Abstract

Despite the pronounced influence of forest phenology on scalar fluxes, no study to date has analyzed these processes in detail. Here, we compared the energy, water, and carbon dioxide fluxes within a broadleaved deciduous forest in Toyota, and an evergreen-dominated mixed forest in Seto, both located in the temperate region of Japan. This study includes annual flux data, obtained from two micrometeorological towers located above the forest canopy at each site, estimated based on the eddy covariance technique. The sensible heat flux (H) measured largely superseded the latent heat flux (lE) from December to April, when the Toyota deciduous forest was leafless. On the other hand, lE flux greatly exceeded the values measured for any of the other parameters, even for net radiation, in evergreen-dominated forest from October to January. Annual mean lE was significantly higher (P <0.05) in the Seto forest than in the deciduous forest in Toyota. The estimated annual gross primary production, ecosystem respiration, and net ecosystem production in the deciduous forest were 13.97, 8.54, and 5.42 t C ha⁻¹ y⁻¹, respectively. For the evergreen mixed forest, these values were 15.20, 12.28, and 2.91 t C ha⁻¹ y⁻¹, respectively. A greater amount of carbon (C) was released into the atmosphere in Seto forest in low growing winter period, compared to the deciduous Toyota forest; therefore the need to maintain a higher respiration rate throughout winter to support the Seto’s evergreen leaves resulted in less C sequestration. Thus, deciduous vegetation may facilitate carbon sequestration more efficiently than evergreen vegetation in temperate climates.

Keywords carbon sequestration, deciduous, evergreen, normalized respiration, phenology, Quercus serrata, temperate climate.

Author and Article Information

Author info
1. Laboratory of Plant Ecology, Department of Crop Botany, Faculty of Agriculture, Bangladesh Agricultural University. Bangladesh.
2. Laboratory of Forest Meteorology and Hydrology, Graduate School of Bioagricultural Sciences, Nagoya University, Japan.

RecievedApr 16 2014　　AcceptedJun 20 2014　　PublishedAug 6 2014

CitationAwal MA, Ohta T (2014) Energy, water, and carbon dioxide fluxes in broadleaved deciduous and evergreen mixed forests in temperate Japan. Science Postprint 1(1): e00027. doi: 10.14340/spp.2014.08A0001

Copyright©2014 The Authors. Science Postprint is published by General Healthcare Inc. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 2.1 Japan (CC BY-NC-ND 2.1 JP) License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

FundingThe first author was supported by the Japan Society for the Promotion of Science (JSPS) as a Postdoctoral Research Fellow at Nagoya University, Japan. Part of this study was supported by Grants-in-Aid for Scientific Research from the JSPS (Grant Nos. 11213209 and 14206018).

Competing interestThere are no relevant competing interests to disclose.

Ethics StatementThis work does not form part of any study or trial; it is part of our regular work.

Donation messageYour kind support would be highly appreciated to further advance this research.

Corresponding authorMd. Abdul Awal
AddressLaboratory of Plant Ecology, Department of Crop Botany, Faculty of Agriculture, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh.
E-mailawalma7@yahoo.com

Introduction

Temperate biomes extend across the mid-latitudes (25–50^°) of both hemispheres, and they are characterized by large seasonal and diurnal climatic variations. The geometry of the Earth’s orbit, together with changes in its location in relation to the sun throughout the year, determines the intensity of seasonal fluctuations in radiation and temperature at different latitudes. Thus with an increase in latitude within the temperate zone, seasonal fluctuations intensify, and summer and winter represent the growing and non-growing seasons, with photosynthesis tending to thrive during summer. Temperate forests often include both deciduous and evergreen tree species as mixed stands; however, most forests are dominated by deciduous species 1-3. Natural and managed monotypic forests are also common in this region. This zone accounts for approximately 25% of the world’s vegetation, and thus it is particularly important for global cycles of carbon, energy, and water. Overall, temperate forests are expanding in area, while forests from other regions are rapidly decreasing in size, particularly in the tropical regions 4. In addition, a large proportion of the world human population lives within the temperate region, and therefore temperate forests are directly influenced by human interventions, such as increases in CO₂, fertilization, harvesting, reforestation, nitrogen deposition, and urbanization. Set against this background, it is important to assess the role of temperate forests and forest management in carbon sequestration and energy and water fluxes.

Broadly, there are two types of functional vegetation worldwide, deciduous and evergreen. Deciduous species are characterized by marked phenological events, including leaf emergence and senescence 5, which can vary in timing and duration between years. In contrast, in evergreen species, leaf emergence and senescence occur continuously throughout the year, and no other major vegetative changes are observed between seasons. Thus, gas exchange in deciduous forests predominantly occurs during the fraction of the year when leaves are still expanding, before they reach their maximum leaf area. On the other hand, such processes occur continuously throughout the year in evergreen species. In general, gas exchange occurs normally in all vegetation types during summer, as long as the levels of radiation and temperature remain high. In winter, atmospheric limitations may lead to a lower photosynthetic rate, although respiration continues to sustain life, specifically marked by the presence green foliage in evergreen species. Although both deciduous and evergreen forests are primary forests in temperate regions, the seasonal variation in gas exchange differs between them 6. In other words, deciduous and evergreen forests complement each other in terms of energy and gas exchange throughout the year. Therefore, the forest phenological type, deciduous or evergreen, plays an important role in the global energy balance and water and carbon cycles.

Forests act both as a source and sink of atmospheric trace gases and water vapor and thus they actively influence the global carbon and energy balance, and atmospheric chemistry. As one of the major stocks of mobilizable carbon, forests can influence the planetary carbon balance and global and local climate by altering the rate at which they take up and lose carbon, water vapor, and energy 7. Some studies have estimated carbon, water, and energy fluxes in a wide range of climates (e.g., tropical, temperate, polar, or alpine) and vegetation types (e.g., rangeland, woodland, or agriculture land) 8-13. Several studies have examined gas flux between forests and the atmosphere in relation to vegetation type 14-19; however, to date, no study has compared the effect of evergreen and deciduous phenology within the same region and climate.

Here, we compared the energy, water, and CO₂ fluxes between deciduous and evergreen-dominated mixed forests, both located in temperate Japan. This study will help to determine which type of vegetation is more effective in reducing atmospheric CO₂ and increasing ecosystem production in temperate climates.

