aeries, E.B., Musin, K.K., Waili, J.W., Lo, K.S., Shimizu, M., Melling, L.: CO2 balance of a secondary tropical peat swamp forest in Sarawak, Malaysia. Agric. For. Meteorol. 248, 494–501, doi:https://doi.org/10.1016/j.agrformet.2017.10.022, 2018. Kim, D.-H., Sexton, J. O. and Townshend, J. R.: Accelerated deforestation in the humid tropics from the 1990s to the 2000s, Geophys. Res. Lett., 42(9), 3495–3501, doi:10.1002/2014GL062777, 2015. Klein, T.: The variability of stomatal sensitivity to leaf water potential across tree species indicates a continuum between isohydric and anisohydric behaviours, Funct. Ecol., 28(6), 23 1313–1320, doi:10.1111/1365-2435.12289, 2014. Konings, A. G. and Gentine, P.: Global variations in ecosystem‐scale isohydricity, Glob. Chang. Biol., 23(2), 891-905, doi:10.1111/gcb.13389, 2016. Kosugi, Y., Takanashi, S., Ohkubo, S., Matsuo, N., Tani, M., Mitani, T., Tsutsumi, D. and Nik, A. R.: CO2 exchange of a tropical rainforest at Pasoh in Peninsular Malaysia, Agric. For. Meteorol., 148(3), 439–452, 2008. Kosugi, Y., Takanashi, S., Tani, M., Ohkubo, S., Matsuo, N., Itoh, M., Noguchi, S. and Nik, A. R.: Effect of inter-annual climate variability on evapotranspiration and canopy CO2 exchange of a tropical rainforest in Peninsular Malaysia, J. For. Res., 17(3), 227–240, doi:10.1007/s10310-010-0235-4, 2012. Kumagai, T. and Porporato, A.: Strategies of a Bornean tropical rainforest water use as a function of rainfall regime: isohydric or anisohydric?, Plant. Cell Environ., 35(1), 61–71, doi:10.1111/j.1365-3040.2011.02428.x, 2012. Kumagai, T., Saitoh, T. M., Sato, Y., Takahashi, H., Manfroi, O. J., Morooka, T., Kuraji, K., Suzuki, M., Yasunari, T. and Komatsu, H.: Annual water balance and seasonality of evapotranspiration in a Bornean tropical rainforest, Agric. For. Meteorol., 128(1–2), 81–92, doi:10.1016/j.agrformet.2004.08.006, 2005. Lasslop, G., Reichstein, M., Papale, D., Richardson, A. D., Arneth, A., Barr, A., Stoy, P. C., Wohlfahrt, G.: Separation of net ecosystem exchange into assimilation and respiration 24 using a light response curve approach: critical issues and global evaluation, Glob. Change Biol., 16: 187-208, 2010. Leuning, R., van Gorsel, E., Massman, W. J. and Isaac, P. R.: Reflections on the surface energy imbalance problem, Agric. For. Meteorol., 156, 65–74, doi:10.1016/j.agrformet.2011.12.002, 2012. Malhi, Y., Pegorano, E., Nobre, A. D., Pereira, M. G. P., Grace, J., Culf, A. D. and Clement, R.: Energy and water dynamics of a central Amazonian rain forest, J. Geophys. Res., 107(D20), 8061, doi:10.1029/2001JD000623, 2002. Martens, C. S., Shay, T. J., Mendlovitz, H. P., Matross, D. M., Saleska, S. R., Wofsy, S. C., Stephen Woodward, W., Menton, M. C., De Moura, J. and Crill, P. M.: Radon fluxes in tropical forest ecosystems of Brazilian Amazonia: night‐time CO2 net ecosystem exchange derived from radon and eddy covariance methods, Glob. Chang. Biol., 10(5), 618–629, 2004. Massman, W. J.: A simple method for estimating frequency response corrections for eddy covariance systems, Agric. For. Meteorol., 104(3), 185–198, 2000. Medvigy, D., Walko, R. L. and Avissar, R.: Effects of deforestation on spatiotemporal distributions of precipitation in South America, J. Clim., 24(8), 2147–2163, 2011. Monteith, J. L., & Unsworth, M. (1990). Principles of environmental physics. London: Arnold. 