- Journal Details
- Format
- Journal
- eISSN
- 2353-3978
- First Published
- 30 Jul 2013
- Publication timeframe
- 2 times per year
- Languages
- English
Agriculture is the largest consumer of water globally, accounting for 70% of total withdrawals (UNEP 2007) but is limited by shortfalls in water quality and quantity (Hoekstra & Chapagain 2007; Postel 2000). Limited water supplies, and continued population growth, demand more efficient water management (Sample 2009; IEEE 2010).
Demand for ornamental plants has been significantly increasing during the past decade (Incrocci et al. 2014) for purposes of urban decoration and gardening and also for environmental restoration and bioremediation (Denys et al. 2006; Sun et al. 2011). This fact, in combination with the abovementioned aspects, implies a double challenge for the plant nursery industry, which will have to increase its production (Savé et al. 1999) and, at the same time, reduce its water inputs through resource optimization.
Historically, one of the most important innovations in the nursery sector has been the switch from mineral soil to soilless cultivation in plastic containers filled with inert substrate that simplified the task of controlling and optimizing water inputs (Di Lorenzo et al. 2013). However, nursery stock in organic substrates are still over-irrigated and leach slow-release fertilizers (Ristvey et al. 2004), which wastes water and pollutes waterways (Stewart-Wade 2011).
Appropriate nursery irrigation requires scheduling in order to provide plants with the adequate amount of water at the correct time. Such scheduling can be based on (1) climate – the evaporative demand and its effects on soil–water balance; see Allen et al. (1998); (2) substrate – its water-holding properties, and the monitoring of substrate moisture (Θ) status; (3) plant water use traits – centered on the relationship between crop water stress and soil water deficit; see Jones (2004) or on their integration. In particular, mathematical modeling is a way to combine the abovementioned strategies with an integrated approach (Gu et al. 2020).
Although microclimate and substrate water-holding properties are quantifiable, plant water use traits are complex and vary among species and cultivars (Knox 1989; Schuch & Burger 1997; Mugnai et al. 1999). Few guidelines are nowadays available for the planning and management of irrigation, and in the reality of nursery production, these topics are addressed most of the times based on the empirical observations more than on modeling (Di Lorenzo et al. 2013).
The aim of the present work was to study the behavior of three ornamental species in containers under increasing water stress, through a continuous monitoring of their ecophysiological responses. The final aim is to highlight differences among the studied species and classify them based on their strategies of stress response, in order to optimize water investments and nursery practices according to the plant type.
Three common ornamental shrubs were chosen for this experiment:
Plants, 14 months of age, were cultivated in plastic containers, with a 18-cm top diameter, 6.5 dm3 of volume, filled with a peat–pumice (1 : 1, v : v) mixture, and placed in a non-heated iron/PVC greenhouse, whose coordinates were 43°70′43″ North and 10°42′75″ West; the period of the year was September.
A five-day dry-down cycle versus control was performed. Plants were watered daily by hand as a partial or total root zone refilling of water lost through evapotranspiration. Control plants were kept at the 100% of container capacity during the whole trial, whereas stressed plants were subjected to 90%, 80%, 70%, and 60% of container capacity along the study period. Six replicates were used for each species and water treatment combination. The entire cycle was replicated three times, subsequently, on new specimens not previously subjected to drought stress.
Air temperature (T) and relative humidity (RH) in the greenhouse were constantly monitored with a data logger FT-105/RF-LCD (Econorma®, Trento, Italy). For each day of trial, the average value of three measurements corresponding to the hours of the day with highest radiation (11:30, 12:30, and 13:30) was calculated. Vapor pressure deficit (VPD) was calculated from T and RH for the same hours.
Incident radiation was measured using a pyranometer (Delta-T Devices, Burwell, England) connected to the data logger. Global incident radiation during the period of measurements of the physiological parameters was calculated daily as the difference between the cumulated radiation values at 14:00 and 11:00. Climatic data are presented in Figure 1.
Fig. 1
Trend of climatic data during the days of experiment: Temperature (closed circles), vapor pressure deficit (VPD, open circles), and global radiation (RG) are shown. Error bars for each average value of temperature and VPD come from 3 measurements, taken at the hours of the day with highest radiation (11:30, 12:30, and 13:30). RG values refer to the interval 11:30–13:30
Water use was gravimetrically determined daily as the difference in weight between 14:00 (end of measurements) and 11:00 (beginning of measurements). Hourly evapotranspiration was then calculated.