Materials and Methods

Site

Measurements were conducted in two secondary broadleaved forests. A predominantly deciduous forest in Toyota (35^°02^′N, 137^°11^′E) and a mixes deciduous/evergreen forest in Seto (35^°15^′N, 137^°05^′E), both within Aichi Prefecture, Central Japan (Figure 1). These forests are separated by 25 km and are approximately 70–80 years old. They have grown naturally, with no human intervention or active management since the end of World War II in 1945. The Japanese red pine (Pinus densiflora), an evergreen coniferous species, was the pioneer and dominant species in both forests; however, this species has gradually declined in recent decades, mostly due to the incidence of different pine diseases, including needle blight (caused by Dothistroma sp.), brown spot (Sricca sp.), needle cast (Lophodermium sp.), tip blight (Sphaeropsis sp.), and wilt (caused by the wood nematode Bursaphelenchus xylophilus). Following outbreaks of these diseases, both forests had been left to naturally regenerate, and ever since the red pine has gradually been replaced by a deciduous broadleaf oak (Quercus serrata), followed by colonization of the evergreen oak Quercus glauca. Currently, several evergreen species, including Ilex pedunculosa (longstalk holly), Eurya japonica, and Cleyera japonica are also mixed and coexist with different deciduous species other than oaks in the Seto forest. Thus, the Seto forest is configured as a mixed forest, and includes distinct proportions of evergreen and deciduous species. In this forest, the ratio of basal area (measured as diameter at breast height, at 1.2 m height from the ground level) of evergreen to deciduous stands is approximately 1.5 (i.e., evergreen species occupy approximately 1.5 more area than deciduous stands). Thus, hereafter, the Seto forest is referred to as an evergreen-dominated mixed forest.

Following natural succession, the deciduous species Quercus serrata remained the dominant species in the Toyota forest for approximately 35–40 years. This forest is considered predominantly deciduous (and referred to as such hereafter), as the basal area ratio of evergreen/deciduous stands in the Toyota forest is very small (0.2) 20. Table 1 includes the site characteristics, canopy composition, and other features such as leaf area index (LAI), and basal area for each site.

Leaf emergence in deciduous species began in the second week of April in both forests; they were both in full leaf by the end of May and showed maximum leaf area index value by mid-July. Deciduous leaves began to change color around the third to the fourth week of October. Leaf senescence commenced in the first week of December, and deciduous stands were entirely defoliated within that month.

Instrumentation and data acquisition

Turbulent flux of energy, water, and CO₂ data were recorded using an open-path CO₂/H₂O infrared gas analyzer IRGA (LI-7500; LICOR, Inc., Lincoln, Nebraska, USA). Three-dimensional (3D) wind speeds were recorded using a triaxial ultrasonic anemometer (model DAT-310; Kaijo for Toyota, and DAT-540; Kaijo, Tokyo, Japan for Seto). One IRGA and one anemometer were installed 20 m above the ground on a walk-up scaffold micrometeorological tower at each site. Full details of the other instrumental set-up are reported in Matsumoto et al. (2008) 21 for the Seto site and in Awal et al. (2010) 22 for the Toyota site. In brief, downward short-wave solar radiation was measured using a Pyranometer (a MS-42; Eiko, Seiki, Tokyo, Japan was installed in the Toyota forest tower, and a CNR-1; Kipp and Zonen, Delft, the Netherlands was installed in the Seto forest tower). Dry- and wet-bulb temperatures were measured using an aspirated psychrometer installed in the Toyota tower (MH-020T; Eiko, Seiki). Air temperature and relative humidity in Seto site were measured using a combined temperature and humidity probe (HMP-350; Vaisala, Vantaa, Finland).

Data were recorded using a 10 Hz sampling rate for Seto and a 20 Hz rate for Toyota. All the data were transmitted using sequential digital connections, to prevent lags or time delays. All the signals were processed using a data logger (CR23X; Campbell Scientific, Logan, UT, USA), and the data were stored on the hard disk of a portable computer, located during the field campaign in a hut at the base of each tower. The averaging interval for computing turbulent fluxes was 30 min. Because the eddy flux data were collected by open-path IRGAs, these data were corrected for the effect of density fluctuations 23. A positive flux density was generated when CO₂ was transferred into the atmosphere from the forest (i.e., carbon source), whereas a negative value denoted the reverse situation (i.e., carbon sink). The data were recorded from May 2003 to April 2004 in both sites. A high energy balance closure (i.e., the ratio between the sum of latent and sensible heat fluxes and the available heat, or the fraction of turbulent heat flux in the total available energy) was found for the whole period for both sites 20-21.

Data processing and analyses

Due to unexpected circumstances such as system failures, and the time required for data acquisition, the eddy flux data obtained included missing data. All data recorded during periods of rainfall and within 10 hours of a rainfall event were rejected because of known IRGAs malfunctions related to these circumstances 24. The CO₂ flux (F_c) data more than 3 standard deviations away from the estimated 10-day moving average were regarded as outliers and rejected. Finally, F_c with low turbulence (friction velocity, u_* ≤0.20 ms⁻¹) at night were also rejected, as F_c data below 0.20 ms⁻¹ were found to be mainly drained or underestimated. Consequently, F_c data were not available for all time series. During the daytime, missing F_c values (or NEE, net ecosystem exchange) were estimated using the empirical model described by Hollinger et al. (1994) 25 at the whole-canopy scale based on the Michaelis−Menten kinetics, as follows:

where R_d represents dark respiration, P_max is the potential maximum rate of photosynthesis, K_m is the Michaelis-Menten constant, and Q is the amount of photosynthetically active radiation (PAR).

The light compensation point, Q_comp (the PAR at which photosynthetic and respiration rates become equal, i.e., NEE = 0), can be derived from Eq (1) as follows:

In addition, light-use efficiency or quantum yield, α (slope of the light response curve at zero PAR) can be estimated as:

Table 2 shows the best fitted values for the parameters R_d, P_max, K_m, Q_comp, and α.

In order to estimate the values of the half-hourly data missing at night-time, the following linear relationship, between air temperature (T_a) and NEE from a high turbulence period, i.e., outside the threshold limit (i.e., u_* >0.20 ms⁻¹), was used:

where b₁ (intercept) and b₂ (slope) are the fitted parameters (Table 3). Equations 1 and 4 were used for filling the gaps both in annual and seasonal basis, where seasons divided into spring (March-April-May, MAM), summer (June-July-August, JJA), fall (September-October-November, SON), and winter (December-January-February, DJF).

