Introduction
The growing popularity of gardening as both an activity and a lifestyle has broadened the scope of the gardening industry, prompting policy initiatives to promote gardening culture and support industry development through enhanced garden infrastructure (2nd Garden Promotion Basic Plan (2021~2025), 2021). Moreover, rising national income levels have increased public interest in garden plants and invigorated urban gardening culture (Choi et al., 2017). As green spaces that can be established even in limited areas, urban gardens contribute to urban biodiversity and play a role in addressing various environmental challenges (Mo, 2021).
Over 70% of Korea’s annual precipitation occurs between June and August, resulting in a prolonged dry period from October through May. Urban areas experience chronic dryness due to the prevalence of impervious surfaces, such as asphalt, as well as energy consumption from vehicles and heating systems (Kim et al., 2014). Moreover, abiotic stress on plants is expected to intensify as climate change accelerates the frequency and severity of extreme weather events. Reduced precipitation, which leads to soil moisture deficiency, is the most critical environmental factor limiting plant growth (Araus, 2002; Fedoroff et al., 2010). Under optimal growth conditions, plants can improve light energy use efficiency and enhance productivity through photosynthesis. However, exposure to abiotic stresses—such as excessive light, temperature extremes, or deficiencies in water and nutrients—can cause significant damage to the photosynthetic system (Pereira, 2016; Zhu, 2016). Water deficiency, in particular, is widely recognized as a key environmental factor limiting photosynthetic activity. It disrupts turgor pressure, thereby inhibiting leaf expansion and stem elongation, and often induces stomatal closure or leaf abscission to minimize water loss (Kim et al., 2020; Lee et al., 2014). Furthermore, plants exposed to drought stress are known to accumulate reactive oxygen species (ROS), which trigger the activation of various antioxidant defense mechanisms. These include increased activities of enzymes such as superoxide dismutase (SOD), catalase (CAT), and ascorbate peroxidase (APX), as well as elevated proline accumulation to mitigate oxidative stress (Foyer et al., 1994; Heuer, 2010; Mittler, 2002). However, under severe drought conditions, photosynthesis is significantly inhibited, osmotic regulation is disrupted, and major metabolic dysfunctions may occur, ultimately leading to substantial physiological damage (Frary, 2015). As such, drought stress inhibits photosynthetic activity by inducing stomatal closure, reducing water use efficiency, and causing imbalances in photochemical reactions and electron transport. Consequently, chlorophyll fluorescence analysis (or chlorophyll fluorescence transients) can be employed as an effective tool for diagnosing stress responses in crops (Strasser et al., 2000).
Chlorophyll fluorescence is a phenomenon observed when chlorophyll molecules are exposed to light of specific wavelengths and intensities. After a period of dark adaptation—typically ranging from 15 minutes to 1 hour—excited chlorophyll molecules return to their ground state (Kautsky and Hirsch, 1931; Strasser, 1978). This process is characterized by the OJIP fluorescence transient, also known as the Kautsky effect, OJIP curve, fluorescence induction or fluorescence decay. In this transient, the O step (origin) represents the initial fluorescence level immediately after dark adaptation; the J step (jump) corresponds to the first inflection step where minimal fluorescence change occurs; the I step (intermediate) indicates an intermediate fluorescence level; and the P step (peak) reflects the maximum fluorescence intensity (Govindje, 1995). Furthermore, the JIP-test—a set of fluorescence parameters that quantitatively analyzes the Kautsky effect —is widely employed to investigate the relationship between chlorophyll fluorescence parameters and abiotic stress conditions (Stirbet and Govindjee, 2011).
Ham et al. (2018) reported that drought stress in nursery seedlings of tomato and cucumber can be diagnosed using six chlorophyll fluorescence parameters—Fv/Fm, DIo/RC, ET2o/RC, RE1o/RC, PI_ABS, and PI_total ABS—through chlorophyll fluorescence analysis (Ham et al., 2018). Kim et al. (2020) assessed water stress in Cnidium officinale Makino using Fv/Fm, Rfd_Lss, and NPQ_Lss, while Yoo et al. (2013) analyzed the photochemical responses of red pepper (Capsicum annuum L.) via the OJIP test, introducing PI_ABS and the drought factor index (DFI) (Kim et al., 2020; Yoo et al., 2013). Additionally, Liu et al. (2018) investigated the response of the photosynthetic electron transport chain (ETC) in maize under varying levels of drought stress, revealing that under moderate drought stress, the photochemical activity of photosystem II (PSII) declined while photosystem I (PSI) remained unaffected; however, under severe drought stress, the entire ETC was inhibited, indicating that PSII is more sensitive to drought stress than PSI (Liu et al., 2018).
