Chlorophyll Pigment and Leaf Macronutrient Trait Variation of Four Salix Species in Elevated CO₂, under Soil Moisture Stress and Fertilization Treatments

Abstract

Leaf chlorophyll pigment and macronutrient concentrations were quantified for four willow species (Salix cordata (COR), S. discolor (DIS), S. eriocephala (ERI) and S. interior (INT)) while growing under 2 × 2 factorial of ambient and elevated CO₂ and well-watered and drought-stressed soil moisture treatments (Expt. 1). After the first year, we saw no difference in stem biomass between CO₂ treatments. In the following year, a second experiment was conducted on a subset of well-watered willows as a 2 × 2 factorial of atmospheric CO₂ and soil fertilization (FERT). For both years of Expt. 1, chlorophyll a, b, a + b (TCC) and carotenoids (CAR) significantly downregulated in response to elevated CO₂ (eCO₂) and upregulated in response to drought (DRT). In Expt. 2, FERT mitigated CO₂ downregulation of TCC and CAR, and upregulated TCC and CAR. Across species, ERI had the greatest pigment concentrations followed by either COR or DIS. Except for one case, INT had the lowest pigment concentrations. A significant species x FERT interaction was due to magnitude effects. The CHLa:b ratio was not affected by CO₂ or DRT but did increase in response to FERT. INT had the greatest CHLa:b ratio followed by DIS, then either ERI or COR. In the second year, TCC:CAR ratio decreased in response to eCO₂ and increased in response to DRT and FERT. In Expt. 1, leaf N was the only nutrient to significantly downregulate in response to eCO₂; whereas all other nutrient levels remained unchanged. In response to DRT, leaf N and Mg upregulated; whereas leaf P, K, and Ca were downregulated. In response to eCO₂ in Expt. 2, again only leaf N downregulated; whereas all other nutrients remained unchanged. All leaf nutrients upregulated in response to FERT. Of the four species, INT had the greatest leaf N and K, and the lowest Ca. Species variation was important, but so to was clonal variation in response to change. Indeed, INT leaf chlorophyll and macronutrients are significantly different or segregated from the other three willow species and this may be related to the evolutionary origins of INT, and other species of the taxonomic section Longifoliae, in the arid southwest USA and Mexico. Furthermore, under low nutrient conditions, it may be necessary to fertilize the plants to see a biomass response to eCO₂.

1. Introduction

Chlorophyll is a defining photosynthetic pigment of plants and varies widely among species, environment, and leaf age [1,2,3,4]. Lower chlorophyll amounts intercept less light by the plant, thereby decreasing photosynthesis and growth [4,5]. Multiple stresses alter the pigment composition and ratios of plants, often resulting in changes or loss of chlorophyll [3,6]. We know that the short-term response to an increase in atmospheric CO₂ stimulates photosynthesis [7,8]. However, downregulation of the photosynthetic apparatus (e.g., chlorophyll and biochemical efficiency traits) can occur in response to elevated CO₂ (eCO₂) and can result in the reduction in carboxylation efficiencies and maximum assimilation, along with other related traits [9,10,11,12]. Studies have shown that photosynthetic apparatus downregulation generally ranges from no downregulation to complete downregulation (when A is the same under eCO₂ as it is under ambient levels), depending on species and study [13,14]. With climate warming, frequency and severity of drought have been predicted to increase [15]. Water is a predominant factor in determining the geographic and landscape distribution of vegetation, and drought tolerance is an important factor in plant growth and mortality [16]. Physiologically, water stress and heat often reduce photosynthesis, increase photo-oxidative stress resulting in leaf abscission, and decrease leaf and plant growth rate [17,18]. Tree physiological responses to water availability vary among and within species [17,19].

Nitrogen (N) is an essential nutrient for the components of chlorophyll, enzymes, proteins and other organic compounds, and N levels control growth [20]. Nitrogen has been positively associated with chlorophyll pigment content [21], photosynthesis [10], and overall productivity [22]. Other macronutrients, such as phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg), are essential elements in various plant physiological functions, and as structural components of DNA and RNA, osmotic potential generation for growth, cell wall constituents, and chlorophyll constituents [20]. Many studies have noted variation in species requirements for each of these elements [23,24]. In a review of tree nutrition, [25] postulated that although tree nutrition has not been the primary focus of climate change studies, there are likely to be changes to nutritional status and growth rates due to the effects from eCO₂.