At night, NEE equals RE (ecosystem respiration) and the Eq. (4) was extrapolated to estimate daytime RE. The half-hourly values of gross ecosystem exchange (GEE) was calculated as the difference between NEE and RE (i.e., GEE = NEE − RE) during daytime, and assumed to be GEE = 0 at night. Net ecosystem production (NEP) and gross primary production (GPP) were estimated as negative NEE and negative GEE values, respectively.

Stomatal conductance (g_s) was calculated using an inversion of the Penman-Monteith’s big leaf model (i.e., P-M canopy conductance model).

Results

Micrometeorological parameters

The annual data obtained for the main micrometeorological parameters, including the incidence of photosynthetically active radiation (PAR), vapor pressure deficit (VPD), air temperature (T_a), and soil temperature (T_s) showed a distinct seasonal trend. The highest daily values where recorded in summer, while the lowest values were observed in winter, thus presenting the typical characteristics of solar temperate climates from the Northern Hemisphere (Figure 2a–h). Overall, the microclimatic data was similar in the Seto and the Toyota sites. Most rainfall occurred during the summer season, from June to August, while the winter season, from December to February, was characterized by the lowest level of rainfall (Figure 3a). The volumetric soil water content (SWC) was higher during summer, and showed strong similitude between sites in relation to magnitude and trend (Figure 3b). However, higher SWC values were recorded in the Toyota site from December to April than in the Seto forest, possibly because the deciduous stands in the Toyota forest utilized less soil water during the leafless winter period.

Energy partitioning and balance components

Figure 4 shows the components of the diurnal energy balance during two contrasting seasons—summer and winter. In both sites, all energy balance parameters showed distinct diurnal trends, reaching maximum levels of net radiation (R_n), latent energy (lE), and ground heat flux (G) during summer. On the other hand, these fluxes reached their minimum values during the winter season, while the sensible heat flux (H) increased in winter and decreased in summer. During both summer and winter seasons, the Seto site exhibited higher H flux (upward during the daytime and downward during the night-time) than the Toyota site. During winter, the Seto site was characterized by a distinctly higher upward lE flux, particularly during the daytime, than the Toyota site. The parameters associated with the diurnal energy balance showed intermediate values during spring and fall, compared to summer and winter seasons (data not shown).

The daily mean energy partitioning components also showed pronounced seasonal changes in both forests, associated with spring leafing and autumn leaf fall (Figure 5). The sensible heat flux (H) remained significantly low from May to September, when leaf area gradually increases; however, the latent heat (lE) flux remained high and continued to dominate over the H flux until the completion of leaf senescence, between November and December. During the active growing season, the upward H flux was higher in the Seto than in the Toyota site. The H flux greatly exceeded the latent heat from December to April in the Toyota forest, when the deciduous trees were leafless. The upward lE flux (mean difference 17.6 W m⁻²; P <0.01) in low-growing season was remarkable higher at the Seto site than at Toyota (the half-hourly mean throughout the season was 42.7 W m⁻²), highlighting the contribution of evergreen vegetation to the upward lE flux. From October to January, daily mean lE in the Seto forest greatly exceeded the values measured for any of the other parameters, even for net radiation (R_n). The annual average of the upward lE flux calculated based on half-hourly data, was significantly higher in the Seto site (60.5 W m⁻²; mean difference 11.7 W m⁻²; P <0.01) than in the Toyota site (48.8 W m⁻²); however, the differences between the two sites were not significantly different for the other three energy balance components (R_n, G, and H; Figure 5).

Responses of gas exchange to vapor pressure deficit (VPD)

lE increased with VPD in both forests (Figure 6a, b). The trend lines show that lE was distinctly higher in the Seto forest, compared to the Toyota site, for the same VPD during winter. However, lE did not differ significantly between the two forests during summer.

In both forests, g_s declined with increasing VPD (Figure 6c, d). During winter, the evergreen mixed forest in Seto showed higher g_s values, for the same level of VPD, compared to the deciduous Toyota forest; however, no remarkable differences were found between the two sites in the summer season.

The net assimilatory CO₂ flux (i.e., negative NEE) increased with VPD, up to a certain limit, and decreased slightly thereafter (Figure 6e, f). In winter, the net assimilation of CO₂, for a similar VPD level, was significantly higher in the Seto forest than in the Toyota forest; however, such differences were not found for the summer season.

CO₂ flux responses to radiation and temperature

Figure 7 shows the response of daytime NEE (i.e., CO₂ flux) to the level of PAR registered during summer (JJA) and winter (DJF) seasons. In addition, the characteristic responses to sun radiation, for all months and seasons are given in Table 2. During summer, the net assimilation of CO₂, for the same radiation level, was slightly higher in the Toyota site. On the other hand, during winter, the net uptake of CO₂ per unit of PAR was distinctly larger in the Seto site than at the Toyota site. Higher R_d, P_max, and quantum yield or light-use efficiency (α) were recorded during the summer months, including May, and gradually decreased toward the fall (SON). The minimum values for these parameters were reached during the winter months but increased again during the following spring (MAM; Table 2). R_d, P_max, and α varied seasonally with the trend JJA > MAM > SON > DJF in both sites. Compared to the Toyota deciduous forest, the evergreen mixed forest in Seto showed comparatively higher R_d and P_max values throughout the year. In contrast, the Toyota deciduous forest showed comparatively greater α than the Seto forest during from May to August; however, the trend was inversed, with lower values in the Toyota forest compared to the Seto site, during the rest of the year.

The Michaelis-Menten constant (K_m) and the light compensation point (Q_comp) did not maintain any specific pattern across months or seasons; however, the evergreen Seto forest showed a distinctly higher Q_comp than the deciduous Toyota forest (Table 2). CO₂ flux can switch from respiratory mode and assimilatory mode, depending on the month or the season, (i.e., sink mode for negative CO₂ flux values) when PAR > 107–201 μmol photon m⁻² s⁻¹ in the Seto site. On the other hand, the switch between modes only required a PAR > 11–131 μmol photon m⁻² s⁻¹ for the Toyota site.