Therefore, this study aimed to analyze chlorophyll fluorescence responses in ten garden plant species to identify universal photophysiological indicators applicable to drought stress assessment across species.
Research Methods
Experimental materials and drought stress treatment
This study was conducted from July to August 2023 in a greenhouse at the Hankyung National University Affiliated Farm, located in Anseong, Gyeonggi Province, South Korea (37°00′42″N, 127°19′12″E). Ten species of garden plants were obtained from The K Flower (Yongin-si, Gyeonggi-do) and planted in experimental pots (143 × 143 × 160 mm): Polygonatum odoratum var. pluriflorum (Miq.) Ohwi, Carex maculata Boott, Pachysandra terminalis Siebold and Zucc., Hosta longipes (Franch. and Sav.) Matsum., Lilium lancifolium Thunb., Allium dumebuchum H.J. Choi, Iris sanguinea Donn ex Hornem., Sedum take-simense Nakai, Aster spathulifolius Maxim., and Mukdenia rossii (Oliv.) Koidz. The pots were filled with silty loam soil. To minimize the influence of increased soil moisture through ground contact, all pots were placed in triplicate on pallets (Fig. 1A). Angled drip spikes were also installed in each pot to deliver a consistent daily amount of water via drip irrigation. After a 15-day acclimatization period, the plants were cultivated under normal watering conditions for 5 days, which served as the control period (Control). Following this, irrigation was suspended to induce drought stress (Fig. 1B). Drought severity was evaluated in four stages ( DS 1 t o DS 4 ) at 5 -day intervals. Throughout the experiment, ambient temperature was recorded hourly using a data logger (testo 174 H, Testo Saveris GmbH, Baden-Württemberg, Germany) to monitor environmental conditions. Additionally, four pots were randomly selected, and their soil moisture content was measured using a WaterScout SM 100 soil moisture sensor (Spectrum Technologies, Inc., Aurora, IL 60504, USA). A data logger (WatchDog 1000 Series Watermark Irrigation Stations, Spectrum Technologies, Inc., Aurora, IL 60504, USA) was used to record the data. Soil moisture content was determined using the calibration formula embedded in the soil moisture sensor.
Chlorophyll fluorescence measurement
Chlorophyll fluorescence was measured on fully expanded leaves (the upper third and fourth leaves) using a chlorophyll fluorescence meter (FP-100, Photon System Instruments, Czech Republic). Three plants were measured in duplicate for each treatment, and measurements were conducted across five drought stress stages (Control, DS 1, DS 2, DS 3, and DS 4). Prior to measurement, the selected leaves were dark-adapted for 30 minutes using leaf clips to ensure that all PSII reaction centers were open and that energy had returned to the ground state. Fluorescence measurements were then performed in OJIP mode, targeting a 4 mm² leaf area with saturating light at an intensity of 2,400 μmol m−2 s−1 (Ayyaz et al., 2020).
The minimal fluorescence (Fo) is measured at 50 μs, when all PSII reaction centers are open. Then, the fluorescence passes through the J step (2 ms) and the I step (30 ms), reaching the P step within 1 second, during which all PSII reaction centers are closed and the maximum fluorescence (Fm) is attained. This sequence is well-documented in chlorophyll fluorescence kinetics studies of photosystem II (Strasser et al., 2000). In this study, chlorophyll fluorescence data for each species were comparatively analyzed across drought stress stages based on their respective O-J-I-P responses. Chlorophyll fluorescence parameters were derived using the JIP-test (Stirbet and Govindjee, 2011; Table 1).