Salix cordata Michaux (COR), S. discolor Muhl. (DIS), S. eriocephala Michx. (ERI), and S. interior Rowlee (INT) are shrub willows native to eastern and central Canada and each of these species appear promising as fast-growing sources of woody biomass production [26,27]. Salix cordata, also known as sand dune willow, occurs in dune landscapes associated with the North American Great Lakes, but also occurs on riverbanks, and lakeshores in sandy, silty or gravelly soils. Salix interior is also widespread in western North America, ranging from Mexico to Alaska and has been observed to be comparatively saline tolerant [28,29]. Salix discolor is commonly found in wet areas on a wide variety of disturbed sites, whereas ERI and INT are more commonly associated with fast-flowing streams and rivers.

The predicted doubling of CO₂ this century could affect drought responses and comparative fitness among plant species. There are few, if any, climate change response studies on willow chlorophyll pigments and the leaf macronutrients involved in photosynthesis. Photosynthesis can be downregulated in response to eCO₂ [30,31], but some willow species have shown increased photosynthesis in response to eCO₂ [32,33]. Our goal was to examine and compare chlorophyll pigment and leaf macronutrients in four willow species under the interactive effect of atmospheric CO₂ and soil moisture treatments and atmospheric CO₂ and soil fertilization. We hypothesized that eCO₂, soil moisture stress, and fertilization responses would differ among willows. Our specific objectives were to (1) quantify genetic variation in chlorophyll pigment and leaf macronutrient traits of four willow species; COR, DIS, ERI and INT, (2) examine their physiological responses and interactions to eCO₂ and soil moisture stress treatments, and (3) further examine their physiological responses and interactions to eCO₂ and soil fertilization treatments.

2. Materials and Methods

This greenhouse study exposed rooted willow cuttings to different CO₂ and soil moisture treatments over two growing seasons (2017 and 2018). The above-ground biomass was harvested after the first year, and the plants were then allowed to coppice (resprout) and grow for a second year of CO₂ and soil moisture treatments. In the second year a sub-set of plants grown under well-watered conditions were given different levels of fertilizer.

3. Plant Material, Growing Conditions and Treatment Delivery

Sixteen genotypes (clones), four clones from each of four willow species, COR, DIS, ERI, and INT were selected for this study (see clone

Table 1. Species and clones of Salix used in 2017 CO₂ × soil moisture and 2018 CO₂ × fertilization experiments.

				2017		2018
Species	Clone ID	Provenance	Lat-Long	Chlorophyll	Nutrients	Chlorophyll	Nutrients
S. cordata	BIG-C4m	Big Sandy Bay, ON	44°06′ N–77°43′ W	X	X	X
	BIG-C5f	Big Sandy Bay, ON	44°06′ N–77°43′ W	X		X
	NOR-C4f	North Beach Park, ON	43°56′ N–77°31′ W	X	X	X	X
	OUT-C1m	Outlet Beach, ON	43°53′ N–77°13′ W	X		X	X
S. discolor	HAW-D4m	Hawkesbury, ON	45°36′ N–74°36′ W	X	X	X
	HAW-D5f	Hawkesbury, ON	45°36′ N–74°36′ W	X		X	X
	MON-D1f	Montmagny, QC	46o58′ N–70o33′ W	X	X	X	X
	RIC-D2m	Richmond Fen, ON	45°07′ N–75°42′ W	X		X
S. eriocephala	ALL-E2m	Allumette Island, QC	45°54′ N–77°06′ W	X	X	X
	FRE-E1f	Fredericton, NB	45°57′ N–66°38′ W	X		X	X
	GRE-E1f	Green River, NB	47°34′ N–68°19′ W	X	X	X	X
	WAI-E1m	Wainfleet, ON	42°55′ N–79°20′ W	X		X
S. interior	LAF-I2f	Ottawa, ON	45°25′ N–75°41′ W	X	X	X	X
	LAF-I12m	Ottawa, ON	45°25′ N–75°41′ W	X		X
	WAI-I1m	Wainfleet, ON	42°55′ N–79°20′ W	X	X	X
	WAI-I2f	Wainfleet, ON	42°55′ N–79°20′ W	X		X	X

