Long-term, short-interval measurements of the frequency distributions of the photosynthetically active photon flux density and net assimilation rate of leaves in a cool-temperate forest

Agricultural and Forest Meteorology 152 (2012) 1–10

Contents lists available at ScienceDirect

Agricultural and Forest Meteorology journal homepage: www.elsevier.com/locate/agrformet

Long-term, short-interval measurements of the frequency distributions of the photosynthetically active photon flux density and net assimilation rate of leaves in a cool-temperate forest Ayana Miyashita ∗, Daisuke Sugiura, Koichiro Sawakami, Ryuji Ichihashi 1, Tomokazu Tani 2, Masaki Tateno Nikko Botanical Gardens, Graduate School of Science, University of Tokyo, Nikko, Tochigi 321-1435, Japan

a r t i c l e

i n f o

Article history: Received 4 February 2011 Received in revised form 21 July 2011 Accepted 5 August 2011 Keywords: PPFD Photosynthetic capacity Leaf productivity Forest understory Light availability

a b s t r a c t Long-term, short-interval measurements of incident photosynthetically active photon flux density (PPFD; ␮mol m−2 s−1 ) on the forest floor are essential for estimating the leaf carbon gain of understory plants. Such PPFD data, however, are scarce. We measured PPFD at 1-min intervals for more than 12 months in cool-temperate forest sites and reported the data as a PPFD frequency distribution. We chose five sites: an open site (OPN), the understory of a deciduous broad-leaved tree stand with no visible gaps (DCD), that of an evergreen conifer stand (EVG), that of a deciduous broad-leaved tree stand with a gap of approximately 80 m2 (GAPDCD), and that of an evergreen conifer stand with a gap of approximately 100 m2 (GAPEVG). DCD were divided into three sub sites (DCD1–3) to investigate variation within a small area. GAP-sites were consisted of two sub sites (GAPDCD1–2 and GAPEVG1–2) differing in the distance from the gap center. Using the PPFD data, we estimated the summer seasonal (May–October) net assimilation rate of leaves (NARL ) at each site for various photosynthetic capacities (Amax : ␮mol m−2 s−1 ) and other parameters of a light response curve of CO2 assimilation rates. At OPN, the average daily accumulated PPFD (mol m−2 day−1 ) was highest in May (28.2) and lowest in December (8.2). Even at OPN, the class of instantaneous PPFD that contributed most to NARL was 250–300 ␮mol m−2 s−1 . Such a large contribution of lower PPFD is suggested to be an important feature of a field light-availability. At DCD, the relative PPFD (RPPFD, %) to OPN was 7.2 during canopy closure and 49.4 after leaf shedding (averaged for 3 sites). EVG had the lowest light environment throughout the year. Its average RPPFD was 3%. For GAP sites, summer seasonal RPPFD (%) was 15.6, 18.8, 6.4 and 15.6 for GAPDCD1, GAPDCD2, GAPEVG1 and GAPEVG2, respectively. At OPN, the NARL increased with Amax (which ranged from 1 to 40), suggesting that plants at OPN do not maximize NARL . In contrast, at DCD and EVG, Amax values were attained that did maximize NARL , suggesting that plants at these sites could maximize the NARL . Amax –NARL relationships for GAPDCD and GAPEVG showed similar trend to closed canopy sites, DCD and EVG, while NARL and A∗max of GAP sites were larger than at these sites. Among DCD1–3, the daily accumulated PPFD (mol m−2 day−1 ) averaged in summer ranged 1.3–1.8 and the maximum NARL value differed up to 1.5 times. It indicates that Amax and NARL can be various among plants under a similar canopy conditions. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Knowledge of the frequency distributions of photosynthetic photon flux density (PPFD) is critical for determining leaf carbon gain (Tang et al., 1999). Leaf productivity per unit time can

∗ Corresponding author. Tel.: +81 288 54 0206; fax: +81 288 54 3178. E-mail address: [email protected] (A. Miyashita). 1 Present address: Fukuoka Research Forest, Kyushu University, Sasaguri, Fukuoka 811-2415, Japan. 2 Present address: Department of Science Education (Biology), Joetsu University of Education, 1-Yamayashiki, Joetsu, Niigata 943-8512, Japan. 0168-1923/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.agrformet.2011.08.001

be estimated using a light-response curve of the CO2 assimilation rate with incident PPFD on the leaf surface. Results can differ with the frequency distribution of PPFD, even if the light response curve remains the same, which suggests that any ecophysiological estimation based on leaf productivity, such as the whole-plant growth rate or nutrient use efficiency, depends on the frequency distribution of PPFD. Takenaka (1989) theorized that the relative frequency distribution of PPFD rather than its average critically affects the predicted optimal photosynthetic capacity at which leaf carbon gain is maximized. Therefore, estimations of leaf productivity with accurate frequency distributions of PPFD can be crucial. In particular, for field-grown plants, estimations of leaf productivity require long-term PPFD measurements because under field