Containers were weighted again every day at sunset in order to calculate the amount of water needed for the reintegration up to water content planned for the next day. A water retention curve was calculated for the substrate used in the experiment through tensiometric cassette and Richard's plates according to the Piemonte Region analytical methods (ARPA 1992; Cassel & Klute 1986). The relationships between container capacity and substrate water tension are given in Table 1.
Substrate water tension (SWT) versus percentage of container capacity for the three investigated species
| SWT | % C.C. | SWT | % C.C. | SWT | % C.C. |
|---|---|---|---|---|---|
| 8 | 100 | 18 | 100 | 23 | 100 |
| 26 | 89.5 | 35 | 90.7 | 41 | 90.7 |
| 46 | 78.8 | 41 | 85.6 | 65 | 85.6 |
| 78 | 69.6 | 65 | 76.1 | 90 | 76.1 |
| 117 | 59.8 | 101 | 65.3 | 135 | 65.3 |
Leaf stomatal conductance was measured with a transit-time diffusion porometer (Mk3, Delta-T Devices, Burwell, England) and then transformed, for each species, into relative values (gsrel = [(gsi/gsmax) · 100], where gsi was one random measurement and gsmax was the maximum value recorded for a species.
Foliar temperature was measured through a portable infrared thermometer (Cyclops Compac 3 Minolta/Land, Sheffield, UK), assuming an emissivity (ε) of 0.93 for plant tissues (Styles et al. 2002). Both leaf conductance and temperature were measured daily on two marked leaves for each plant in the interval between 11:30 and 13:30.
At the end of the trial, the leaf area was measured for each plant with a digital planimeter (ΔT Area Meter Mk2, Delta-T Devices, Burwell, England).
Data were subjected to analysis of variance (ANOVA) using SAS version 9.1 (SAS Institute, USA). 3-way ANOVA was performed with species × day (container water capacity) × cycle as descriptors. The significance of differences between means was determined using Duncan's multiple range test (SAS 1990).
For each measurement of leaf stomatal conductance (gspor), the corresponding modeled value gscal was calculated from the transpiration rate and VPD following the equation of Pearcy et al. (1991):
ANOVA of the eco-physiological plant variables (ET/RG, gs, and ΔT) showed no significant differences among cycles (p > 0.05) and no significant interaction between cycle and the two other factors (day and species). Therefore, data from the first cycle are presented here to show the trends of ET/RG, gs, and ΔT, as representative of the three cycle repetitions. The decrease in container water capacity in treated plants is expressed as the substrate water tension.
Fig. 2
Leaf area by species. Values marked with the same letter are not statistically different at p < 0.05, according to Duncan's multiple range test. Error bars were calculated from six measurements (replicates) for each average value
Fig. 3
ET/RG ratio versus substrate water tension
Note: See Fig. 2
Stomatal conductance declined for all species as soil water availability diminished (Fig. 4), whereas control plants kept their values at 100% ± 5% (which means an average fluctuation of ± 5% around the first-day-optimal value) during the whole trial (data not shown in the graph). For all the days of experiment,
Fig. 4
Relative stomatal conductance (gsrel) versus substrate water tension
Note: See Fig. 2
Fig. 5
Linear correlation between measured and calculated stomatal conductance; for each species, a single regression line, with slope and R2, is presented in the graph; the central dotted line represents the ideal 1 : 1 relationship between measured and calculated gs
ΔT (Tleaf − Tatm) increased for all species as water stress got more severe (Fig. 6), whereas control plants kept stable values (data not shown in the graph).
Fig. 6
ΔT versus substrate water tension. Note: See Fig. 2
The results point out
This phenomenon is also explained by the capacity of
The similar leaf areas for
On the other side, the hydraulic behavior of
The increase in ΔT from day 1 to day 5 is a typical consequence of reduced foliar transpiration because of stomatal closure from water stress, which reduces evaporative cooling and so leaf temperature rises (Mugnai 2004; Zuccarini et al. 2011). Low Tleaf − Tatm in
Few studies have reported on hydraulic behavior of
Björkman et al. (1980) already showed gas exchange rates of
This study suggests that
The present data point out
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Substrate water tension (SWT) versus percentage of container capacity for the three investigated species
| SWT | % C.C. | SWT | % C.C. | SWT | % C.C. |
|---|---|---|---|---|---|
| 8 | 100 | 18 | 100 | 23 | 100 |
| 26 | 89.5 | 35 | 90.7 | 41 | 90.7 |
| 46 | 78.8 | 41 | 85.6 | 65 | 85.6 |
| 78 | 69.6 | 65 | 76.1 | 90 | 76.1 |
| 117 | 59.8 | 101 | 65.3 | 135 | 65.3 |