). Cuttings approximately 20 cm long were collected either in the field in April 2017 and kept frozen (−8 °C), or in early May 2017 and kept cool at 4 °C, in sealed plastic bags until 7 June 2017, when all cuttings were moved to a common 4 °C cooler. Cuttings were placed in water at room temperature for two days, prior to planting in the greenhouse on 13 June 2017. CO₂ treatments were ambient (aCO₂—no CO₂ added, approximately 400 ppm) and elevated (eCO₂—target of 800 ppm), with eCO₂ treatment maintained by a delivery system based on the opening or closing of solenoid valves to control CO₂ into the outside air stream. Irrigation treatments were well-watered (WW, volumetric water content target of 20–25%) and drought (DRT, VWC 10–15%). Irrigation levels were controlled by manually powered valves delivering water as needed to maintain treatment VWC target levels. On 19 July 2018, sixteen plants from each of the four well-watered (W) greenhouse chambers were randomly selected, two ramets from each of two clones (genotypes) from each of the four species. Paired ramets in each chamber were selected to be of similar in size. For the remainder of the experiment, no fertilizer (NOFERT) plants received the same watering regime as all other W plants via spigots, while fertilized (FERT) plants were hand-watered with an equal volume of water (typically 1 L two times per week) containing fertilizer (Plant Prod “Forestry Special” 20:8:20 delivered at 100 ppm N). Please see Section 2.1 in [31] for detailed growing conditions of treatment delivery. Eight bulk soil samples (one bulked from FERT and one from NONFERT pots for each chamber) were collected in early September 2018 from NONFERT and FERT treatments and sent to the University of New Brunswick nutrient laboratory for analyses. The soil analysis procedure was the same used previously [34].

4. Chlorophyll Determinations—2017 and 2018

On 24–25 October 2017 (Day 118–119 of treatments), two leaves were collected from each of 128 plants (four species × four clones × four treatments × two blocks = 128) for pigment determinations. Leaves were collected in the early morning, kept cool and dark, and samples were prepared immediately that day in the laboratory. The following procedure was repeated twice: four leaf disks (0.75 cm diameter), two from each leaf, taken mid-blade, were weighed fresh to the nearest 0.00001 g, sealed in 1.5 mL microcentrifuge tubes and immediately placed in the dark at −80 °C. In addition, some empty tubes were included as blanks. Working quickly, leaf material was exposed to the light for less than 3 min prior to freezing. The fresh weight was taken from remaining leaf material, then reweighed following 48 h at 65 °C, for determining fresh and dry mass ratio.

Chlorophyll a and b, and total carotenoids were extracted and measured in N-N-Dimethyl-formamide (DMF) [29,35], following two weeks in the freezer. Each sample received 1500 µL of DMF (D1194, Fisher Scientific, Lot 172209) Microcentrifuge tubes were then resealed and shaken at 125 rpm (Innova 2000 platform shaker, New Brunswick Scientific, Edison, NJ, USA) for 23 h at room temperature while covered with aluminum foil. Each extract was then well-mixed by inversion, and 750 µL drawn off and diluted 1:1 with DMF in a 2 mL microcentrifuge tube. Absorbance values were measured at 480, 647, and 664 nm in matched quartz cuvettes on a spectrophotometer (Ultrospec 2100 pro, Biochrom Ltd., Holliston, MA, USA). “Extracts” from empty tubes measured approximately every 15–20 samples had absorbances at all three wavelengths of 0.000–0.003, so no blank correction was applied. Concentrations (ug·mL⁻¹) of chlorophyll a and chlorophyll b, total chlorophyll (chlorophyll a + b) and total carotenoids were calculated according to the equations and absorbance coefficients for DMF with resolution of 1–4 nm, published by [36]. Pigment concentrations were calculated on fresh weight (ug·g⁻¹) of punched samples (0.0001). Pigment concentrations for a given plant were taken as the mean of the two replicated subsamples for that plant.

In 2018, we followed the same procedures as 2017. Leaf samples were collected on 8 August 2018. The 128 samples for the CO₂ × watering study in 2018 included all 32 NO FERT plants. In addition, all 32 FERT plants were sampled, thus yielding an overall total of 160 chlorophyll samples.

5. Leaf Nutrient Analyses—2017 and 2018

Dried leaf material from the 31 October 2017 harvest for the 64 plants sampled for gas exchange was ground and analyzed for elemental macronutrient levels. Clones used for leaf analyses in each year, see Table 1. The Laboratory for Forest Soils and Environmental Quality at the University of New Brunswick used standard protocols (e.g., Method numbers TP-SSMA 15.3.1, 15.3.3, and 15.4 [37], for foliage analysis of P, K, Ca, Mg and Na). Total N was determined for each sample using an elemental analyzer (CNS-2000, LECO Corporation, St. Joseph, MI, USA). Details of nutrient analysis methods are found in Section 2 [38]. Note that we also analyzed for leaf Na as we have found a willow species effect in a previous study [28]. The same methods were used for foliar analyses of the 64 plants in the 2018 CO₂ x fertilizer sub-study.