25 Motulsky, H. J. and Ransnas, L. A.: Fitting curves nonlinear regression : a practical and nonmathematical review, FASEB J., 1(5), 365–374, 1987. Mrad, A., Sevanto, S., Domec, J., Liu, Y., Nakad, M. and Katul G. G.: A dynamic optimality principle for water use strategies explains isohydric to anisohydric plant responses to drought. OSF Preprints, doi: 10.31219/osf.io/dsxm8. Novick, K. A., Ficklin, D. L., Stoy, P. C., Williams, C. A., Bohrer, G., Oishi, A. C., Papuga, S. A., Blanken, P. D., Noormets, A., Sulman, B. N., Scott, R. L., Wang, L. and Phillips, R. P.: The increasing importance of atmospheric demand for ecosystem water and carbon fluxes, Nat. Clim. Chang., 6, 1023–1027, doi: 10.1038/nclimate3114, 2016. Oishi, A. C., Oren, R., Novick, K. A., Palmroth, S. and Katul, G. G.: Interannual invariability of forest evapotranspiration and its consequence to water flow downstream, Ecosystems, 13(3), 421–436, doi:10.1007/s10021-010-9328-3, 2010. Papale, D., Reichstein, M., Aubinet, M., Canfora, E., Bernhofer, C., Kutsch, W., Longdoz, B., Rambal, S., Valentini, R., Vesala, T. and Yakir, D.: Towards a standardized processing of Net Ecosystem Exchange measured with eddy covariance technique: algorithms and uncertainty estimation, Biogeosciences, 3, 571–583, 2006. Pastorello, G., Papale, D., Chu, H., Trotta, C., Agarwal, D., Canfora, E., Baldocchi, D. and Torn, M.: A new data set to keep a sharper eye on land-air exchanges, Eos (Washington. 26 DC)., 2017. Reichstein, M., Falge, E., Baldocchi, D., Papale, D., Aubinet, M., Berbigier, P., Bernhofer, C., Buchmann, N., Gilmanov, T., Granier, A., Grunwald, T., Havrankova, K., Ilvesniemi, H., Janous, D., Knohl, A., Laurila, T., Lohila, A., Loustau, D., Matteucci, G., Meyers, T., Miglietta, F., Ourcival, J.-M., Pumpanen, J., Rambal, S., Rotenberg, E., Sanz, M., Tenhunen, J., Seufert, G., Vaccari, F., Vesala, T., Yakir, D. and Valentini, R.: On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm, Glob. Chang. Biol., 11(9), 1424–1439, doi:10.1111/j.1365- 2486.2005.001002.x, 2005. Richardson, A. D., Hollinger, D. Y., Burba, G. G., Davis, K. J., Flanagan, L. B., Katul, G. G., Munger, J. W., Ricciuto, D. M., Stoy, P. C., Suyker, A. E., Verma, S. B. and Wofsy, S. C.: A multi-site analysis of random error in tower-based measurements of carbon and energy fluxes, Agric. For. Meteorol., 136(1–2), 1–18, 2006. da Rocha, H. R., Goulden, M. L., Miller, S. D., Menton, M. C., Pinto, L. D. V. O., de Freitas, H. C. and de Silva Figueira, A. M.: Seasonality of water and heat fluxes over a tropical forest in eastern Amazonia, Ecol. Appl., 14, 22–32, doi:10.1890/02-6001, 2004. Stoy, P., Mauder, M., Foken, T., Marcolla, B., Boegh, E., Ibrom, A., Arain, M. A., Arneth, A., Aurela, M., Bernhofer, C., Cescatti, A., Dellwik, E., Duce, P., Gianelle, D., van Gorsel, E., Kiely, G., Knohl, A., Margolis, H., McCaughey, H., Merbold, L., Montagnani, L., 27 Papale, D., Reichstein, M., Saunders, M., Serrano-Ortiz, P., Sottocornola, M., Spano, D., Vaccari, F. and Varlagin, A.: A data-driven analysis of energy balance closure across FLUXNET research sites: The role of landscape scale heterogeneity, Agric. For. Meteorol., 171, 137–152, 2013. Stoy P. C., El-Madany, T., Fisher, J. B., Gentine, P., Gerken, T., Good, S. P., Liu, S., Miralles, D. G., Perez-Priego, O., Skaggs, T. H., Wohlfahrt, G., Anderson, R. G., Jung, M., Maes, W. H., Mammarella, I., Mauder, M., Migliavacca, M., Nelson, J., A., Poyatos, R., Reichstein, M., Scott, R. L. and Wolf. S.: Reviews and synthesis: Turning the challenges of partitioning ecosystem evaporation and transpiratoin into