2

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

conditions, long-term frequency distributions of PPFD are unpredictable due to influences of weather conditions and canopy tree phenology, as well as other incidental elements. Thus, PPFD measurements should at least encompass an entire growing season. However, the availability of long-term (a year or a growing season) PPFD data measured at short intervals under field conditions is limited, although many studies have shown detailed short-term (only a few days or months) PPFD patterns at various field sites (Brantley and Young, 2009; Canham et al., 1994; Chazdon and Fetcher, 1984; Chazdon and Field, 1987; Muraoka et al., 2001; Pearcy, 1983; Pfitsch and Pearcy, 1989; Tang et al., 1999). Furthermore, most studies estimating leaf productivity have not applied such PPFD data to the estimation (e.g., at a cool-temperate forest: Chen and Klinka, 1997; Ida and Kudo, 2008; Kobe and Hogarth, 2007; Muller et al., 2005; Washitani, 1992). Here, we addressed two main objectives. Our first goal was to obtain year-round high-resolution PPFD data for cool-temperate forest sites. For this purpose, we measured PPFD at one high-light site, two low-light sites and two gap sites within a cool-temperate forest in Japan: an open site (OPN), the understory of a deciduous broad-leaved tree stand (DCD), that of an evergreen conifer stand (EVG), that of a deciduous broad-leaved tree stand with a gap (GAPDCD), and that of an evergreen conifer stand with a gap (GAPEVG). DCD was divided into three sub sites (DCD1–3) to investigate the variation of light availability under the similar canopy conditions. Each GAP site was consisted of two sub sites (GAPDCD1&2 or GAPEVG1&2) differing in the distance from the gap center. These PPFD data are reported as a frequency distribution of PPFD. Our second objective was to estimate the summer seasonal net assimilation rate of leaves (NARL : mol CO2 m−2 season−1 ) at OPN, DCD, EVG, GAPDCD and GAPEVG. For this purpose, we calculated the NARL using measured PPFD data for various photosynthetic capacities (Amax ) and other parameters of a light response curve of CO2 assimilation rates, such as initial slope, convexity and dark respiration rate to Amax ratio. The results of the present study provide a useful basis for understanding plant growth and ecophysiological traits under various light availability conditions in a cool-temperate forest.

2. Materials and methods 2.1. Study sites The study was conducted in the Nikko Botanical Gardens (NBG), Tochigi, Japan (36◦ 45 N, 139◦ 35 E, 647 m above sea level), from 2007 to 2011. The mean annual air temperature and precipitation during the experimental period was 12.26 ◦ C and 2144 mm, respectively. NBG is located in a cool-temperate zone, and the surrounding natural forest is dominated by deciduous broad-leaved (Fagus japonica, Quercus crispula, and Acer spp.) and evergreen coniferous (Abies firma and Tsuga sieboldii) tree species. PPFD measurements were conducted at five sites: an open site (OPN); the understory of a deciduous closed canopy stand (DCD) of Fagus crenata Blume and F. japonica Maxim., both planted about 30 years ago and grown under natural conditions; the understory of an evergreen closed canopy stand (EVG) of A. firma Siebold et Zucc. and Cryptomeria japonica [L.f.] D. Don, both planted about 100 years ago and grown under natural conditions; the understory of a deciduous stand of Acer spp. and other some deciduous tree species planted about 100 years ago with a gap about 10 m in diameter (GAPDCD); and the understory of A. firma stand planted about 100 years ago mixed with some deciduous trees with a gap, which is north–south rectangular and about 100 m2 in area (GAPEVG). DCD, GAPDCD and GAPEVG were divided into two or three sub-sites. DCD was divided into DCD1, DCD2 and DCD3 and the centers of the sites are about

5 m apart from each other. DCD1–3 were under the similar canopy conditions. GAPDCD and GAPEVG were consisted of two sub sites. GAPDCD1 and GAPDCD2 were established at the south-west edge of the gap and the sites were 5 m apart from each other. GAPDCD1 was canopy-side and GAPDCD2 was gap-side. GAPEVG1 was under the canopy and 5 m apart from GAPEVG2 established at the north edge of the gap. At DCD1–3 and EVG, the canopies were closed and no visible gaps were present. At each of the nine sites, OPN, DCD1–3, EVG, GAPDCD1–2 and GAPEVG1–2, we established an approximately 5m2 sampling site from which we removed all understory vegetation to minimize any shading effect on our understory light measurements, although understory vegetation was scarce at DCD1–3 and EVG. 2.2. Measurements at the study sites 2.2.1. PPFD We placed a quantum sensor (for OPN, DCD1 and EVG: Quantum Light Sensor, Spectrum Technologies, Inc., Plainfield, IL, USA; for DCD2–3, GAPDCD1–2 and GAPEVG1–2: Photosynthetic Light (PAR) Sensors, Onset Computer Corp.) in the center of each site. The sensors were set at a height of 50 cm above the ground and carefully leveled. A datalogger (for OPN, DCD1 and EVG: WatchDog Data Logger, Model 400, Spectrum Technologies, Inc.; for DCD2–3, GAPDCD1–2 and GAPEVG1–2: HOBO Micro Station Data Logger, Onset Computer Corp.) connected to each sensor was used to sample the sensor output every minute. We used a 1-min interval because we considered it to be sufficiently short to capture shorttime fluctuations such as “sunflecks” while keeping the quantity of data manageable (cf. Pearcy, 1983). These measurements were conducted from 1 May 2007 to 26 September 2008 for OPN, DCD1 and EVG, from 1 June 2010 to 27 May 2011 for DCD2–3, and from 23 June 2010 to 30 June 2011 for GAPDCD1–2 and GAPEVG1–2. We defined the summer growing season as 1 May–31 October and the winter season as the rest of the year because leaf expansion starts at the end of April and leaf shedding begins in early November at NBG. In winter we have some days with snow cover, but the depth of snow cover did not exceed the height of the sensor during the experimental period. For EVG, measurements ceased on 29 February 2008 because a large A. firma tree was uprooted near the site at the end of February 2008, leaving a large gap. For GAPDCD2 and GAPEVG1–2, the data from 27 November 2010 to 4 February 2011 and from 20 December 2010 to 4 February 2011, respectively, are lacking due to the sensor and/or logger troubles. From the 1-min interval PPFD data, frequency distributions of PPFD were calculated for each month at each site. We had PPFD data for 1–2 years for each month; in the latter case, the data were averaged. Averaged daily accumulated PPFD (mol m−2 day−1 ) was determined every month. We multiplied each measured PPFD by 60 s and then summed them across the month, and then divided it by the number of days with available data in the month. Daily accumulated PPFD averaged in summer or winter season was also calculated from summed PPFD across the season. Relative PPFD (RPPFD; %) was calculated for DCD1–3, EVG, GAPDCD1–2 and GAPEVG1–2 as the ratio of accumulated PPFD to that of OPN. 2.2.2. Air temperature Air temperature (T, ◦ C) was also monitored and logged at 1-min intervals at each site (for OPN, DCD1 and EVG: Temperature (Micro) Sensor, Spectrum Technologies, Inc.; for DCD3–2 and GAP-sites: 12-bit Temperature Smart Sensors, Onset Computer Corp.). 2.2.3. Leaf temperature For the estimation of NARL , leaf temperature (LT, ◦ C) was estimated from T and PPFD. We first measured the LT using three thermocouples (Type T 6 ft Beaded Thermocouple Sensor, Onset