6. Statistical Analysis

This study was established as a completely randomized block experimental design. For the chlorophyll pigment analysis for the CO₂ × water treatment experiment (1) there were 128 samples. Willow species, CO₂, soil water treatments and genotype nested within species were considered as fixed effects. The data were subjected to analyses of variance (ANOVA) using the following ANOVA model:

Y_ijklmn = μ + B_i + S_j +C_k + W_l + SC_jk + SW_jl + CW_kl + SCW_jkl +G_m(j) + CG_km(j) + WG_lm(j) + CWG_klm(j) + e_ijklmn

(1)

where Y_ijklmn is the dependent seedling trait of the i^th greenhouse block, of the j^th willow species, of the k^th CO₂ treatment, of the l^th water treatment, of the m^th genotype, n^th seedling, and μ is the overall mean. B_i is the effect of the i^th greenhouse block (i = 1, 2), S_j is the effect of the j^th species (j = 1, 2, 3, 4), C_k is the effect of the k^th CO₂ treatment (k = 1, 2), W_l is the effect of the l^th water treatment (l = 1, 2), SC_jk is the interaction effect of the j^th willow species and the k^th CO₂ treatment, SW_jl is the interaction effect of the j^th willow species and l^th water treatment, CW_kl is the interaction effect of the k^th CO₂ treatment and the l^th water treatment, SCW_jkl is the interaction effect of the j^th willow species, k^th CO₂ treatment and l^th water treatment, G_m(j) is the effect of the m^th genotype nested within the j^th willow species, CG_km(j) is the interaction of the k^th CO₂ treatment and m^th genotype nested within the j^th willow species, WG_lm(j) is the interaction of the l^th water treatment and m^th genotype nested within the j^th willow species, CWG_klm(j) is the interaction of the k^th CO₂ treatment, l^th water treatment and m^th genotype nested within the j^th willow species, and e_ijklmn is the random error component. Effects were considered statistically significant at the p = 0.05 level, although individual p values are provided so that readers can make their own interpretations. The data satisfied normality and equality of variance assumptions. The general linear model from Systat (Chicago, IL) was used for analysis. Species were tested using Tukey’s mean separation test (p = 0.05). The variance component analysis was conducted using the sum of squares as outlined by ([39] pp. 55–57) Genotype is nested within species, but when we discuss genotype (species) we will refer to this as the clone effect.

For the 2018 FERT x CO₂ chlorophyll pigment experiment, (2) and the macronutrient analysis there were 64 samples. The 2017 macronutrient analysis was on 64 samples using the same model substituting water for FERT treatments. Willow species, CO₂ and soil FERT treatments were all considered as fixed effects. The data were subjected to analyses of variance (ANOVA) using the following ANOVA model:

Y_ijklm = μ + B_i + S_j +C_k + F_l + SC_jk + SF_jl + CF_kl + SCF_jkl + e_ijklm

(2)

where Y_ijklm is the dependent seedling trait of the i^th greenhouse block, of the j^th willow species, of the k^th CO₂ treatment, of the l^th FERT treatment, of the m^th seedling, and μ is the overall mean. B_i is the effect of the i^th greenhouse block (i = 1, 2), S_j is the effect of the j^th species (j = 1, 2, 3, 4), C_k is the effect of the k^th CO₂ treatment (k = 1, 2), F_l is the effect of the l^th FERT treatment (l = 1, 2), SC_jk is the interaction effect of the j^th willow species and the k^th CO₂ treatment, SF_jl is the interaction effect of the j^th willow species and l^th FERT treatment, CF_kl is the interaction effect of the k^th CO₂ treatment and the l^th FERT treatment, SCF_jkl is the interaction effect of the j^th willow species, k^th CO₂ treatment and l^th FERT treatment. and e_ijklm is the random error component.

We examined the relationship between TCC and maximum carboxylation efficiency (V_cmax) from 31 to leaf N under CO₂, soil moisture stress and fertilizer treatments. To assist interpretation, covariate analysis was used to evaluate relationships among TCC, V_cmax in relations to leaf N concentration. The Covariate methodology is described previously [38]. Results were considered statistically significant at p = 0.10 due to fewer (8) data points, although individual p values are provided for all traits so that readers can make their own interpretations of significance.