Statistical analysis
All statistical analyses in this study were performed using the R software (R Core Team, 2021). Chlorophyll fluorescence parameters estimated at each drought stress stage of garden plants were analyzed using Pearson correlation analysis and principal component analysis (PCA).
Variable importance assessment
Based on the above analyses, an evaluation of variable importance was conducted to identify suitable photophysiological indicators for effectively assessing the drought stress responses of garden plants. Using 546 data points with missing values excluded, a multiple linear regression model was constructed. To minimize the influence of differences in measurement units among chlorophyll fluorescence parameters, z-score normalization was applied. The explanatory power of the model was assessed using the coefficient of determination (R2) and the adjusted coefficient of determination (Adjusted R2). Predictive accuracy was evaluated using the root mean square error (RMSE) and mean absolute error (MAE) (Equations 1–4). In addition, the model’s performance was compared with that of a baseline model based on the simple mean, to verify its predictive superiority. To address potential multi-collinearity among predictor variables, a relative weights analysis was conducted following the method proposed by Johnson (2000). Through this procedure, key photo-physiological indicators suitable for assessing drought stress in garden plants were derived.
(yi: observed data, ŷl: predicted data, ȳ: mean of observed data, N: total number of samples, p: number of independent variables)
Results and Discussion
Environmental monitoring
Environmental monitoring was performed at one-hour intervals to measure air temperature and soil moisture content inside the greenhouse (Fig. 2). During the cultivation period, the daily mean air temperature ranged from a minimum of 22.3 °C at 28 days after treatment (DAT) to a maximum of 39.9 °C at 6 DAT, with an overall average of 31.6 °C. The daily mean soil moisture content under normal cultivation conditions (Control) was 26.9%. However, it declined sharply immediately after the onset of the drought stress treatment (DS 1).
Chlorophyll fluorescence (OJIP) analysis across drought stress stages
In general, each phase of the chlorophyll fluorescence transient reflects specific processes occurring in the PSII reaction center: the O-J phase corresponds to the reduction of quinone A (QA), the J-I phase indicates the reduction and reoxidation of QA, and the I-P phase represents the rapid reduction of the plastoquinone (PQ) pool (Stirbet et al., 1998). A comparative analysis of chlorophyll fluorescence in garden plants across different drought stress stages revealed that, in all species, Control exhibited the highest fluorescence intensity up to the maximum fluorescence (Fp), indicating efficient electron transport from PSII to PSI. After the onset of drought stress, Fp gradually declined ( Fig. 3 ). At the DS 1 s tage, Fp decreased by a minimum of 6.9% in Hosta longipes (Franch. and Sav.) Matsum and a maximum of 29.4% in Carex maculata Boott. At the DS 2 stage, the reduction ranged from 25.2% in Iris sanguinea Donn ex Hornem to 51.8% in Sedum takesimense Nakai. At the DS 3 stage, Fp had decreased by a minimum of 52.0% in Mukdenia rossii (Oliv.) Koidz and a maximum of 88.5% in Carex maculata Boott. At the DS 4 stage, reductions ranged from 59.2% in Hosta longipes (Franch. and Sav.) Matsum to 95.1% in Carex maculata Boott. In this study, at the DS 4 stage, seven out of ten species exhibited more than a 70% reduction in Fp compared to Control, with the exceptions being Hosta longipes (Franch. and Sav.), Mukdenia rossii ( Oliv.) Koidz, and Lilium lancifolium Thunb. Notably, Carex maculata Boott showed a reduction exceeding 90%, indicating severe inhibition of electron transport. Under drought stress, plants commonly respond by closing their stomata to conserve water, which in turn limits CO2 availability. This limitation delays the reduction and reoxidation of QA, leads to inactivation of reaction centers (RCs), and impairs the overall function of PSII (Baker, 2008; Lee, 2018; Shanker et al., 2022; Yan et al., 2024). Additionally, the reduced rate of QA− reoxidation suppresses electron transport and inactivates the oxygen- evolving complex (OEC). These disruptions promote the accumulation of reactive oxygen species (ROS), further impairing the early stages of the photosynthetic electron transport chain. As a result, water-splitting activity ceases, and electron flow to the RC is interrupted (Asada, 1999; Cifre et al., 2005; Lidon and Henriques, 