Night-time CO₂ efflux (i.e., net respiratory flux or efflux) varied with air temperature (T_a) in both forests (Figure 8). No obvious differences in the function of night-time CO₂ flux over T_a were found between sites for the summer period (data not shown). However, the CO₂ efflux, for same T_a, was distinctly higher in the Seto site than in the Toyota site during winter (see Figure 9 for details).

Daily, seasonal, and annual CO₂ fluxes

During the summer season, a net uptake of CO₂ (i.e., carbon sink) was distinctly observed during the daytime and a net release of CO₂ (i.e., source) occurred during the night-time (Figure 9a). Compared to the Seto site, the Toyota forest had a significantly lower CO₂ flux (i.e., lager carbon sink) during the daytime (half-hourly seasonal mean −9.14 μmol CO₂ m⁻² s⁻¹, where the negative sign indicates a net uptake of CO₂ by the forest ecosystem from the atmosphere; mean difference 1.61 μmol CO₂ m⁻² s⁻¹; P <0.01). The difference in respiratory CO₂ efflux during the night-time was not significantly different between sites (P = 0.05); however, both forests behaved as net CO₂ sinks during the daily balance, where the Toyota forest (−7.50 μmol CO₂ m⁻² s⁻¹) showed a net CO₂ uptake 45% greater (P <0.01) than the Seto site (−4.13 μmol CO₂ m⁻² s⁻¹).

During the winter season, the CO₂ flux at the Seto site exhibited a distinct diurnal trend (i.e., a net uptake of CO₂ during the daytime and a net release of CO₂ during the night) (Figure 9b). However, no distinct diurnal patterns were observed for net CO₂ flux for the Toyota site. Both the net assimilatory flux of CO₂ during the daytime (−0.45 and −1.92 μmol CO₂ m⁻² s⁻¹ for Toyota and Seto, respectively) and the net respiratory CO₂ flux during the night (0.39 and 2.01 μmol CO₂ m⁻² s⁻¹ for Toyota and Seto, respectively) were significantly higher (P <0.01) in the Seto forest than in the Toyota site. On the basis of the overall daily carbon balance, the Seto forest appeared to be a net source of CO₂, while the Toyota forest acted as a slight net sink. The nature and magnitude of the diurnal net CO₂ flux during spring and fall showed intermediate values between those observed during summer and winter (data not shown).

Daily night-time CO₂ efflux reached the lowest values during winter and the highest values in summer. Similar to other parameters, intermediate values between these two were found in spring (from March to May) and autumn (from September to November; Figure 10). The Seto forest showed a significantly higher (P <0.01) night-time respiratory flux, particularly during the low-growing season, from October to April (for example, 2.04 g CO₂ m⁻² day⁻¹ for the Toyota forest and 5.52 g CO₂ m⁻² day⁻¹ for the Seto site). Throughout the year, night-time ecosystem respiration was also significantly higher in the Seto forest than at the Toyota site. The annual average night-time respiratory flux in the Seto site (5.84 g CO₂ m⁻² day⁻¹) was approximately 46% higher (P <0.01) than that of the Toyota site (3.16 g CO₂ m⁻² day⁻¹).

Compared to the Seto forest, the Toyota site (−14.78 g CO₂ m⁻² day⁻¹) maintained a higher net daytime uptake of CO₂ during the active growing season (mean difference 2.06 g CO₂ m⁻² day⁻¹; P <0.05; Figure 10). However, during the low-growing season, the Seto site stored a larger amount of carbon (−5.78 g CO₂ m⁻² day⁻¹) than the Toyota forest (−3.79 g CO₂ m⁻² day⁻¹; P <0.01). No significant differences in the annual average daytime CO₂ influx were found between sites (P = 0.05).

The daily net flux (i.e., sum of daily daytime uptake and night-time release) in the evergreen Seto forest showed that this site behaved as a net source of CO₂ during the winter season (1.31 g CO₂ m⁻² day⁻¹), while the deciduous Toyota forest was a minor sink of CO₂ during winter (−0.88 g CO₂ m⁻² day⁻¹; Figure 11). Both sites behaved as a net sink of CO₂ in the other three seasons, between March and November; however, the Toyota site (−6.67 g CO₂ m⁻² day⁻¹) stored a comparatively larger amount of carbon than the Seto forest (mean difference 2.46 g CO₂ m⁻² day⁻¹; P <0.01). The annual mean net CO₂ flux was approximately 46% higher (P <0.01) in the Toyota forest (−5.23 g CO₂ m⁻² day⁻¹) than in the Seto site (−2.84 g CO₂ m⁻² day⁻¹).

A summary of the seasonal and the annual NEE data, including their partitioning into GPP and RE, and normalized respiration data are shown in Table 4. The Seto site showed a higher gross primary production (GPP) than the Toyota forest for all seasons, except for summer, when the Toyota site showed higher values of gross photosynthesis. Ecosystem respiration (RE) throughout the year was significantly higher in the Seto site; however, the differences between the two sites were not significantly different for the spring and summer seasons. The Toyota site had a distinctly higher net ecosystem production (NEP) during summer and fall than the Seto site. A net loss of carbon (>0.3 t C ha⁻¹) was recorded in the Seto site during winter, even though a slight net gain of carbon was common in Toyota. Compared to the Toyota site, GPP was approximately 9% higher in the Seto site; however, RE was 44% larger, resulting in a significantly lower NEP. Overall, the annual NEP was larger in the Toyota forest (5.42 t C ha⁻¹ y⁻¹) than that at Seto site by 1.86-fold (Table 4).

The normalized respiration (i.e., RE/GPP ratio) was smaller during MAM or JJA, and gradually increased toward the winter reaching maximum values in DJF (Table 4). The normalized respiration was significantly larger in the Seto site than in the Toyota forest, both seasonally and annually. In addition, this difference was most pronounced during DJF, when the RE/GPP ratio in the Seto site was higher than one.

Discussion

Latent heat (i.e., water vapor) flux commonly reaches a maximum during the summer period, when the leaves are present in deciduous forests, or the leaves are present in much larger volumes (in evergreen mixed site). On the other hand, sensible heat flux increases in early spring, when the level of foliage is minimum 26. Soil moisture was depleted faster, and consequently the water vapor flux was larger, in the Seto site than in the Toyota forest during the low-growing season, due to the presence of foliage during winter in the former. A strong seasonal fluctuation of the energy partitioning components, particularly latent heat, was detected in the deciduous Toyota forest, possibly associated with leaf emergence and senescence 27, because transpiration rate decreased to a minimum during the period when the oak (Quercus serrata) and other deciduous species were leafless, typically during winter. In fact, these energy partitioning characteristics, in relation to vegetation types, are often important in regional water cycles, including the recycling of precipitation and soil moisture through evapotranspiration.