7. Results

There were no soil nutrient differences between FERT and NOFERT for organic matter, C, N, CN ratio, Ca, Na, and S (

Table 2. Soil properties (mean ± SE) during the 2018 NOFERT and FERT treatments. Treatments with different letters are significantly different using ANOVA test, α = 0.05.

Treatment	Organic Matter (%)	Carbon (%)	Nitrogen (%)	Potassium (meq/100 g)	Calcium (meq/100 g)	Magnesium (meq/100 g)	Phosphorus (ppm)
NOFERT	0.67 ± 0.40 a	0.39± 0.02 a	0.085 ± 0.009 a	0.04 ± 0.01 b	1.21 ± 0.08 a	0.050 ± 0.004 b	6.50 ± 0.50 b
FERT	0.63 ± 0.40 a	0.37 ± 0.02 a	0.090 ± 0.009 a	0.16 ± 0.01 a	1.08 ± 0.08 a	0.068 ± 0.004 a	9.50 ± 0.50 a
Treatment	Sodium (%)	Sulfur (%)		pH		C:N ratio
NOFERT	0.11 ± 0.01 a	0.020 ± 0.007 a		6.8 ± 0.1 a		4.7 ± 0.2 a
FERT	0.11 ± 0.01 a	0.013 ± 0.007 a		6.2 ± 0.1 b		3.9 ± 0.2 a

). FERT soil nutrient traits were greater than NOFERT for K, Mg, and P. NOFERT had higher pH than FERT with 6.8 and 6.2, respectively. The 2017 CO₂ × water stress treatments ANOVA results for chlorophyll a, b and TCC appear in (

Table 3. 2017 CO₂ × soil moisture experiment: chlorophyll trait variance components and ANOVAs including source of variation, degrees of freedom (df), mean square values (MS), variance component (VC), p values, and coefficient of determination (R²). p values < 0.05 are in bold print.

Source of Variation	df	Chlorophyll a (mg·g−1)			Chlorophyll b (mg·g−1)			Carotenoids (mg·g−1)
		MS	VC (%)	p-Value	MS	VC (%)	p-Value	MS	VC (%)	p-Value
Block	1	0.15	0.4	0.343	0.023	0.4	0.365	0.008	0.7	0.165
SP 1	3	2.165	17.0	<0.001	0.449	20.1	<0.001	0.076	21.2	<0.001
CO2	1	3.049	8.0	<0.001	0.516	7.7	<0.001	0.089	8.2	<0.001
Water	1	7.149	18.7	<0.001	1.135	17.0	<0.001	0.227	21.0	<0.001
SP × CO2	3	0.067	0.5	0.750	0.010	0.5	0.777	0.001	0.2	0.927
SP × Water	3	0.169	1.3	0.387	0.032	1.4	0.334	0.005	1.3	0.323
Water × CO2	1	1.786	4.7	0.002	0.331	4.9	0.001	0.034	3.2	0.005
SP × Water × CO2	3	0.210	1.6	0.290	0.037	1.6	0.277	0.006	1.7	0.215
Clone(SP)	12	0.279	8.7	0.089	0.047	8.4	0.090	0.008	8.4	0.047
Clone(SP) × CO2	12	0.145	4.5	0.571	0.025	4.4	0.564	0.004	4.4	0.444
Clone(SP) × water	12	0.122	3.8	0.707	0.021	3.7	0.698	0.003	3.2	0.708
Clone(SP) × water * CO2	12	0.120	3.8	0.717	0.021	3.7	0.707	0.003	3.6	0.621
Error	63	0.164	27.0		0.028	26.1		0.004	22.9
R2				0.731			0.740			0.772
Source of variation	df	Total chlorophyll (mg·g−1)			Chlorophyll a:b ratio			Chlorophyll: carotenoids ratio
		MS	VC (%)	p-value	MS	VC (%)	p-value	MS	VC (%)	p-value
Block	1	0.291	0.4	0.348	0.000	<0.1	0.905	0.034	0.1	0.769
SP	3	4.581	17.9	<0.001	0.156	36.3	<0.001	3.887	20.5	<0.001
CO2	1	6.075	7.9	<0.001	0.002	0.2	0.500	0.091	0.2	0.630
Water	1	13.982	18.2	<0.001	0.002	0.2	0.479	0.030	0.1	0.782
SP × CO2	3	0.127	0.5	0.761	0.006	1.3	0.288	0.556	2.9	0.242
SP × Water	3	0.347	1.4	0.371	0.008	1.9	0.152	0.188	1.0	0.695
Water × CO2	1	3.654	4.8	0.001	0.012	0.9	0.101	1.946	3.4	0.029
SP × Water × CO2	3	0.420	1.6	0.287	0.005	1.2	0.347	0.235	1.24	0.614
Clone(SP)	12	0.548	8.6	0.093	0.026	24.5	<0.001	0.611	12.9	0.123
Clone(SP) × CO2	12	0.288	4.5	0.568	0.004	3.8	0.541	0.244	5.2	0.810
Clone(SP) × water	12	0.242	3.8	0.705	0.006	5.8	0.194	0.405	8.5	0.422
Clone(SP) × water × CO2	12	0.239	3.7	0.714	0.002	2.0	0.921	0.055	1.2	1.000
Error	63	0.326	26.8		0.005	22.0		0.388	43.0
R2				0.733			0.781			0.571