opportunities. Biogeosciences Discussions, doi: 10.5194/bg-2019-85, 2019. Sulman, B. N., Roman, D. T., Yi, K., Wang, L., Phillips, R. P. and Novick, K. A.: High atmospheric demand for water can limit forest carbon uptake and transpiration as severely as dry soil, Geophys. Res. Lett., 43, 9686–9695, doi:10.1002/2016GL069416, 2016. Tang, A. C. I., Stoy, P. C., Hirata, R., Musin, K. K., Aeries, E. B., Wenceslaus, J. and Melling, L.: Eddy covariance measurements of methane flux at a tropical peat forest in Sarawak, Malaysian Borneo. Geophysical Research Letters. doi: 10.1029/2017GL076457, 2018. Vick, E. S. K., Stoy, P. C., Tang, A. C. I. and Gerken, T.: The surface-atmosphere exchange of carbon dioxide, water, and sensible heat across a dryland wheat-fallow rotation, Agric. 28 Ecosyst. Environ., 232, 129–140, doi:doi:10.1016/j.agee.2016.07.018, 2016. Webb, E. K., Pearman, G. I. and Leuning, R.: Correction of flux measurements for density effects due to heat and water vapour transfer, Q. J. R. Meteorol. Soc., 106, 85–100, 1980. Werth, D.: The local and global effects of Amazon deforestation, J. Geophys. Res., 107(D20), doi:10.1029/2001JD000717, 2002. Wilczak, J. M., Oncley, S. P. and Stage, S. A.: Sonic anemometer tilt correction algorithms, Boundary-Layer Meteorol., 99(1), 127–150, 2001. Wilson, K., Goldstein, A., Falge, E., Aubinet, M., Baldocchi, D., Berbigier, P., Bernhofer, C., Ceulemans, R., Dolman, H., Field, C., Grelle, A., Ibrom, A., Law, B. E., Kowalski, A., Meyers, T., Moncrieff, J., Monson, R., Oechel, W., Tenhunen, J., Valentini, R. and Verma, S.: Energy balance closure at FLUXNET sites, Agric. For. Meteorol., 113, 223–243, 2002. Wolf, S., Eugster, W., Majorek, S. and Buchmann, N.: Afforestation of tropical pasture only marginally affects ecosystem-scale evapotranspiration, Ecosystems, 14(8), 1264–1275, 2011a. Wolf, S., Eugster, W., Potvin, C., Turner, B. L. and Buchmann, N.: Carbon sequestration potential of tropical pasture compared with afforestation in Panama, Glob. Chang. Biol., 17(9), 2763–2780, 2011b. Wolf, S., Eugster, W., Potvin, C. and Buchmann, N.: Strong seasonal variations in net ecosystem CO2 exchange of a tropical pasture and afforestation in Panama, Agric. For. 29 Meteorol., 151(8), 1139–1151, 2011c. Wu, J., Guan, K., Hayek, M., Restrepo-Coupe, N., Wiedemann, K.T., Xu, X., Wehr, R., Christoffersen, B.O., Miao, G., da Silva, R., de Araujo, A.C., Oliviera, R.C., Camargo, P.B., Monson, R.K., Huete, A.R., Saleska, S.R.: Partitioning controls on Amazon forest photosynthesis between environmental and biotic factors at hourly to interannual timescales. Glob. Chang. Biol., 23, 1240–1257, doi:10.1111/gcb.13509, 2017. 30 Tables Table 1. Tropical rainforest eddy covariance sites in the FLUXNET2015 database including the MY-MLM site explored in the present study Site ID Name Latitude Longitude Observation period (years of measurement) References AU-Rob Robson Creek, Australia -17.1175 145.6301 2014-2014 (1) Bradford et al. (2014) BR-Sa1 Santarem-Km67- Primary Forest, Brazil -2.8567 -54.9589 2002-2011 (9) Martens et al. (2004) BR-Sa3 Santarem-Km83- Logged Forest, Brazil -3.0180 -54.9714 2000-2004 (5) Goulden et al. (2006) GF-Guy Guyaflux, French Guyana 5.2777 -52.9288 2004-2014 (11) Bonal et al. (2008) GH-Ank Ankasa, Ghana 5.2685 -2.6942 2011-2014 (4) Albinet et al. (2015) MY-PSO Pasoh Forest Reserve, Malaysia 2.9730 102.3062 2003-2009 (7) Kosugi et al. (2008b) PA-SPn Sardinilla Plantation, Panama 9.3181 -79.6346 2007-2009 (3) Wolf