The timing of the plant response to light is associated with the leaf area index (LAI). Dark respiration (R_d) is usually lower during the low-growing months, possibly due to the lower temperatures and smaller LAI registered during this period 28. The R_d in light response curve was higher for evergreen forest, probably due to the lower degree of inhibition of dark respiration by light (i.e., the Kok effect) showed by evergreen species compared to deciduous species 29. The maximum rate of photosynthesis (P_max) and quantum yield (α) are clear positive functions of LAI. On the other hand, the low P_max and α values recorded during the low-growing months, particularly winter, are believed to be associated with leaf yellowing, senescence, and defoliation. These parameters showed their higher values during the low-growing season in the evergreen Seto forest and, consequently, this site assimilated a larger amount of CO₂ during the daytime than the Toyota site (Figure 9, 10). Compared to the Seto site (an evergreen-dominated mixed forest), the Toyota site showed a lower light compensation point (Q_comp), and a higher quantum yield or light-use efficiency (α), suggesting the higher photosynthetic capacity of the deciduous stands in the Toyota site, particularly during the high-growing season (when both sites showed their full physiological potential, maximum LAI, higher level of PAR, and higher temperatures). The R_d and P_max values obtained were similar to those previously reported 28, 30-32. Light-use efficiency (α) ranges between 0.04 and 0.055 μmol CO₂/μmol photon for a mixed deciduous Harvard forest in mid-summer 33, between 0.029 and 0.044 μmol CO₂/μmol photon for a Douglas fir- dominated evergreen European coniferous forest 34, between 0.025 and 0.063 μmol CO₂/μmol photon for a Danish beech forest in summer 35, and between 0.04 and 0.07 μmol CO₂/μmol photon for a broadleaf deciduous forest in central Japan 36. In boreal ecosystems, Griffis et al. 16 also found a higher α in a deciduous aspen forest (α ≥ 0.08μmol CO₂/μmol photon) in JJA, while evergreen black spruce (α ≥ 0.06 μmol CO₂/μmol photon) and evergreen jack pine (α ≥ 0.03 μmol CO₂/μmol photon) forests showed comparatively lower α. Our results are consistent with those previously reported.

The CO₂ balance, expressed as the net ecosystem exchange (NEE), represents a fine balance between the uptake of CO₂ (influx) through photosynthesis and the release of CO₂ (efflux) through respiration. Despite the higher annual gross primary production (GPP) observed in the Seto forest, the concomitant larger ecosystem respiration (RE) rate led to a comparatively smaller net annual carbon sequestration (NEE). In contrast, the net annual carbon balance was larger in the deciduous Toyota forest, despite showing a lower amount of GPP, due to a relatively small RE (Table 4). Therefore, respiration influences NEE more strongly than GPP, which is consistent with the findings by Valentini et al. 37. In the temperate sites studied here, photosynthesis can be limited by environmental factors, reaching very low levels for approximately 45% of the year; however, respiration is maintained throughout the year at a considerable rate in evergreen species. Therefore, daytime net CO₂ influx could not offset the release of CO₂ recorded during the night for evergreen-dominated sites, as the night lasts longer than the day during winter in temperate latitudes (Figure 10b), resulting in a loss of CO₂ (source) to the atmosphere. For this reason, even a small increase in the number of evergreen stands can lead to an increase in RE, and subsequently negatively influence NEE. The larger CO₂ efflux observed in the Seto forest than in the Toyota site during winter nights (Figure 9b) clearly indicates that respiration strongly dominates over photosynthesis, due to the greater proportions of evergreen species 15, 18. Green foliage in evergreen stands behaves as a strong source and sink of atmospheric CO₂ throughout the year; however, they become dominantly a carbon source whenever environmental limitations restrict photosynthesis during the low growing winter. This can be supported by the larger normalized respiration recorded in the Seto site compared to the Toyota forest (Table 4).

The net loss of carbon from the forest (i.e., CO₂ efflux) observed at the Seto site, compared to the Toyota forest, during winter was related to an enhanced RE (Table 4), caused by the greater proportion of evergreen species. At the end of the growing season (CO₂ sink period), the CO₂ stored in the evergreen foliage is released, ultimately lowering the amount of carbon sequestration. Indeed, this is exemplified by the fact that the level of C sequestration at the Seto site was approximately 46% lower than the NEP observed in the Toyota forest. During the low-growing season (CO₂ source period), the deciduous stands may be able to avoid carbon loss (through respiration) more efficiently than evergreen species, due to their leafless canopy. Moreover, any additional respiratory loss due to other circumstances (e.g., increasing temperatures) may also be lower in deciduous stands than in evergreen ones. Therefore, annual carbon sequestration in temperate zones is largely influenced by the proportion of evergreen and deciduous species present. This fact might be useful in future studies, to avoid errors when comparing the extent of carbon sequestration between sites.

The differences in carbon sequestration capacity observed between the Seto and the Toyota forests also deserve further attention, to confirm that these differences are not the result of differences between individual stands (Table 1). Canopy height, LAI, and basal area were larger at the deciduous Toyota site than that in the evergreen Seto forest. This variability could be corrected by standardizing the respiration values based on the gross primary production (i.e., the RE/GPP ratio), because net ecosystem carbon exchange is mainly influenced by respiration 37. The greater normalized respiration, particularly during winter, recorded at the Seto site compared to the Toyota forest, can be attributed to the presence of evergreen stands. Higher dark respiration (R_d) values observed in the Seto forest (Table 2), compared to the Toyota site, may strengthen this fact. Thus, evergreen forests in temperate zones probably decrease the capability to act as a sink for atmospheric CO₂ by accelerating respiration. The smaller RE/GPP ratio recorded in the deciduous Toyota site, compared to the evergreen-dominated Seto site, is also supported by the conclusions by Griffis et al. 16. These authors showed that less carbon was invested in respiration by deciduous species (due to seasonal leaf fall) compared to evergreen species, in a study comparing deciduous aspen and evergreen black spruce forests. Therefore, deciduous forests are characterized by a higher net production by burning less carbon through respiration throughout the year.