¹ Species.

). The three traits followed the same trend and thus only TCC findings are presented below. The 2018 CO₂ × water stress treatments ANOVA results are similar to the 2017 findings and are found in supplementary section (Table S1). There were often significant clone effects, but no significant clone × CO₂, clone × water or clone × CO₂ × water interactions. The significant species clonal results are presented in a separate section below, rather than repeat results for each parameter.

8. Chlorophyll and Carotenoid Content

From the 2017 CO₂ x water experiment, there were significant differences in TCC, for the effects of species, CO₂, water, clones and the water x CO₂ interaction which accounted for 17.9, 7.9, 18.2 and 8.6% of the total variation (Table 3), respectively. Overall, COR, DIS and ERI were not significantly different with an average of 2.65 versus INT, which was the lowest at 1.91 mg g⁻¹, 28% less in TCC. All four species downregulated TCC under eCO₂ with an average decrease in 16% (Figure 1a). Drought increased TCC for all species on average by 31% (Figure 1b). The water x CO₂ interaction was a result eCO₂ mitigating the drought effect.

In the 2018 CO₂ × FERT experiment, variation in TCC was significant for species and FERT effects and the species × FERT, CO₂ × FERT and species × FERT × CO₂ interactions (

Table 4. 2018 CO₂ x fertilizer experiment: chlorophyll trait variance components and ANOVAs including source of variation, degrees of freedom (df), mean square values (MS), variance component (VC), p values, and coefficient of determination (R²). p values < 0.05 are in bold print.

Source of Variation	df	Chlorophyll a (mg·g−1)			Chlorophyll b (mg·g−1)			Carotenoids (mg·g−1)
		MS	VC (%)	p-Value	MS	VC (%)	p-Value	MS	VC (%)	p-Value
Block	1	0.430	2.8	0.002	0.089	3.6	0.001	0.007	2.9	0.008
SP 1	3	0.326	6.4	<0.001	0.068	8.2	<0.001	0.008	10.3	<0.001
CO2	1	0.002	0.0	0.842	0.002	0.1	0.639	0.001	0.4	0.247
FERT2	1	10.418	68.0	<0.001	1.613	64.7	<0.001	0.134	55.1	<0.001
SP × CO2	3	0.039	0.8	0.412	0.007	0.8	0.397	0.001	1.2	0.328
SP × FERT	3	0.252	4.9	0.001	0.042	5.1	0.002	0.005	6.6	0.001
CO2 × FERT	1	0.187	1.2	0.037	0.023	0.9	0.076	0.005	2.1	0.024
SP × CO2 × FERT	3	0.179	3.5	0.008	0.028	3.4	0.014	0.004	4.9	0.005
Error	47	0.040	12.4		0.007	13.3		0.001	16.5
R2				0.876			0.867			0.835
Source of variation	df	Total chlorophyll (mg·g−1)			Chlorophyll a:b ratio			Chlorophyll: carotenoids ratio
		MS	VC (%)	p-value	MS	VC (%)	p-value	MS	VC (%)	p-value
Block	1	0.912	3.0	0.002	0.014	1.4	0.283	3.254	3.9	0.002
SP	3	0.687	6.8	<0.001	0.070	20.5	0.002	0.46	1.7	0.205
CO2	1	0.006	0.0	0.779	0.002	0.2	0.669	5.13	6.2	<0.001
FERT	1	20.228	67.2	<0.001	0.127	12.4	0.002	56.886	68.7	<0.001
SP × CO2	3	0.080	0.8	0.405	0.008	2.4	0.554	0.09	0.3	0.817
SP × FERT	3	0.498	5.0	0.001	0.013	3.9	0.360	0.27	1.0	0.433
CO2 × FERT	1	0.343	1.1	0.045	0.023	2.2	0.174	0.357	0.4	0.272
SP × CO2 × FERT	3	0.348	3.5	0.009	0.006	1.8	0.687	0.355	1.3	0.310
Error	47	0.081	12.6		0.012	55.2		0.289	16.4
R2				0.874			0.449			0.836