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

3

Computer Corp., Bourne, MA, USA) attached to the abaxial sides of leaves. Readings were logged at 1-min intervals (HOBO Thermocouple Logger, Onset Computer Corp.). PPFD and T were also monitored (Photosynthetic Light (PAR) Sensors and 12-bit Temperature Smart Sensors, Onset Computer Corp.) and logged (HOBO Micro Station Data Logger, Onset Computer Corp.) simultaneously. These measurements were conducted in August 2009 at OPN with leaves of an Acer diabolicum seedling. Results of this monitoring showed that at night, LT nearly equaled T (LT = T − 2.0 × 10−10 , r2 = 0.9590). Daytime LT (PPFD > 0) also closely resembled T, yet a multiple regression with T and PPFD produced a better correlation (LT = 0.900775508 × T + 0.002368479 × PPFD + 2.200475896, r2 = 0.9575).

Section 2.2.3, LT during a summer season was estimated as follows:

2.3. Calculation of NARL

Rd (LTK ) = Rd (25) · exp

2.3.1. Parameters of light-response curves Summer seasonal net assimilation rate of leaves (NARL : mol CO2 m−2 season−1 ) was calculated by accumulating instantaneous net assimilation rates of leaves (A, ␮mol CO2 m−2 s−1 ) during the season. The results were shown as the value of 184 days (1 May–30 October). For the calculation, the May–October PPFD data for OPN, DCD1–3, EVG, GAPDCD1–2 and GAPEVG1–2, and hypothetical light-response curves were used. A light-response curve is the steady-state rate of CO2 assimilation, which increases asymptotically with increasing irradiance. The shape of a light-response curve can be described by a non-rectangular hyperbola (Thornley, 1976). A at a given PPFD is

where Rd (25) is the value of Rd at 25 ◦ C, Ea is the activation energy of Rd (66.405 kJ mol−1 ; Farquhar et al., 1980), R is the gas constant (8.314 J mol−1 K−1 ), and LTK is the LT in K. In the summer season in NBG, more than 65% of measured T were within the range of 15–25 ◦ C and relatively few days occurred with T and LT > 30 ◦ C. In addition, the PPFD dependence of Rd was added as follows (Oguchi et al., 2008): Rd was not modified when PPFD = 0 ␮mol m−2 s−1 , but Rd was multiplied by 0.4 when PPFD > 0. The LT dependence on gross assimilation rates was not included in this work for the following reason. Sugiura and Tateno (unpublished results) measured the gross assimilation rates of leaves of some deciduous tree species by varying the LT and PPFD, and expressed the gross assimilation rate relative to the value at 25 ◦ C as a function of LT. They calculated NARL with or without the LT dependence of gross assimilation rates and showed that these values of NARL differed by only 5% or less.

A=

˚ ∗ PPFD + Amax −



2

(˚ ∗ PPFD + Amax ) − 4˚ ∗ PPFD ∗  ∗ Amax 2

− Rd

(1)

where Amax is the light-saturated rate of gross CO2 assimilation at infinitely high irradiance, ˚ is the initial slope of the line representing the ration of moles of CO2 assimilated to moles of incident PPFD (mol CO2 mol quanta−1 ),  is the curvature factor (nondimensional), and Rd is dark respiration rate of leaves (␮mol CO2 m−2 s−1 ). We substituted our PPFD data and calculated the seasonal CO2 assimilation rate, NARL , assuming that the PPFD lasted for 1 min. For the calculations, Amax , ˚,  and Rd to Amax ratio (Rd /Amax ) were changed separately. Amax was changed stepwise from 1 to 40 ␮mol CO2 m−2 s−1 . ˚ was set at 0.04, 0.05 or 0.07 mol CO2 mol quanta−1 , that falls within the commonly observed range. The maximum value of the quantum yield based on the incident light measured at CO2 saturation is generally held to be remarkably constant among non-stressed C3 vascular species and to have a value of 0.08–0.09 (Bjorkman and Demmig, 1987). However, actual values of ˚ measured at ambient CO2 mostly fall in the range of 0.04–0.07 (Ehleringer and Bjorkman, 1977; Hikosaka et al., 1999; Hirose and Werger, 1987; Koyama and Kikuzawa, 2010; Miyazawa and Kikuzawa, 2005; Veneklaas and Poorter, 1998). The value of  for the photosynthetic light-response curve can be assumed to be an empirical factor (Cannell and Thornley, 1998; Ogren and Evans, 1993). We chose a value for  of 0.5, 0.8 or 0.99, that covers the range of commonly observed values (Cannell and Thornley, 1998; Hikosaka et al., 1999; Hirose and Werger, 1987; Koyama and Kikuzawa, 2010; Veneklaas and Poorter, 1998). Rd apparently depends on the leaf nitrogen content, and in previous studies, the values of Amax and Rd were expressed as linear functions of area-based nitrogen content (Hikosaka et al., 1999; Hirose and Werger, 1987), where the Rd /Amax ratio ranged from 1/10 to 1/20. Here, Rd /Amax was set at a constant 1/10, 1/15 or 1/20 at a LT of 25 ◦ C. 2.3.2. Leaf temperature dependence of Rd and gross assimilation rates The dark respiration rate, Rd , changed depending on the LT relative to the value at 25 ◦ C. Based on the calculations described in

LT = 0.900775508 × T + 0.002368479 × PPFD + 2.200475896 (PPFD > 0)