1993; Strasser et al., 2000; Strasser et al., 2004). Consequently, inhibition of electron acceptor activity under drought stress results in QA accumulation (O-J phase), suppression and stagnation of electron transfer from QA to quinone B (QB) (J-I phase), and limited electron transport from QB to the plastoquinone (PQ) pool (I-P phase) (Kalaji et al., 2017; Shu et al., 2024; Tóth et al., 2007). In this study, following drought stress treatment, Carex maculata Boott, Iris sanguinea Donn ex Hornem, and Aster spathulifolius Maxim exhibited a transient increase in chlorophyll fluorescence during the O-J phase at the DS 2 stage, which was subsequently suppressed. This suppression was accompanied by a marked decline in fluorescence during the J-I and I-P phases, resulting in chlorophyll fluorescence induction curves (OJIP curves) that progressively shifted toward a more linear profile. These results suggest that the initial rise in the O-J phase, followed by reductions in the J-I and I-P phases under drought conditions, reflects rapid photochemical damage to OEC and QA. This impairs the recovery mechanisms of the photosynthetic system, leading to a more gradual slope in the OJIP curves of these garden plants (Ham et al., 2018; Lee et al., 2014; Oukarroum et al., 2007).
Identification of key photophysiological indicators for drought stress assessment
The flow of energy in plants typically begins when light energy absorbed by chlorophyll is transferred through the antenna complex to the oxygen-evolving complex (OEC), ultimately exciting the reaction center (RC). Once excited, the RC releases an electron to pheophytin, which then reduces the primary electron acceptor, quinone A (QA). The electron is subsequently transferred to quinone B (QB), during which QA is re-oxidized and QB is reduced to plastoquinone (PQ) by accepting protons from the stroma. The PQ pool then mediates electron transfer to plastocyanin (PC) via the cytochrome b6f complex, undergoing a redox cycle. During this process, protons are translocated into the thylakoid lumen. These resulting proton gradient is subsequently utilized in the synthesis of adenosine triphosphate (ATP) and nicotinamide adenine dinucleotide phosphate (NADPH) (Frary, 2015; Johnson, 2025).
To identify key photophysiological indicators for assessing drought stress in garden plants, species-specific chlorophyll fluorescence parameters derived from the JIP-test) were calculated and subjected to Pearson correlation analysis against soil water content (Fig. 4). The analysis revealed that five parameters—the number of active reaction centers per cross-sectional area (RC/CS), which reflects changes in energy flux per unit leaf area; light energy absorption per cross-sectional area (ABS/CS); trapped excitation energy per cross-sectional area (TRo/CS); and two electron transport-related parameters (ET2o/CS and RE1o/CS)—were consistently and positively correlated with soil moisture across all species, with most correlation coefficients exceeding 0.8 (Eom et al., 2024; Oh and Koh, 2013; Shin, 2022). In addition, the maximum quantum yield of PSII (Fv/Fm) and the performance indices associated with energy conversion efficiency in the electron transport chain (PI_ABS and PI_total ABS) have previously been reported to decline under drought stress. Consistent with these findings, this study also observed significant correlations in seven or more plant species, indicating similar physiological trends (Guo et al., 2022; Jia et al., 2020; Rathod, 2011; Wang et al., 2025). In contrast, five parameters representing energy flux per PSII reaction center—including ABS/RC, DIo/RC, ET2o/RC, RE1o/RC, and TRo/RC—showed weak correlations with soil moisture, suggesting that they are less suitable as indicators for evaluating drought stress in garden plants.
A principal component analysis (PCA) was performed using 11 photophysiological parameters—Fv/Fm, RC/ABS, PI_ABS, PI_total ABS, DF_total ABS, RC/CS, ABS/CS, TRo/CS, ET2o/CS, DIo/CS, and RE1o/CS—to identify common indicators for assessing drought stress in ten garden plant species. The PCA results showed that PC1 (69.3%) and PC2 (16.4%) explained a total of 85.7% of the variance, and the distribution exhibited a trend of shifting from the negative to the positive direction along PC1 in response to soil moisture levels (Fig. 5). PC1 was characterized by strong negative loadings (< −0.25) for all parameters except DIo/CS, which instead exhibited the highest positive loading (0.65) on PC2. These findings indicate that all parameters included in the analysis are associated with drought stress responses in garden plants (Table 2).
