Introduction

Forests have experienced unequivocal climate warming in recent decades (IPCC 2013). Rising temperatures contribute to an increase in surface evapotranspiration; therefore, a change in the surface water balance is predicted (Knapp et al. 2015). Trees are considered vulnerable to drastic changes in soil moisture conditions because their roots cannot move freely. Although various forest disturbance factors (e.g., beetle attacks) could be involved in tree decline or mortality, drought related to global warming has been proposed as a major trigger (Allen et al. 2010; Smith et al. 2015). Furthermore, when a drought period is long, intense, or frequently occurring, it can lead to tree mortality (Brzostek et al. 2014; Cailleret et al. 2014).

In Korea, Korean fir (Abies koreana) in sub-alpine forests have suffered severe decline from the early 2000s. Abies koreana is categorized as an endangered species by the International Union for Conservation of Nature (IUCN) (Kim et al. 2011) and designated as a climate-sensitive biological indicator species (CBIS) in Korea (Lee et al. 2010). The most likely cause of A. koreana decline is related to the drought-induced by warming in early spring (Lim et al. 2007; Woo et al. 2008). However, knowledge regarding the physiological processes that lead to A. koreana mortality is limited. Therefore, intensive studies on physiological responses to environmental changes in A. koreana are required to understand and predict their responses to climate change, as well as for restoration.

The rapid decline of A. koreana has led to new environmental conditions in which the trees are exposed to more light. Therefore, it is necessary to examine the effect of light conditions on the physiological traits to determine the causes responsible for the increase in the decline rate of A. koreana. Light absorption and utilization are the first phases of photosynthesis, which are directly affected by changes in the light environment (Bjorkman 1981; Pearcy and Sims 1994). Consequently, the structural and functional properties of leaves are adapted to an efficient use light for tree survival and growth (Nobel 2009; Terashima et al. 2011; Vogelmann and Gorton 2014; Kwon and Woo 2016; Valladares et al. 2016). These traits respond sensitively to different light conditions depending on species-specific traits such as extent of shade tolerance and sun/shade dependence (Murchie and Horton 1997; Valladares et al. 2000; Fini et al. 2014; Sancho-Knapik et al. 2014).

Changes in leaf traits to adapt to varying light conditions could affect drought sensitivity. For example, Jiménez et al. (2009) reported that Quercus suber trees grown in moderate shade (15% light) have a higher tolerance to drought conditions than those grown in deep shade (1% light). Fraxinus ornus trees display higher dissipation of excess radiant energy (non-photochemical quenching) under high light irradiance and overcome oxidative damage to a greater extent compared to the trees grown in shade under drought (Fini et al. 2014).

Chlorophyll a fluorescence transient measurement (OJIP test) has been reported to be a sensitive and reliable method to detect the changes in the photosynthetic apparatus (Schansker et al. 2005; Shao et al. 2010; Wang et al. 2012). The OJIP fluorescence transient reflects the state of QA, QB, and PQ pool. Therefore, it is possible to provide detailed information on the functional traits of the photosynthetic apparatus such as absorption and trapping photons, dissipation of excess energy, electron transport, and level of photosynthetic performance (Strasser et al. 2004; Lepeduš et al. 2011; Stirbet 2011). Drought conditions are known to reduce the ability of absorption and trapping photons, electron transport by inactivated and/or damaged reaction centers and increases the quenching reaction center to dissipate of excess energy (van Rensburg et al. 1996; Gallé et al. 2007; Albert et al. 2011; Wang et al. 2012). However, the interactive effect of light and drought on the photosynthetic apparatus has not been well-addressed. In particular, drought responses of A. koreana growing under different light conditions have not been previously reported.

We examined the changes in functional traits of the photosynthetic apparatus using chlorophyll fluorescence traits to better understand the response of A. koreana to light and drought conditions. We also analyzed the changes in photosynthetic rate, and composition and amount of photopigments under different light regimes and water status in A. koreana seedlings, which are more sensitive to environmental stresses than adult plants. We hypothesized that the photosynthetic apparatus of A. koreana would change under different light conditions and affect the response to progressive drought and recovery.