At the leaf level, deciduous stands also show a greater net photosynthetic activity than evergreen species 39-41. Therefore, forests with a greater proportion of deciduous stands are likely to switch from a CO₂ source to a sink regime earlier in their stand development. From the climatic and adaptive perspective, temperate forests are mainly dominated by deciduous species; however, some forests may also include only evergreen species, or present a mixture of deciduous and evergreen species 1-3. For example, temperate coniferous evergreen and broadleaved deciduous forests cover approximately 40% and 45% of the total forested area in Japan, respectively 42. Increasing the presence of deciduous species in temperate forests might increase carbon sequestration, as a result of smaller winter respiratory loss due to low temperatures, the presence of leafless canopies during winter, and a higher photosynthetic rate during the longer days (as much as 15 hours) characteristic of the high-growing season (i.e., summer). Moreover, the abundance of soil nutrients and water (Figure 3b; because of the reduced uptake, due to leaf fall, during the non-growing season), along with the presence of newly emerged young green foliage, may result in a greater photosynthetic rate and a subsequent faster growth during the growing season 43, thus creating a larger difference between photosynthesis and respiratory fluxes. A small proportional change in these fluxes increase carbon sequestration. In the present study, the deciduous Toyota site showed the effects from this processes, resulting in a higher carbon balance, and thus sequestering carbon from the atmosphere 44-46. The differences in the net carbon assimilation during the high-growing season observed between the two sites are consistent with the results by Ohtani et al. 47, who reported a 1.5 times larger NEP in a deciduous broadleaved forest compared to an evergreen coniferous forest; however, their results regarding the low-growing season differed from those presented in this study. Deciduous forests have also been shown to sequester significantly more carbon than evergreen forests in boreal ecosystems 19, 48. Nevertheless, the annual NEP values observed at both sites in this study lie within the NEP ranges reported for other temperate forests elsewhere 33, 49 and in Japan 50.

Promoting the presence of deciduous vegetation in temperate regions could significantly help to offset the carbon released into the atmosphere associated with anthropogenic emission and forest clearing. However, rapidly replacing evergreen stands with deciduous stands would lead to the conversion of old forests into young ones, further increasing the amount of carbon emission by releasing the carbon stored within old forests 51. Nevertheless, management activities associated with afforestation, reforestation, social forestation, woodlands, roadside forests, agroforests, public recreation forests, and parks should consider a predominance of deciduous trees to offset carbon emissions. Additionally, limiting thinning or understory clearing operations in mixed forest could also be limited to evergreen stands. Selective thinning, encouraging the growth of deciduous stands, may result in higher carbon sequestration, as has been shown previously 52. Young stands require a certain period to switch from carbon sources to sinks; however, once the stands are replaced by deciduous stands, these will become important net carbon sinks. Nevertheless, these strategies should be combined with efficient biodiversity conservation measures, particularly focused on the conservation of endangered species.

Conclusions

These results, based on annual eddy covariance data, showed distinctive seasonal patterns in radiation, temperature, water vapor pressure deficit, and precipitation that broadly coincide with physiological events associated with seasonal gas fluxes, including evapotranspiration, net CO₂ assimilation, etc., in a deciduous forest. On the other hand, the maintenance of a high respiration rate throughout the winter in the evergreen forest studied limited its potential as a carbon sink. The disproportionate differences between deciduous and evergreen vegetation in temperate climates may lead to substantial differences in energy partitioning and carbon sequestration.

Our results showed that the sensible heat flux remained significantly lower during the high growing season, from May to September, which also showed the larger leaf area. During the summer and early fall, the latent heat flux (water vapor) also remained high, being higher than the sensible heat flux until leaf senescence, around November or December. The Seto evergreen-dominated forest was characterized by a higher sensible heat flux during the high-growing season and a higher latent heat flux during the low-growing season, compared to the Toyota forest. The sensible heat flux greatly exceeded that of latent heat from December to April, when the deciduous forest was leafless; however, from October to January, the latent heat flux in the evergreen mixed forest exceeded even the net radiation value. Energy partitioning characteristics in relation to vegetation type are often important in the regional recycling of precipitation and soil moisture through transpiration.

Although the annual gross primary production of the evergreen mixed forest was higher than that of the deciduous forest, a distinctly greater ecosystem respiration resulted in a lower net ecosystem production. The greater amount of ecosystem respiration in the evergreen site, compared to the deciduous site, particularly during the low-growing season may have contributed to the low level of carbon sequestration detected. The significantly higher annual carbon sequestration observed for the Toyota forest indicates that promoting the presence of deciduous species in temperate climates could potentially be used as an efficient method to offset the rising concentration of atmospheric CO₂, and subsequently global warming.

Acknowledgements

The contribution of the personnel involved in the work related to the eddy covariance tower and data acquisition management system is gratefully acknowledged. Thanks to the anonymous reviewers for critical comments and suggestions to improve the final version of the manuscript. Finally, the manuscript was proof-read and polished by the editors at Editage (http://www.editage.com).