Based on the results of the Pearson correlation analysis and PCA, a multiple linear regression model was developed to evaluate the relative importance of the variables. The model achieved an R2 of 0.8 or higher, with both RMSE and MAE reduced to less than half of those observed in the baseline model, indicating high predictive reliability (Table 3). However, multicollinearity was identified among certain independent variables. To minimize inter-variable influence and accurately determine the relative contribution of each, Johnson’s (2000) relative weights analysis was applied, and the results were visualized in Fig. 6. Among the eleven parameters, TRo/CS exhibited the greatest explanatory power for drought stress in garden plants, accounting for 15.1% of the total R2. This was followed by ABS/CS (14.8%), RC/CS (14.5%), ET2o/CS (13.5%), and RE1o/CS (11.9%). Collectively, these top five parameters explained over 70% of the total R2. These findings suggest that the five chlorophyll fluorescence parameters—TRo/CS, ABS/CS, RC/CS, ET2o/CS, and RE1o/CS—can serve as effective indicators for evaluating drought stress responses in garden plants. In contrast, DF_total ABS accounted for only 1.4% of the R2, while RC/ABS (4.5%) and PI_total ABS (4.7%) also demonstrated relatively low contributions.
In this study, five proposed chlorophyll fluorescence parameters were identified as potential indices for assessing drought stress in garden plants composed of various species. However, since these indicators are derived from mean values, certain parameter values may be biased, and species-specific characteristics could be partially obscured. Nevertheless, integrated analyses based on the mean values of extensive datasets collected from diverse species confirmed the potential to establish common indicators applicable across multiple species. Furthermore, as plants exhibit species-dependent variation in factors affecting soil moisture—such as leaf area, canopy coverage, and transpiration rate—future studies that incorporate species-specific sensitivity analyses or account for different planting conditions, while accounting for soil moisture variation, are expected to enhance the practical applicability of these indices as assessment tools.
Conclusion
This study was conducted from July to August 2023 in a greenhouse at the Hankyung National University Affiliated Farm in Anseong, Gyeonggi Province, South Korea, using ten garden plant species as experimental subjects. The plants during the control period (Control) were maintained under normal cultivation conditions via drip irrigation, after which irrigation was withheld to induce drought stress. The drought treatment was applied in five-day intervals, during which chlorophyll fluorescence responses were analyzed. The analysis showed that Control exhibited the highest values of maximum chlorophyll fluorescence, while these values progressively decreased with continued drought stress. To identify key photophysiological indicators for evaluating drought stress in garden plants, species-specific chlorophyll fluorescence parameters were calculated using the JIP-test. Pearson correlation analysis and Principal Component Analysis (PCA) were then performed with respect to soil moisture content. The results demonstrated that eleven parameters—Fv/Fm, RC/ABS, PI_ABS, PI_total ABS, DF_total ABS, RC/CS, ABS/CS, TRo/CS, ET2o/CS, DIo/CS, and RE1o/CS—were associated with drought stress responses in garden plants. Based on these findings, a multiple linear regression model was constructed to assess the relative importance of the variables, demonstrating high predictive reliability. However, multicollinearity was detected among certain independent variables. After excluding inter-variable influence and determining individual contributions, five chlorophyll fluorescence parameters—TRo/CS, ABS/CS, RC/CS, ET2o/CS, and RE1o/CS—were proposed as effective indices for assessing drought stress responses in garden plants. As the analyses in this study were conducted using mean values, certain parameter estimates may have been biased, and species-specific characteristics may have been partially obscured. Nevertheless, this research is significant in that an integrated analysis of extensive data collected from multiple species enabled the identification of common indicators applicable to drought stress assessment in garden plants. Future studies are expected to further improve the practical applicability of the proposed indicators by incorporating sensitivity analyses that consider species-specific factors affecting soil moisture content—such as leaf area, canopy coverage, and transpiration rate—as well as through validation under diverse planting conditions.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