(2.1)

or LT = T

(PPFD = 0)

(2.2)

The LT dependence of Rd was fitted using the Arrhenius model following Hikosaka et al. (2006):

 E · (LT − 298)  a K 298 · R · LTK

(3)

2.4. Amax data collecting To evaluate our results, Amax values of plants grown under field conditions or natural light resources were collected from published works and our own measurements. The result was shown as a frequency distribution of Amax in Fig. 8. References for “sun leaves”: Bazzaz and Carlson (1982) 5 herbaceous species and 9 deciduous tree species “grown in full sunlight”; Koike (1988) “sun leaves” from 28 deciduous broad-leaved tree species; Reich et al. (1995) “sun leaves in high light microenvironments” from 28 deciduous broad-leaved tree species; and Osone and Tateno (2003) “leaves exposed to the sun” from 41 herbaceous species. References for “shade leaves”: Bazzaz and Carlson (1982) 5 herbaceous species and 9 deciduous tree species grown in understory, where 1% of full sunlight; Reich et al. (1998) 4 deciduous broad-leaved trees species grown at 5% of full sun shaded by cloth in green house; and our own measurements explained below. We measured Amax for various deciduous species naturally occurring at DCD to compensate for the luck of Amax data published. 10 perennial herbs [Angelica pubescens, Boehmeria silvestrii, Cardiandra alternifolia, Cimicifuga simplex, Dioscorea nipponica (liana), Elatostema japonicum, Fallopia japonica, Geum japonicum, Persicaria filiformis and Persicaria thunbergii] and 4 deciduous tree species [Akebia quinata (liana), Juglans mandshurica (seedling), Rubus palmatus and Smilax china (liana)]. Amax were measured for fully expanded, young, mature leaves in the early to late morning using an infrared gas-exchange analyzer (Li-6400, Li-Cor, Lincoln, NE, USA) with an ambient CO2 concentration of 380 ␮mol mol−1 , leaf temperature of 25 ◦ C, relative humidity of 75%, and saturating PFD of 700 ␮mol m−2 s−1 in August 2010.

4

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

5 4

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

3 2 1

frequency (103)

0 4 3 2 1 0 4 3 2 1 0

0

500

1000

1500

2000

0

500

1000

1500

2000

0

PPFD (µmol

500

m-2

1000

1500

2000

0

500

1000

1500

2000

2500

s-1)

Fig. 1. Frequency distributions of the measured PPFD at OPN for each month. PPFD was logged at 1-min intervals from 1 May 2007 to 26 September 2008. We did not include readings of 0 ␮mol m−2 s−1 , and the class width of the frequency distribution is 50 ␮mol m−2 s−1 .

3. Results 3.1. Characteristics of light availability at the study sites OPN was a high-light environment throughout the year. The average accumulated PPFD was 21.8 mol m−2 day−1 in the summer (May–October) and 15.0 in the winter (November–April). However, even in this environment, most PPFD was <500 ␮mol m−2 s−1 (Fig. 1); in the summer, PPFD <500 ␮mol m−2 s−1 accounted for 69.5% of the accumulated frequency and 27.6% of the accumulated PPFD. Light availability at OPN is maximal in May (Table 1, Fig. 1) because mid-June to mid-July, when peak irradiance occurs, falls within the rainy season at the study site. NBG usually has the most rain in July. At DCD, the frequency distributions of PPFD changed seasonally. Monthly patterns of PPFD showed that during June–October,

the canopy trees were in full leaf and most values of PPFD were <50 ␮mol m−2 s−1 (Table 1, Fig. 2). RPPFD of DCD1–3 during the summer season was 6.0, 8.3 and 7.3%, respectively. However, after leaf-expansion completing (June–October), RPPFD fell to 2.7, 1.6 and 1.5% for DCD1–3, respectively. At these sites, sunflecks (defined here as PPFD exceeding 50 ␮mol m−2 s−1 ; cf. Pearcy, 1983, 1988) accounted for 14.1, 21.7 and 21.3% of the accumulated PPFD frequency and 73.9, 79.8 and 77.7% of the accumulated PPFD in summer, respectively. These values are larger than those published (e.g., tropical rain forests: Chazdon and Fetcher, 1984; Pearcy, 1983, 1988; temperate forests: Canham et al., 1994). In the winter season, RPPFD increased to 50.5, 48.2 and 49.5% for DCD1–3, respectively, because the canopy trees had shed their leaves. Among DCD1–3, there were no explicit differences in the shape of PPFD frequency distributions and the yearly trend of monthly averaged daily accumulated PPFD. However, there were up to 2.3 times differences in

Table 1 Averaged daily accumulated PPFD for each month at each site. For more details of PPFD measurement, see legend to Fig. 4. Month

Averaged daily total PPFD (mol m−2 day−1 ) OPN

January February March April May June July August September October November December –: no data.

11.2 15.8 19.0 26.0 28.2 23.6 17.8 24.8 18.4 11.7 9.1 8.2

DCD

EVG

1

2

3

5.4 8.2 11.9 15.0 4.2 0.8 0.3 0.5 0.6 0.6 2.6 3.7

6.3 9.0 11.1 15.2 9.5 0.7 0.5 0.5 0.4 0.4 4.1 5.6

6.6 9.7 12.7 13.5 8.1 0.5 0.4 0.4 0.4 0.5 4.1 6.1

GAPDCD 1

0.8 1.0 – – 1.3 0.2 0.1 0.2 0.2 0.2 0.4 0.5

7.2 11.8 15.2 15.1 7.1 2.9 2.9 2.8 2.6 2.7 5.9 5.8

GAPEVG 2 – 12.2 14.5 17.0 9.2 3.4 3.9 3.5 2.5 2.5 6.2 –

1

2

– 3.4 6.0 4.5 2.4 0.9 0.9 1.6 1.8 1.1 1.8 2.5

– 3.8 7.4 10.1 6.3 2.5 2.9 4.0 3.6 1.3 2.5 3.3

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

5

5

Jan

4

Feb

Mar

Apr

3 2 1

frequency (103)

0

10.4

8.2

May

4

7.6

Jun

8.9

Jul

Aug

3 2 1 0

9.2

10.2

Sep

4

8.1

Oct

5.8

Nov

Dec

3 2 1 0

0

500

1000

1500

2000

0

500

1000

1500

2000

0

500

1000

1500

2000

0

500

1000

1500

2000

2500

PPFD (µmol m-2 s-1) Fig. 2. Frequency distributions of the measured PPFD at DCD1 for each month. The measurement was conducted from 1 May 2007 to 26 September 2008. The numerical value in the upper left of each panel shows the frequency value of the minimum class, which overs the y-axis scale. For more details, see legend to Fig. 1.