et al., (2011a, 2011b, 2011c) MY-MLM Maludam National Park, Malaysia 1.4536 111.1494 2011-2014 (4) Tang et al., (2018) Table 2. The amount of variability (r2) of the Bowen ratio (Bo) explained by volumetric water content (VWC) and vapor pressure deficit (VPD) for the tropical forest ecosystems described in Table 1 and the study ecosystem, MY-MLM. VWC data were not available for BR-Sa1. Site ID VWC VPD AU-Rob 0.31 0.19 BR-Sa1 - < 0.01 BR-Sa3 0.05 0.01 GF-Guy 0.04 0.07 GH-Ank < 0.01 < 0.01 MY-PSO 0.03 < 0.01 PA-SPn 0.23 0.20 MY-MLM < 0.01 < 0.01 31 Figures Figure 1: Map of the Maludam National Park, Malaysia tropical peat forest research site (MY- MLM) and tropical rainforest FLUXNET tower sites used in this study (see Table 1). 32 Figure 2: The seasonal pattern of net radiation (Rn), latent heat flux (LE), and sensible heat flux (H) at the MY-MLM tropical peat forest ecosystem in Sarawak, Malaysian Borneo. Dots represent mean daily values for the four-year measurement period and lines represent a Savitzky-Golay filter applied to the average flux sum for each day of the calendar year. 33 Figure 3: The cumulative sum of evapotranspiration (ET) in a tropical peat forest ecosystem in Sarawak, Malaysian Borneo (MY-MLM) with uncertainty estimates displayed as +/- 1 standard deviation from the annual sum. Shaded areas represent the dry season. Annual ET sums are not significantly different from each other. ;J" " `6&0:'"JZ">')%"A60:%)("'=)E3:)26='"M:)7263%"[23E]")%A"C3B'%":)263%"[X3223H]"M3:"28'"A:4" )%A"" B'2"?')?3%"6%")"2:3E67)("E')2"M3:'?2"'73?4?2'H"6%"1):)B)G.">)()4?6)%"C3:%'3"[>fT >K>]+" 35 Figure 5: The response of the Bowen ratio to net radiation (Rn) across the dry and wet seasons in a tropical peat rainforest ecosystem in Sarawak, Malaysian Borneo (MY- MLM). Vertical bars represent the standard error for each bin. 36 Figure 6: The average diurnal pattern of the shortwave albedo for different measurement years in a tropical peat rainforest ecosystem in Sarawak, Malaysian Borneo (MY-MLM). 37 Figure 7: Mean half-hourly LE against mean half-hourly VPD for dry and wet seasons for 2011, 2012, 2013 and 2014, respectively. Vertical bars represent the standard error for each bin. 38 Figure 8: Daily averaged Bowen ratio against daily averaged VPD for dry and wet seasons for 2011, 2012, 2013 and 2014, respectively. Vertical bars represent the standard error for each bin. 39 Figure 9: Energy balance closure, calculated as the daily sum of sensible (H) and latent heat flux (LE) versus the daily sum of net radiation (Rn) minus the model for soil heat flux (G) across the dry and wet seasons for a four-year measurement period in a tropical peat rainforest ecosystem in Sarawak, Malaysian Borneo (MY-MLM). 40 Figure 10: The seasonal patterns of (a) latent heat flux (LE), (b) sensible heat flux (H), and (c) the Bowen ratio, across the eight eddy covariance research sites investigated here (see Table 1). Lines represent a Savitzky-Golay filter applied to average daily fluxes and ratios across all measurement years for each ecosystem. Values for MY-MLM follow Figure 2. 41 Figure 11: The relationship between vapor pressure deficit (VPD) and the residual of a linear model between net radiation and latent heat flux (LE residual). Results are presented for VPD values greater than 1 kPa for which canopy conductance is known to be sensitive to VPD across global ecosystems (Lasslop et al., 2010). All relationships were significant at the common P < 0.05 level.