Author Contributions

Awal MA: Carried out the main part of the work
Ohta T: Research supervisor

References

1. Axelrod DI (1966) Origin of deciduous and evergreen habits in temperate forests. Evolution 20: pp. 1–15.
2. Woodward FI (1987) Climate and Plant Distribution, Cambridge Studies in Ecology. Cambridge University Press, UK.
3. Baldocchi D, Xu L (2005) Carbon exchange of deciduous broadleaved forests in temperate and Mediterranean regions. In: Griffiths H, Jarvis PG, eds. Carbon Balance of Forest Biomes. pp. 187–215, Taylor & Francis Group, London. ISBN: 978-1859962145.
4. Dixon RK, Solomon AM, Brown S, Houghton RA, Trexler MC, Wisniewski J (1994) Carbon pools and flux of global forest ecosystems. Science 263: pp. 185–190. doi: 10.1126/science.263.5144.185.
5. Hutchison BA, Matt DR, McMillen RT, Gross LJ, Tajchman SJ, Norman JM (1986) The architecture of a deciduous forest canopy in eastern Tennessee, USA. J. Ecol. 74: pp. 635–646.
6. Larcher W (2003) Physiological Plant Ecology: Ecophysiology and Stress Physiology of Functional Groups (4th ed.). Springer-Verlag, Berlin. ISBN: 978-3540435167.
7. Musselman RC, Fox DG (1991) A review of the role of temperate forests in the global CO₂ balance. J. Air Waste Manage. Assoc 41: pp. 798–807. doi: 10.1080/10473289.1991.10466876.
8. Baldocchi D, Vogel CA (1996) Energy and CO₂ flux densities above and below a temperate broad-leaved forest and a boreal pine forest. Tree Physiol. 16: pp. 5–16. doi: 10.1093/treephys/16.1-2.5.
9. Reich PB, Walters MB, Ellsworth DS (1997) From tropics to tundra: Global convergence in plant functioning. PNAS 94 (25): pp. 13730-13734.
10. Malhi Y, Baldocchi DD, Jarvis PG (1999) The carbon balance of tropical, temperate and boreal forests. Plant Cell Environ. 22: pp. 715–740. doi: 10.1046/j.1365-3040.1999.00453.x.
11. Reich PB, Bolstad P (2001) Productivity of evergreen and deciduous temperate forests. In Roy J, Saugier B, Mooney HA, eds. Terrestrial Global Productivity. pp. 245–283, Academic Press, San Diego, CA. ISBN: 978-0125052900.
12. McGuire AD, Wirth C, Apps M, Beringer, J, Clein J, Epstein H, et al. (2003) Environmental variation, vegetation distribution, carbon dynamics and water/energy exchange at high latitudes. J. Veg. Sci. 13: pp. 301–314. doi: 10.1111/j.1654-1103.2002.tb02055.x.
13. Richardson AD, Hollinger DY, Dail DB, Lee JT, Munger JW, O'Keefe J (2009) Influence of spring phenology on seasonal and annual carbon balance in two contrasting New England forests. Tree Physiol. 29: pp. 321–333. doi: 10.1093/treephys/tpn040.
14. Ohta T, Suzuki K, Kodama Y, Kubata J, Kominami Y, Nakai Y (1999) Characteristics of heat balance above the canopies of evergreen and deciduous forests during snowy season. Hydrol. Process. 13: pp. 2383–2394.
15. Arain MA, Black TA, Barr AG, Jarvis PG, Massheder JM, Verseghy DL, Nesic Z (2002) Effects of seasonal and interannual climate variability on net ecosystem productivity of boreal deciduous and conifer forests. Can. J. For. Res. 32: pp. 878–891. doi: 10.1139/x01-228.
16. Griffis TJ, Black TA, Morgenstern K, Barr AG, Nesic Z, Drewitt GB, et al. (2003) Ecophysiological controls on the carbon balances of three southern boreal forests. Agr. Forest Meteorol. 117: pp. 53–71. doi: 10.1016/S0168-1923(03)00023-6.
17. Stoy PC, Katul GG, Siqueira MBS, Juang JY, McCarthy HR, Kim HS, et al. (2005) Variability in net ecosystem exchange from hourly to inter-annual time scales at adjacent pine and hardwood forests: a wavelet analysis. Tree Physiol. 25 (7): pp. 887–902. doi: 10.1093/treephys/25.7.887.
18. Kljun N, Black TA, Griffis TJ, Barr AG, Gaumont-Guay D, Morgenstern K, et al. (2006) Response of net ecosystem productivity of three boreal forest stands to drought. Ecosystems 9: pp. 1128–1144. doi: 10.1007/s10021-005-0082-x.
19. Welp LR, Randerson JT, Liu HP (2007) The sensitivity of carbon fluxes to spring warming and summer drought depends on plant functional type in boreal forest ecosystems. Agr. Forest Meteorol. 147: pp. 172–185. doi: 10.1016/j.agrformet.2007.07.010.
20. Park H, Hattori S (2004) Modeling scalar and heat sources, sinks, and fluxes within a forest canopy during and after rainfall events. J. Geophys. Res.-Atmos. (1984–2012) 109 (D14). doi: 10.1029/2003JD004360.
21. Matsumoto K, Ohta T, Nakai Y, Kuwada T, Daikoku K, Iida S, et al. (2008) Responses of surface conductance to forest environments in the Far East. Agr. Forest Meteorol. 148: pp. 1926–1940. doi: 10.1016/j.agrformet.2008.09.009.
22. Awal MA, Ohta T, Matsumoto K, Toba T, Daikoku K, Hattori S (2010) Comparing the carbon sequestration capacity of temperate deciduous forests between urban and rural landscapes in central Japan. Urban Forestry & Urban Greening 9(3): pp. 261–270. doi: 10.1016/j.ufug.2010.01.007.
23. Webb EK, Pearman GI, Leuning R (1980) Correction of flux measurements for density effects due to heat and water vapour transfer. Q. J. Roy. Meteor. Soc. 106: pp. 85–100. doi: 10.1002/qj.49710644707.
24. Toba T, Ohta T (2005) An observational study of the factors that influence interception loss in boreal and temperate forests. J. Hydrol. 313: pp. 208–220. doi: 10.1016/j.jhydrol.2005.03.003.
25. Hollinger DY, Kelliher FM, Byers JN, Hunt JE, McSeveny TM, Weir PL (1994) Carbon dioxide exchange between an undisturbed old-growth temperate forest and the atmosphere. Ecology 75: pp. 134–150.
26. Hogg EH, Price DT, Black TA (2000) Postulated feedbacks of deciduous forest phenology on seasonal climate patterns in the western Canadian interior. J. Climate 13: pp. 4229–4243. doi: 10.1175/1520-442(2000)013<4229:PFODFP>2.0.CO;2.