5

10.1

Jan

10.7

Feb

11.7

May

5.6

Jun

4 3 2 1

frequency (103)

0 4

Jul

Aug

3 2 1 0

Sep

4

5.5

Oct

8.1

Nov

Dec

3 2 1 0

0

500

1000

1500

2000

0

500

1000

1500

2000

0

500

1000

1500

2000

0

500

1000

1500

2000

2500

PPFD (µmol m-2 s-1) Fig. 3. Frequency distributions of the measured PPFD at EVG for each month. The measurement was conducted from 1 May 2007 to 29 February 2008. The numerical value in the upper left of each panel shows the frequency value of the minimum class, which overs the y-axis scale. For more details, see legend to Fig. 1.

the values of monthly averaged daily accumulated PPFD (Table 1)

6

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

4

OPN (21.8)

3

EVG (0.3)

2

1

4

5.3

GAPDCD1 (3.4)

3

4.8

GAPDCD2 (4.1)

frequency (104)

2

1

4

7.1

GAPEVG1 (1.4)

3

6.9

GAPEVG2 (3.4)

2

1

4

5.8

DCD1 (1.3)

3

5.5

5.6

DCD3 (1.6)

DCD2 (1.8)

2 1 0 0

500

1000

1500

2000

2500

500

1000

1500

PPFD (µmol

m-2

2000

2500

500

1000

1500

2000

2500

s-1)

Fig. 4. Summer seasonal frequency distributions of the measured PPFD at each site. The “summer season” was defined as 1 May–31 October. The class width of the frequency distribution is 50 ␮mol m−2 s−1 . We did not include readings of 0 ␮mol m−2 s−1 . The numerical values in parenthesis show seasonal average of daily accumulated PPFD (mol m−2 day−1 ) at each site. The numerical value in the upper left of each panel shows the frequency value of the minimum class, which overs the y-axis scale. PPFD was logged at 1-min intervals at each site. The measurements were conducted from 1 May 2007 to 26 September 2008 for OPN and DCD1, from 1 May 2007 to 29 February 2008 for EVG, from 1 June 2010 to 27 May 2011 for DCD2–3, and from 23 June 2010 to 30 June 2011 for GAPDCD1–2 and GAPEVG1–2.

and 1.4 times in the summer seasonal averaged daily accumulated PPFD (Fig. 4). EVG was the darkest site throughout the year (Table 1, Fig. 3). RPPFD was 1.5% and sunflecks accounted for 8.4% of the accumulated PPFD frequency and 53.5% of the accumulated PPFD during the summer. The results show that less light was available in the understory of the closed evergreen coniferous canopy than in that of the deciduous broad-leaved canopy. Light availability appeared to be higher in winter. At the GAP sites, most of the measured PPFD were <100 ␮mol m−2 s−1 and the shape of the frequency distributions were similar to that of closed canopy sites, DCD1–3 and EVG, rather than that of OPN. However, for the PPFD ≥50 ␮mol m−2 s−1 , the fre-

quency was larger than that of DCD1–3 and EVG. At GAPDCD1 and 2, RPPFD of the summer season were 15.6 and 18.8%, respectively. In winter, RPPFD was around 67%, on average, which was higher than that of DCD1–3. At GAPEVG1 and 2, summer seasonal RPPFDs were 6.4 and 15.6%, respectively. In winter, light availability was better at GAPEVG because of existence of deciduous trees in evergreen conifers. 3.2. NARL at OPN, DCD1–3, EVG, GAPDCD1&2 and GAPEVG1&2 The NARL with changing Amax of deciduous leaves exhibited different patterns for high-light environments (OPN) and low-light environments (DCD1–3 and EVG) (Fig. 5, for the case

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

25

120

12

OPN

GAPEVG

GAPDCD

100

7

10

20

8

80 15

6

NARL (mol CO2 m-2 season-1)

60 10

4

40 5

20 0

2

0

0

10

20

30

0

40

0

5

10

15

20

0

5

10

15

1

5

DCD

EVG

4

3 0 2

1

-1

0

0

2

4

6

8

10

0

1

2

3

4

5

Amax (µmol m-2 s-1) Fig. 5. Effect of Amax on NARL for deciduous leaves at OPN, GAPDCD, GAPEVG, DCD and EVG. In the panels of GAPDCD and GAPEVG, bold line and solid line represents site-1 and -2, respectively. In the panel of DCD, bold line, solid line and dashed line represents site-1, -2 and -3, respectively. All results were calculated based on PPFD data of summer seasonal (1 May–31 October). Other parameters of light-curve, ˚,  and Rd /Amax was set at 0.05, 0.8 and 1/15, respectively.