Materials and methods

Plant material and experimental conditions

This experiment was conducted at the National Institute of Forest Science (NIFoS), Seoul, Korea (37°35′N, 127°2′W) from April 4 to August 26, 2014. Three-year-old seedlings of A. koreana were planted in 2.5-L pots (one seedling per pot) containing forest soil. After 2 weeks, the seedlings were exposed to artificial shade levels produced through the use of different shading nets, fixed in wireframe facilities with a transparent roof as a rain shelter to exclude the effects of rain. These nets allowed for the 35% shading (1,000–1,400 µmol m−2 s−1, min–max) and 75% shading (200–400 µmol m−2 s−1) of full sunlight (1,500–1,900 µmol m−2 s−1). The photosynthetic photon flux density (PPFD) was converted from light intensity using the HOBO pendant temperature/light data logger (UA-002-64, Onset, MA, the USA) according to Wagner and McGraw (2013). Temperature and relative humidity readings were recorded for the shading treatments every 10 min using a HOBO Pro v2 logger (U23-002, Onset, MA, the USA). Mean air temperature and relative humidity for full sunlight, 35% shading, and 75% shading were 30.2 ± 2.1 °C and 61.4 ± 9.0%, 29.5 ± 2.8 °C and 60.5 ± 10.2%, and 30.4 ± 2.6 °C and 62.0 ± 11.1%, respectively. Light intensity, mean air temperature, and relative humidity were measured at the upper extremity of plants between 08:00 and 18:00 h on 24 clear days during the experimental period.

Watering treatments

The seedlings were watered daily to field capacity until the beginning of the drought experiment. On the first day of the experiment (July 10, 2014), half of the seedlings from each shading treatment were selected for well-watering treatment (W), and the other half were no-watering treatment (NW) subjected to a progressive drought. When one of the light treatments reached a mean photosynthetic rate (net CO2 gas exchange rate on leaf dry weight basis) of 0 µmol g−1 s−1 or less due to the drought, all the NW treatments were re-watered. During the experiment, soil volumetric water content (SVWC, %) was measured daily twice (Field Scout TDR 100, Spectrum Technologies, Inc., Plainfield, IL, the USA). The SVWC in W was maintained between 12 and 20% during the experiment. However, the SVWC in the NW treatment drastically decreased until 15 days after no-watering and was maintained below 3% for 1 week until re-watering occurred (Fig. 1).

Fig. 1
figure 1

The average volumetric soil moisture contents in each treatment. Black dash and grey dash lines indicate well-watering (W) and no-watering (NW), respectively

Full size image

Photosynthetic rate measurements

Measurements of photosynthetic rate (PN) were conducted weekly from the start of the watering treatment. Recently expanded terminal leaves were chosen and repetitively measured using a portable photosynthesis systems equipped with light-emitting diode (LED) light sources and the standard 2 × 3 cm chambers (Li-6400XT and Li-6400-02B LED, Li-Cor Inc., Lincoln, the USA) during the experimental period. During measurements, CO2 concentration, temperature in the chamber, flow rate of air, and PPFD were maintained at 400 μmol mol−1, 27 °C, 500 μmol s−1, and 1,000 μmol m−2 s−1, respectively. After the experiment, measured needles were removed from the trees and dried and weighed for calculation of the PN using needle dry weight according to von Caemmerer and Farquhar (1981).

Photopigment analysis

Photopigments were measured after no-watering for 21 days. The fresh needle disk (0.1 g) was immersed in 10 ml of dimethyl sulfoxide (DMSO) (99%) for 6.5 h at 65 °C. The absorbance of its extracts was measured using a spectrophotometer (DU730 UV VIS Beckman Coulter, the USA) at wavelengths of 470, 645, and 663 nm for chlorophyll a (Chla), chlorophyll b (Chlb), carotenoid (Car.), and xanthophyll (Xan.) contents, respectively. Chlorophyll and accessory pigment contents were calculated according to Wellburn and Lichtenthaler (1984).

Chlorophyll a fluorescence transient analysis

Measurements of chlorophyll a fluorescence transient (OJIP test) were performed at the same time of photosynthetic rate measurements. Chlorophyll a fluorescence transient was measured using a Pocket-PEA fluorimeter (Hansatech Instruments Ltd., the UK) on plants that were dark adapted for at least 30 min. A total of 5–8 seedlings per treatment were used. The Pocket-PEA fluorimeter was at O (50 µs), J (2 ms), and I (30 ms) as the intermediate stage, and P (300 ms) as the peak. The obtained data were used in the OJIP test (Strasser et al. 2004) to calculate the following parameters of photosystem II (PSII) photochemistry (Table 1). Data are presented as spider plots, indicating the mean of the relative change of the selected parameters in relation to their corresponding well-watering values. ∆FIP (fluorescence amplitude of the I to P phases) was calculated using the difference in P and I fluorimetry (Ceppi et al. 2012).