27. Blanken PD, Black TA, Yang PC, Neumann HH, Nesic Z, Staebler R, et al. (1997) Energy balance and canopy conductance of a boreal aspen forest: Partitioning overstorey and understorey components. J. Geophys. Res. 102 (D24): pp. 28915–28927. doi: 10.1029/97JD00193.
28. Wang H, Saigusa N, Yamamoto S, Kondo H, Hirano T, Toriyama A, et al. (2004) Net ecosystem CO₂ exchange over a larch forest in Hokkaido, Japan. Atmos. Environ. 38: pp. 7021–7032. doi: 10.1016/j.atmosenv.2004.02.071.
29. Villar R, Held AA, Merino J (1995) Dark leaf respiration in light and darkness of an evergreen and a deciduous plant species. Plant Physiol. 107: pp. 421–427. doi: 10.1104/pp.107.2.421.
30. Ruimy A, Jarvis PG, Baldocchi DD, Saugier B (1995) CO₂ fluxes over plant canopies and solar radiation: a review. Adv. Ecol. Res. 26: pp. 1–68.
31. Chen WJ, Black TA. Yang PC, Barr AG, Neumann HH, Nesic Z, et al. (1999) Effect of climatic variability on the annual carbon sequestration by a boreal aspen forest. Glob. Change Biol. 5: pp. 41–53.
32. Hirano T, Hirata R, Fujinuma Y, Saigusa N, Yamamoto S, Harazono Y, et al. (2003) CO₂ and water vapor exchange of a larch forest in northern Japan. Tellus B 55: pp. 244–257. doi: 10.1034/j.1600-0889.2003.00063.x.
33. Wofsy SC, Goulden ML, Munger JW, Fan SM, Bakwin PS, Daube BC, et al. (1993) Net exchange of CO₂ in a mid-latitude forest. Science 260: pp. 1314–1317. doi: 10.1126/science.260.5112.1314.
34. Aubinet M, Chermanne B, Vandenhaute M, Longdoz B, Yernaux M, Laitat E (2001) Long term carbon dioxide exchange above a mixed forest in the Belgian Ardennes. Agr. Forest Meteorol. 108: pp. 293–315. doi: 10.1016/S0168-1923(01)00244-1.
35. Pilegaard K, Hummelshøj P, Jensen NO, Chen Z (2001) Two years of continuous CO₂ eddy-flux measurements over a Danish beech forest. Agr. Forest Meteorol. 107: pp. 29–41. doi: 10.1016/S0168-1923(00)00227-6.
36. Saigusa N, Yamamoto S, Murayama S, Kondo H, Nishimura N (2002) Gross primary production and net ecosystem exchange of a cool-temperate deciduous forest estimated by the eddy covariance method. Agr. Forest Meteorol. 112: pp. 203–215. doi: 10.1016/S0168-1923(02)00082-5.
37. Valentini R, Matteucci G, Dolman AJ, Schulze ED, Rebmann C, Moors EJ, et al. (2000) Respiration as the main determinant of carbon balance in European forests. Nature 404: pp. 861–865. doi: 10.1038/35009084.
38. Jarvis PG, Leverenz JW (1983) Productivity of temperate, deciduous and evergreen forests. In: Lange OL, Nobel PS, Osmond CB, Ziegler H, eds. Physiological Plant Ecology IV. Springer-Verlag, Berlin, pp. 233–280. ISBN: 978-3642681561.
39. Field CB, Mooney HA (1986) The photosynthesis-nitrogen relationship in wild plants. In Givinish TJ (ed.) On the Economy of Plant Form and Function: Proceedings of the Sixth Maria Moors Cabot Symposium, Evolutionary Constraints on Primary Productivity, Adaptive Patterns of Energy Capture in Plants, Harvard Forest, August 1983 (Vol. 6). Cambridge University Press, Cambridge, UK. pp. 25–55. ISBN: 978-0521262965.
40. Evans JR (1989) Photosynthesis and N relationships in leaves of C₃ plants. Oecologia 78: pp. 9–19. doi: 10.1007/BF00377192.
41. Reich PB, Walters MB, Ellsworth DS (1992) Leaf life-span in relation to leaf, plant, and stand characteristics among diverse ecosystems. Eco1. Monogr. 62: pp. 365–392. doi: 10.2307/2937116.
42. Ministry of Agriculture Forestry & Fisheries (MAFF) (2003) Forestry Census: 1960-2000, Norin Toukei Kyoukai, Tokyo, Japan. ISBN: 978-4541030375.
43. Merino J, Field C, Mooney HA (1982) Construction and maintenance costs of Mediterranean-climate evergreen and deciduous leaves. I. Growth and CO₂ exchange analysis. Oecologia 53: pp. 208–213. doi: 10.1007/BF00545665.
44. Tans PP, Fung IY, Takahashi T (1990) Observational constraints on the global atmospheric CO₂ budget. Science 247: pp. 1431–1438. doi: 10.1126/science.247.4949.1431
45. Ciais P, Tans PP, Trolier M, White WC, Francey RJ (1995) A large northern hemisphere terrestrial CO₂ sink indicated by the ¹³C/¹²C ratio of atmospheric CO₂. Science 269: pp. 1098–1102.
46. Keeling CD, Chin JFS, Whorl TP (1996) Increased activity of northern vegetation inferred from atmospheric CO₂ measurements. Nature 382: pp. 146–149.
47. Ohtani Y, Saigusa N, Yamamoto S, Mizuguchi Y, Watanabe T, Yasuda Y (2005) Characteristics of CO₂ fluxes in cool-temperate coniferous and deciduous broadleaf forests in Japan. Phyton 45: pp. 73–80.
48. Gower ST, Vogel JG, Norman JM, Kucharik CJ, Steele SJ, Stow TK (1997) Carbon distribution and aboveground net primary production in aspen, jack pine, and black spruce stands in Saskatchewan and Manitoba, Canada. J. Geophys. Res.-Atmos. (1984–2012) 102: pp. 29029–29041. doi: 10.1029/97JD02317.
49. Greco S, Baldocchi DD (1996) Seasonal variations of CO₂ and water vapour exchange rates over a temperate deciduous forest. Glob. Change Biol. 2: pp. 183–197. doi: 10.1111/j.1365-2486.1996.tb00071.x.
50. Saigusa N, Yamamoto S, Hirata R, Ohtani Y, Ide R, Asanuma J, et al. (2008) Temporal and spatial variations in the seasonal patterns of CO₂ flux in boreal, temperate, and tropical forests in East Asia. Agr. Forest Meteorol. 148: pp. 700–713. doi: 10.1016/j.agrformet.2007.12.006.
51. Harmon ME, Ferrell WK, Franklin JF (1990) Effects of carbon storage of conversion of old-growth forests to young forests. Science 247: 699–702. doi: 10.1126/science.247.4943.699.
52. Jarvis PG, Ibrom A, Linder S (2005) ‘Carbon forestry’: managing forest to conserve carbon. In: Griffiths H, Jarvis PG, eds. The Carbon Balance of Forest Biomes. Taylor & Francis Group, London. pp. 331–349. ISBN: 978-1135322588.