of 2.8 and 1 ␮mol CO2 m−2 s−1 for DCD1 and EVG, respectively, maximize the NARL values. Amax –NARL relationships of GAP-sites show similar patterns to DCD and EVG, but NARL and A∗max were larger than that of such a closed canopy sites (Fig. 5). The change in ˚,  and Rd /Amax (˚: 0.04–0.08, : 0.5–0.99 and Rd /Amax : 20/1–10/1, respectively) affected the Amax –NARL curves (Fig. 6). At OPN, ˚ and  had the similar effect and more than Rd /Amax did. NARL (mol CO2 m−2 season−1 ) was varied from 59 to

of ˚ = 0.05,  = 0.8 and Rd /Amax = 1/15). For OPN, the NARL increased with increasing Amax . The NARL ranged from 0 to 110 mol CO2 m−2 season−1 . At OPN, NARL behaved as a saturating function of Amax , which ranged from 0 to 40 ␮mol CO2 m−2 s−1 . In contrast, for DCD1–3 and EVG, the NARL showed a convex upward pattern with increasing Amax because large values of Amax correspond to large values of Rd that reduce NARL . There are Amax values that maximize NARL values (A∗max ). For example, the Amax values

6

NARL (mol CO2 m-2 season-1)

200

2

100

0 0

EVG

DCD1

OPN

10

20

30

40

4

1

2

0

0

-1

-2

0

2

4

6

8

10

-2

0

1

2

3

4

5

Amax (µmol m-2 s-1) Fig. 6. Effect of initial slope (˚), convexity () and Rd /Amax on Amax –NARL relationship for deciduous leaves at OPN, DCD1 and EVG. ˚,  and Rd /Amax were changed separately. For each panel, bold lines represent (˚, , Rd /Amax ) = (0.04, 0.5, 1/20), (0.05, 0.8, 1/15 (same as Fig. 5)) and (0.07, 0.99, 1/10), for lower line, middle one and upper one, respectively. Solid lines represent the effect of ˚, where ˚ was set at 0.04 and 0.07 for lower line and upper one, respectively, and other parameters were the same as bold line (middle). Dashed lines represent the effect of , where  was set at 0.5 and 0.99 for lower line and upper one, respectively, and other parameters were the same as bold line (middle). Dotted lines represent the effect of Rd /Amax , where Rd /Amax was set at 1/20 and 1/10 for lower line and upper one, respectively, and other parameters were the same as bold line (middle).

8

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

8

25 OPN

20

4

15

2

10

0

5

100

DCD1 & EVG 80

DCD1 EVG

60

frequency

contribution (%)

sun leaves

Amax 10 Amax 40

6

0 20

shade leaves

40

15

20 0 0

500

1000

1500

2000

2500

10

PPFD (µmol m-2 s-1) Fig. 7. Contributions (%) of each PPFD class to GARL . Upper panel shows the results for OPN. The Amax values used were 10 and 40 ␮mol m−2 s−1 . Lower panel shows the results for DCD and EVG. The Amax values used were 2 and 1 ␮mol m−2 s−1 , respectively. All results were calculated based on light availability of summer season (1 May–31 October).

5 0 0

4

8

12

16

20

24

Amax (µmol 109 when Amax = 20 ␮mol CO2 m−2 s−1 and from 72 to 162 when Amax = 40 ␮mol CO2 m−2 s−1 . At low-light environment, DCD1 and EVG, changes in parameters also affected A∗max value (Fig. 6). At DCD1, change in ˚ positively affected NARL the most and Rd /Amax negatively did the most. When Amax = 4 ␮mol CO2 m−2 s−1 , NARL was varied from 1.0 to 4.9 mol CO2 m−2 season−1 . The most extreme case, A∗max was varied from 2 to 9 ␮mol CO2 m−2 s−1 . At EVG, where NARL values are small, the parameters can be critical for maintaining carbon balance. The effect of Rd /Amax on NARL was especially large. When Amax = 1.5, NARL was varied from −0.04 to 1.3 mol CO2 m−2 season−1 . The most extreme case, A∗max was varied from 0.5 to 2 ␮mol CO2 m−2 s−1 .

28

m-2

32

36

40

42

s-1)

Fig. 8. Frequency distributions of Amax values from wide range of plants around cool temperate zone. Data were collected from published works dealing with plants grown under field conditions or natural light resources and our own measurements (see Section 2.4). The class width of the frequency distribution is 2 ␮mol m−2 s−1 . “Sun leaves”: grown at an open environment and “shade leaves”: grown at 1–5% RPPFD.

The results suggest that Amax and NARL can vary among plants even under the similar canopy conditions. Eventually, precise light-availability at a point is never shown without an actual measurement. However, various light-availabilities from low to high investigated in this paper help us to predict light-availability under some canopy conditions in a cool temperate forest.

4. Discussion

4.2. Predicted Amax and NARL for a cool-temperate forest

4.1. Light availability characteristics at the study sites

Under high light availability, our results for OPN theoretically indicate that a larger Amax results in a larger NARL (Fig. 5). However, few species grown in full sunlight, or sun leaves, exhibit an Amax > 40 ␮mol m−2 s−1 in a cool-temperate zone (Fig. 8; Bazzaz and Carlson, 1982; Koike, 1988; Osone and Tateno, 2003; Reich et al., 1995). This indicates that NARL is not maximized under conditions of high light availability. In such high-light environments, maximizing the NARL (i.e., having A∗max ) is not realistic because A∗max is very large and having such a value of Amax requires a large amount of nitrogen and water per unit leaf area. Thus, plants may maximize whole-plant productivity not by maximizing leaf productivity but by adjusting biomass allocation patterns (i.e., allocation to leaves, stems, or roots) in relation to soil resource availability (Hilbert, 1990; Osone and Tateno, 2003, 2005). Therefore, Amax , and consequently NARL , may vary among species depending on plasticity in biomass allocation (Osone and Tateno, 2003), and also within species depending on nitrogen availability in the habitat (Hilbert, 1990; Osone and Tateno, 2003, 2005). In a low-light environment, however, our results suggest that plants do attain A∗max and maximum NARL . Published data and our own measurements show that the Amax of plants grown in a forest understory or shaded by cloth in a cool-temperate forest approximates the A∗max predicted for DCD (Fig. 8; Bazzaz and Carlson,