Table 1 Definitions and explanations of selected OJIP test parameters
Full size table

Statistical analysis

All data were subjected to one-way analysis of variance (ANOVA) using SPSS 13.0 (SPSS Inc., Chicago, IL, the USA) to determine the difference between treatments. Duncan’s multiple range tests were used to detect significant differences between means at a significance level of P < 0.05.

Results

Photosynthetic rate

Although PN of all light treatments showed a decreasing tendency due to drought, PN of 75% shading showed no difference between W and NW until the 19th day of no-watering. The first time that the photosynthetic rate reached 0 µmol g−1 s−1 was on the 21st day of no-watering under full sunlight. Moreover, no recovery of PN was observed under full sunlight until the 7th day after re-watering, whereas PN was recovered on the 4th day after re-watering for other light treatments (35 and 75% shading) (Fig. 2).

Fig. 2
figure 2

Changes in the photosynthetic rate (PN) in the needles of Abies koreana seedlings grown in different light (full sunlight: a; 35% shading: b; 75% shading: c) and water treatments (W: well-watering; NW: no-watering). Values are mean and error bars represent standard deviation from the mean (n = 7). The single asterisk indicates significant difference at P < 0.05 by ANOVA (DUNCAN)

Full size image

Photopigment characteristics

ChlT and Car. + Xan. were significantly increased with an increase in shading under W. The same patterns were also observed with NW. Decrease tendency in Chla/Chlt and increase tendency in Chlb/Chlt also showed an increase in shading under W. However, these tendencies were not observed under light treatment combined with NW. A significant difference between W and NW for Chla/Chlb was observed only under full sunlight (Table 2).

Table 2 Photopigment content and quality in the needles of Abies koreana seedlings grown in different light and water treatments (W: well-watering; NW: no-watering) at the end of drought (21st day of no-watering)
Full size table

Chlorophyll a fluorescence transient traits (OJIP test)

The OJIP curves reflect the major changes in photosynthetic electron transport. NW showed lower fluorescence values in the OJIP curves than W especially in phases I and P. This indicates that drought led to inhibition of electron transport from primary (QA) to secondary quinone acceptor (QB). These differences in the IP phase (∆FIP) between W and NW in 75% shading were the smallest among light treatments (full sunlight and 35% shading) (Fig. 3).

Fig. 3
figure 3

Effects on the kinetics of the OJIP transients and fluorescence amplitude of the I to P phase (∆FIP) measured in the needles of Abies koreana seedlings grown in different light (full sunlight: a; 35% shading: b; 75% shading: c) and water treatments (W: well-watering; NW: no-watering) at the end of drought (21st day of no-watering). Black circles and bars indicate W, and white circles and bars indicate NW. Values are mean (n = 5–7) and the single asterisk indicates a significant difference at P < 0.05 by ANOVA (DUNCAN)

Full size image

The patterns of spider-plot presentation in chlorophyll a fluorescence transient parameters showed similar shapes under full sunlight and 35% shading by water treatment. However, those values were different under 75% shading (Fig. 4). Similarly, under full sunlight and 35% shading, the reductions in yield (ΦP0, Ψ0, and ΦE0), area (Sm and N), performance index (PIabs), and electron transport flux further than QA (ET0/RC) and the increases in F dynamic (F0, VJ, and dV/dt0) and parts of flux per reaction center (RC) (ABS/RC, DI0/RC, TR0/RC) were significant in NW on the 21st day of no-watering (Fig. 4a). Under 75% shading, only the reduction in PIabs caused by NW was similar to the other light treatments (full sunlight and 35% shading) (Fig. 4a). On the 11th day after re-watering, decreases in ΦE0 and PIabs and increases in VJ and parts of flux per RC (ABS/RC, DI0/RC, TR0/RC) were observed under full sunlight and NW (Fig. 4b). At the same time, an increase in DI0/RC was observed only under 35% shading, and there were no significant differences between W and NW in the case of 75% shading (Fig. 4b). On the 13th day after re-watering, there were no significant differences in any of the light treatments (Fig. 4c). Apparent recovery by re-watering was observed on the 19th day after re-watering, as indicated by the same spider-plot pattern between W and NW (Fig. 4d).

Fig. 4
figure 4

Spider-plot of OJIP parameters in the needles of Abies koreana seedlings grown in different light (full sunlight; 35% shading; 75% shading) and water treatments (W: well-watering; NW: no-watering) on the 21st (a) day of no-watering and the 11th (b), 13th (c), and 19th (d) days after re-watering. Values are mean (n = 5–7) and the single asterisk indicates a significant difference at P < 0.05 by ANOVA (DUNCAN)

Full size image

Discussion

Effects of different light conditions on the photosynthetic rate under drought and re-watering

PN decreased with an increase in the water-withholding treatment period. However, the pattern change was different depending on the light treatment. PN in 75% shading appeared to have no significant differences between W and NW until the 19th day of water withholding whereas those under full sunlight and 35% shading showed significant decline at that time (Fig. 2). Drought usually induces stomatal closure to preserve the water in the plant which contributes to the initial decrease in PN by restricting CO2 input into the leaf (Rodriguez‐Dominguez et al. 2016). The decrease in PN could induce a decreased demand for adenosine triphosphate (ATP) and nicotinamide adenine dinucleotide phosphate (NADPH) in the Calvin-Benson cycle and led to a redox back pressure on PSII (Medrano et al. 2002). Under continuing light-driven electron transport, imbalances between electron use and energy consumption cause the generation of reactive oxygen species (ROS) directly and/or indirectly (Asada 2006). If ROS cannot be scavenged sufficiently, the photosynthetic apparatus can be damaged due to oxidative stress (e.g., destruction of cell structure and components) (Asada 1999; Flexas and Medrano 2002). Therefore, as the extent and period of water stress increase, PN reduces due to the impaired photosynthetic apparatus (Flexas et al. 2006; Xu et al. 2010).

In the present study, reductions in the amount and quality of photopigment and the low capability of light capture and utilization implied that damages were preceded by drought especially under full sunlight (Table 2 and Figs. 3, 4). In A. koreana, full sunlight and 35% shading could likely be the conditions that imposed excessive light energy input. Thus, photoinhibition may have caused a decrease in PN (Powles 1984; Alves and Magalhães 2002). The excess photon energy received by chlorophyll molecules are utilized during photosynthesis or dissipated by non-photochemical quenching (NPQ). However, if the absorption of light exceeds the utilization during photochemical conversion, it is exacerbated the photoinactivation of PSII, which is damaged or incorrectly assembled chlorophyll protein complexes (Santabarbara et al. 2002; Oguchi et al. 2011). Unlike 35% shading and full sunlight treatment, the 75% shading treatment delayed the time when PN reduced to 0 µmol g−1 s−1 due to drought and showed rapid recovery after re-watering (Fig. 2). These results indicate that the effect of drought on photosynthesis in A. koreana depends on light conditions.

Changes in photopigment under different light and water conditions

Both ChlT and accessory pigments (Car. + Xan.) in A. koreana significantly increased under shade treatments (Table 2) which indicated an adaptive response to low light conditions. A higher level of ChlT and Car. + Xan. per leaf mass unit in shade treatments were similar to those previously described in Abies alba (Lichtenthaler et al. 2007), another coniferous tree (Wyka et al. 2008; Barsi et al. 2009), and other broadleaf species (Lichtenthaler et al. 2007; Sarijeva et al. 2007). Accessory pigments are known to be potent quenchers of ROS (Asada 1999). When ROS in trees increase under unhealthy environmental conditions including drought, it plays an essential role in protecting the photosynthetic apparatus (Foyer and Shigeoka 2011; Jahns and Holzwarth 2012). Thus, under severe drought conditions, relatively lower accessory pigment contents under full sunlight compared to shade conditions (35 and 75% shading) would not be sufficient to scavenge the ROS increase. These conditions could have resulted in the low content of photopigment under combined treatment of full sunlight and NW.

A decrease in Chla/Chlb with increasing chlorophyll b (Chlb/ChlT) was also observed in W subjected to shade treatments which implied an enhancement of the efficiency of blue light absorption under reduced light conditions (Yamazaki et al. 2005). Shade tolerant species have a higher photosynthetic capacity and can maintain their growth rates under shading conditions compared to intolerant species (Schrader et al. 2006; Valladares and Niinemets 2006). The observed responses can be assigned to the adaptive ability of A. koreana to cope with shade as shade tolerant species, and these responses led to a reduction in the non-stomatal limitation in terms of the efficiency of light capture and therefore, affected the photosynthetic rate in NW.