To further investigate the characteristics of light availability, we determined the contributions of each class of PPFD to the seasonal gross assimilation rate (GARL ; NARL + accumulated Rd (mol CO2 m−2 season−1 )) for each light availability site (Fig. 7). At OPN, PPFDs < 500 ␮mol m−2 s−1 contributed about 40% of the GARL , even when Amax was 40, although a larger Amax allowed higher PPFD classes to contribute more to the GARL . The class of PPFD that contributed the most to GARL was 250–300 ␮mol m−2 s−1 at OPN. At DCD1 and EVG, the lowest PPFD class (0–50 ␮mol m−2 s−1 ) contributed more than 70% of the GARL . Thus, light compensation points likely do not exceed 50 ␮mol m−2 s−1 at such understory sites. Our results strongly indicate that lower PPFD values are more important than higher PPFD values for leaf productivity under the field conditions, even in a high-light environment. Such characteristics of light availability in the field may not be detected by short-term PPFD measurements. Light regimes can differ at a small spatial scale (Chazdon and Fetcher, 1984; Chazdon and Pearcy, 1991; Chazdon et al., 1988). This is also shown in our results at DCD1–3. As a result, Amax –NARL curves and A∗max value were different among these sites (Fig. 5).

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

1982; Reich et al., 1998), yet Amax data under conditions of low light availability are still scarce. Under conditions of very low light availability, such as the environments of DCD and EVG, A∗max is not very high. Reaching A∗max does not require a large root biomass or an abundance of soil resources, except under severe resource limitation. Therefore, in a low-light environment, plants have Amax values close to A∗max and thus attain maximal NARL . In the case that the Amax –NARL curve is like a saturation curve, having A∗max may not be necessarily adaptive for plants, because around the peak of the Amax –NARL curve, NARL increases inefficiently with increasing Amax . For example, GAPDCD1, where A∗max = 12.5 (␮mol m−2 s−1 ) (Fig. 5), 95% of the maximum NARL is attained when Amax over 7.5 ((␮mol m−2 s−1 ). In such a light-availability, plants may have lower Amax than A∗max . Variety in other parameters of light-curves, ˚,  and Rd /Amax , may be associated with variety in NARL and Amax among species at the same site because these parameters can affect the Amax –NARL relationships especially under a low light-availability. More accurate estimations of leaf productivity under field conditions are required. In the present study, we did not account for photosynthetic induction and efficiency in a fluctuating light environment (Gross, 1982; Pearcy, 1988; Pfitsch and Pearcy, 1989) or leaf orientation and mutual shading (Pearcy et al., 2005). These parameters should be included in models, especially for forest understory sites.

5. Conclusions We measured PPFD at 1-min intervals for more than 12 months at nine sites in a cool-temperate forest (OPN, DCD1–3, EVG, GAPDCD1–2 and GAPEVG1–2) and reported the data as frequency distributions of PPFD for each month. Using these data, we then estimated the summer seasonal net assimilation rate of leaves (NARL ) for each site. Our results characterized in detail the light availability at each site and revealed some notable patterns: OPN was a high-light environment throughout the year, but it was characterized by a large contribution of PPFD <500 ␮mol m−2 s−1 to the frequency of PPFD, accumulated PPFD, and NARL . At OPN, the NARL increased with Amax , suggesting that there is not an optimal Amax value in the 0–40 ␮mol m−2 s−1 range that maximizes NARL at OPN. In contrast, at lower light availability, Amax values which maximized NARL were found, suggesting that plants could do so. Detailed data for PPFD and NARL can be applied to quantifying whole-plant growth rates. Few studies have examined relationships between plant ecophysiological traits, such as resource allocation ratio and leaf habit, and qualitative and quantitative measures of growth rate (Osone and Tateno, 2003, 2005). Further measurements of long-term, short-interval PPFD at various sites will improve our understanding of light availability in cool-temperate forests. Measurements of leaf temperature and the temperature dependence of photosynthetic rates in winter are required to estimate the winter NARL for evergreen species. Such work is required to better understand the variation in growth traits of species and mechanisms of regeneration and species composition in cool-temperate forests.

References Bazzaz, F.A., Carlson, R.W., 1982. Photosynthetic acclimation to variability in the light environment of early and late successional plants. Oecologia 54 (3), 313–316. Bjorkman, O., Demmig, B., 1987. Photon yield of O2 evolution and chlorophyll fluorescence characteristics at 77 K among vascular plants of diverse origins. Planta 170 (4), 489–504.