Changes in chlorophyll a fluorescence transient parameters under different light and water conditions

All treatments exhibited the typical polyphasic OJIP increases. However, OJIP curves indicated a reduction in shape due to drought especially in the IP phase (Fig. 3). The IP phase depends not only on the transient block on the acceptor side of photosystem I (PSI) but also on the flow of the electrons originating in PSII through PSI (Schansker et al. 2005). Therefore, IP phase amplitude can be used as an indicator of PSI content and activity (Ceppi et al. 2012). We calculated fluorescence amplitude of the I to P rise (∆FIP) to analyze elaborate differences in the IP phase. Interestingly, ∆FIP was not affected by drought under 35 and 75% shading. ∆FIP of full sunlight was low with both water treatments. Therefore, it can be concluded that PSI was less damaged even under drought due to the combination of shade conditions.

The changes in fluorescence parameters from the OJIP curves varied greatly between W and NW with the increase in light intensity under NW (Fig. 4). In the case of specific fluxes per active PSII reaction center referring to its function, the parameters describing absorption (ABS/RC), trapping (TR0/RC), and dissipation of excess energy (Di0/RC) were higher. Moreover, the electron transport beyond Q −A (ET0/RC) was lower in NW than those of W under full sunlight and 35% shading on the last day of no-watering (Fig. 4). The increase in functional antenna size (high ABS/RC) indicated the presence of a non-QA reducing reaction center. Strasser et al. (2004) suggested a possibility that non-QA reducing reaction centers act as heat sinks. Similarly, we also showed the increase in Di0/RC by drought under full sunlight and 35% shading in our study. Furthermore, the capability for electron transport beyond Q −A was impaired (low ET0/RC) which indicated that electron transfer was reduced from Q −A to QB. The slope of the induction curve from F0 to FM (dV/dt0) also increased, and implied inhibition of reoxidation of Q −A to QB by drought. These results were related to the accumulation of reduced primary plastoquinone (Q −A ) and inactivation of RC which were confirmed by the increases in Vj, N, and Sm by drought under full sunlight and 35% shading.

The presence of non-QA reducing centers and a low electron transportation rate from QA to QB which are considered to be downregulation mechanisms dissipated the excess absorbed light in a controlled way. These mechanisms could protect the photosynthetic apparatus from photo-oxidation (Li et al. 2015). Matsubara and Chow (2004) and Vass and Cser (2009) also suggested that inactivation of RC is related to quenching reaction in terms of photoprotection, and prevents the photosynthetic apparatus from irreparable damage. This strategy could allow resilience to drought stress after re-watering. In consistent with our results, the photosynthetic apparatus was less damaged by drought, and it was rapidly restored by re-watering in drought-tolerant species (Gallé et al. 2007; Flexas et al. 2009).

The maximum quantum yield of primary photochemistry (ΦP0) and photosynthetic performance index (PIabs) are often used to gauge the occurrence and extent of changes in physiological responses in plants (Bussotti et al. 2007; Lepeduš et al. 2011; Wang et al. 2012). Many studies have reported a positive correlation between ΦP0 or PIabs and drought stress (Shao et al. 2010; Rathod et al. 2011). However, ΦP0 occasionally only responds to severe stress conditions (Oukarroum et al. 2007; Wang et al. 2012) whereas PIabs is more sensitive to stress (Bussotti et al. 2007; Wang et al. 2012). In our study, PIabs was one of the indicators of drought effect on A. koreana under all light conditions. All fluorescence parameters indicated that the photosynthetic apparatus of A. koreana recovered from drought from the 13th day after re-watering. This recovery was also observed in the photosynthetic rate (Figs. 2, 4c). Thus, we can conclude that full sunlight conditions aggravate the damage to the photosynthetic apparatus during drought periods, although a complete recovery from water stress can be achieved after re-watering regardless of light conditions.

Conclusions

The photosynthetic apparatus of three-year-old seedlings of A. koreana grown under different light conditions indicated different responses to drought. Under full sunlight, a more impaired photosynthetic apparatus was observed during drought periods. Although the ability for light absorption and utilization and PN in all light treatments was recovered after re-watering, A. koreana under full sunlight exhibited the lowest photopigment content and quality, PSII activities, and PN compared to those under other light conditions during drought. Taken together, these results indicate that seedlings of A. koreana grown under high light intensity could be more vulnerable to drought events, and shading could decrease the sensitivity to drought by alleviating water stress.