9

Brantley, S.T., Young, D.R., 2009. Contribution of sunflecks is minimal in expanding shrub thickets compared to temperate forest. Ecology 90 (4), 1021–1029. Canham, C.D., Finzi, A.C., Pacala, S.W., Burbank, D.H., 1994. Causes and consequences of resource heterogeneity in forests: interspecific variation in light transmission by canopy trees. Can. J. Forest Res. 24 (2), 337–349. Cannell, M.G.R., Thornley, J.H.M., 1998. Temperature and CO2 Responses of leaf and canopy photosynthesis: a clarification using the non-rectangular hyperbola model of photosynthesis. Ann. Bot. 82, 883–892. Chazdon, R.L., Fetcher, N., 1984. Photosynthetic light environments in a lowland tropical rain forest in Costa Rica. J. Ecol. 72, 553–564. Chazdon, R.L., Field, C.B., 1987. Photographic estimation of photosynthetically active radiation: evaluation of a computerized technique. Oecologia 73 (2), 525–532. Chazdon, R.L., Williams, K., Field, C.B., 1988. Interactions between crown structure and light environment in five rain forest Piper species. Am. J. Bot. 75 (10), 1459–1471. Chazdon, R.L., Pearcy, R.W., 1991. The importance of sunflecks for forest understory plants. Bioscience 41 (11), 760–766. Chen, H.Y.H., Klinka, K., 1997. Light availability and photosynthesis of Pseudotsuga menziesii seedlings grown in the open and in the forest understory. Tree Physiol. 17, 23–29. Ehleringer, J., Bjorkman, O., 1977. Quantum yields for CO2 uptake in C3 and C4 plants: dependence on temperature, CO2 , and O2 concentration. Plant Physiol. 59 (1), 86–90. Farquhar, G.D., Von Caemmerer, S., Berry, J.A., 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149, 78–90. Gross, L.J., 1982. Photosynthetic dynamics in varying light environments: a model and its application to whole leaf carbon gain. Ecology 63 (1), 84–93. Hikosaka, K., Sudoh, S., Hirose, T., 1999. Light acquisition and use by individuals competing in a dense stand of an annual herb, Xanthium canadense. Oecologia 118 (3), 388–396. Hikosaka, K., Ishikawa, K., Borjigidai, A., Muller, O., Onoda, Y., 2006. Temperature acclimation of photosynthesis: mechanisms involved in the changes in temperature dependence of photosynthetic rate. J. Exp. Bot. 57 (2), 291–302. Hilbert, D.W., 1990. Optimization of plant root: shoot ratios and internal nitrogen concentration. Ann. Bot. 66 (1), 91–99. Hirose, T., Werger, M.J.A., 1987. Nitrogen use efficiency in instantaneous and daily photosynthesis of leaves in the canopy of a Solidago altissima stand. Physiol. Plantarum 70 (2), 215–222. Ida, T.Y., Kudo, G., 2008. Timing of canopy closure influences carbon translocation and seed production of an Understorey Herb, Trillium apetalon (Trilliaceae). Ann. Bot. 101, 435–446. Kobe, R.K., Hogarth, L.J., 2007. Evaluation of irradiance metrics with respect to predicting sapling growth. Can. J. Forest Res. 37, 1203–1213. Koike, T., 1988. Leaf structure and photosynthetic performance as related to the forest succession of deciduous broad-leaved trees. Plant Species Biol. 3 (2), 77–87. Koyama, K., Kikuzawa, K., 2010. Geometrical similarity analysis of photosynthetic light response curves, light saturation and light use efficiency. Oecologia 164, 53–63. Miyazawa, Y., Kikuzawa, K., 2005. Winter photosynthesis by saplings of evergreen broad-leaved trees in a deciduous temperate forest. New Phytol. 165 (3), 857–866. Muller, O., Hikosaka, K., Hirose, T., 2005. Seasonal changes in light and temperature affect the balance between light harvesting and light utilisation components of photosynthesis in an evergreen understory shrub. Oecologia 143, 501–508. Muraoka, H., Hirota, H., Matsumoto, J., Nishimura, S., Tang, Y., Koizumi, H., Washitani, I., 2001. On the convertibility of different microsite light availability indices, relative illuminance and relative photon flux density. Funct. Ecol. 15, 798–803. Ogren, E., Evans, J.R., 1993. Photosynthetic light-response curves: 1. The influence of CO2 particle pressure and leaf inversion. Planta 189 (2), 182–190. Oguchi, R., Hikosaka, K., Hiura, T., Hirose, T., 2008. Costs and benefits of photosynthetic light acclimation by tree seedlings in response to gap formation. Oecologia 155 (4), 665–675. Osone, Y., Tateno, M., 2003. Effects of stem fraction on the optimization of biomass allocation and maximum photosynthetic capacity. Funct. Ecol. 17 (5), 627–636. Osone, Y., Tateno, M., 2005. Nitrogen absorption by roots as a cause of interspecific variations in leaf nitrogen concentration and photosynthetic capacity. Funct. Ecol. 19, 460–470. Pearcy, R.W., 1983. The light environment and growth of C3 and C4 tree species in the understory of a Hawaiian forest. Oecologia 58 (1), 19–25. Pearcy, R.W., 1988. Photosynthetic utilisation of lightflecks by understory plants. Aust. J. Plant Physiol. 15, 223–238. Pearcy, R.R., Hiroyuki Muraoka, H., Valladares, F., 2005. Crown architecture in sun and shade environments: assessing function and trade-offs with a threedimensional simulation model. New Phytol. 166, 791–800. Pfitsch, W.A., Pearcy, R.W., 1989. Daily carbon gain by Adenocaulon bicolor, a redwood forest understory herb, in relation to its light environment. Oecologia 80, 465–470. Reich, P.B., Walters, M.B., Kloeppel, B.D., Ellsworth, D.S., 1995. Different photosynthesis-nitrogen relations in deciduous hardwood and evergreen coniferous tree species. Oecologia 104 (1), 24–30. Reich, P.B., Tjoelker, M.G., Walter, M.B., Vanderklein, D.W., Buschena, C., 1998. Close association of RGR, leaf and root morphology, seed mass and shade tolerance in seedlings of nine boreal tree species grown in high and low light. Funct. Ecol. 12, 327–338. Takenaka, A., 1989. Optimal leaf photosynthetic capacity in terms of utilizing a natural light environment. J. Theor. Biol. 139, 517–529.

10

A. Miyashita et al. / Agricultural and Forest Meteorology 152 (2012) 1–10

Tang, Y., Kachi, N., Furukawa, A., Awang, M.B., 1999. Heterogeneity of light availability and its effects on simulated carbon gain of tree leaves in a small gap and the understory in a tropical rain forest. Biotropica 31 (2), 268–278. Thornley, J.H.M., 1976. Mathematical Models in Plant Physiology. Academic Press, London, pp. 92–110. Veneklaas, E.J., Poorter, L., 1998. Growth and carbon partitioning of tropical tree seedlings in contrasting light environments. In: Lambers, H., Poorter,

H., Van Vuuren, M. (Eds.), Inherent Variation in Plant Growth: Physiological Mechanisms and Ecological Consequences. Backhuys Leiden, The Netherlands, pp. 337–362. Washitani, I., 1992. Leaf carbon gain simulation for tree seedlings acclimatized to microsite light regimes of a temperate pine forest. Ecol. Res. 7, 187–191.