¹ Species, ² Fertilizer treatment.

). In the three-way interaction, INT did not respond to the FERT × CO₂ interaction in the same way as the other three species (Figure 1c). The species × FERT interaction as seen in Figure 1d, were a magnitude effect, TCC responded less to FERT in INT than it did in the other species to FERT. The CO₂ × FERT interaction was a rank change; under no FERT, TCC decreased in response to eCO₂ (Figure 1c), whereas, under FERT, TCC increased in response to eCO₂. Overall, there was a strong and significant increase in TCC in response to FERT from 1.24 to 2.39 mg g⁻¹ or a 92% increase (Figure 1d).

9. Carotenoid Content

In CAR content from the 2017 CO₂ x water experiment, species, CO₂, water and clone effects and water x CO₂ interaction were significant accounting for 21.2, 8.2, 21.0 and 3.2% of the total variation (Table 3). Overall, species CAR were quite different: ERI had the greatest CAR, followed by DIS, then COR, followed by INT with the lowest CAR at 370, 337, 316 and 254 ug g⁻¹, respectively (Figure 2a). All four species showed CAR downregulation in response to eCO₂ with an average decrease in 15%. Drought increased CAR for all species an average of 30% (Figure 2b). The water x CO₂ interaction resulted from eCO₂ mitigating the drought effect.

In the 2018 CO₂ × FERT experiment, variation in CAR was significant for species and FERT effects and for the interactions of species × FERT, CO₂ × FERT and species x FERT x CO₂ (Table 4). The CO₂ × FERT was a rank change interaction, similar to that described above and seen in Figure 1c. The species × FERT interaction as seen in Figure 2c is a magnitude effect: in INT, the CAR response is lower than in the other species to FERT. Overall, there was a significant increase in CAR in response to FERT from 177 to 277 ug g⁻¹ or a 56% increase.

10. Chlorophyll a:b and Chlorophyll to Carotenoid Ratio

From the 2017 CO₂ × water experiment, CHLa:b was significant for species and clonal effects accounting for 36.3 and 24.5% of the total variation (Table 3). INT had the greatest CHLa:b ratio at 2.60 compared with DIS and ERI which were similar at 2.48 (Figure 3a). COR had the lowest ratio with 2.43. There were no consistent CO₂ or drought effects on CHLa:b ratio. From the 2018 CO₂ × FERT experiment, CHLa:b was significant for species and FERT accounting for 20.5 and 12.4% of the total variation (Table 4). INT had the greatest CHLa:b ratio at 2.59, followed by DIS, and then by COR and ERI (Figure 3b).

For the TCC:CAR ratio in the 2017 CO₂ x water experiment, species differences had a significant effect as did the CO₂ x water interaction accounting for 20.5 and 3.4 % of the total variation (Table 3). COR had the greatest TCC:CAR ratio with 8.2, followed by DIS with 7.8, then ERI and INT with on average 7.4 (Figure 4a). From the 2018 CO₂ × FERT experiment, CO₂ and FERT effects were significant, accounting for 6.2 and 68.7% of the total variation (Table 4). Elevated CO₂ decreased the TCC:CAR ratio (Figure 4b), whereas FERT increased the ratio (Figure 4c).

11. Clonal Variation

In the 2017 CO₂ × water experiment, clonal variation of chlorophyll pigments accounted for approximately 8.5% of the total variation (Table 3). There was some variation among clones within each species: for example, DIS clones 3 and 4 (order in Table 1 had greater chlorophyll levels than clones 1 and 2. In COR, Clones 1 and 2 were greater than clones 3 and 4; in ERI, clone 4 was lower than the other three clones. For CHLa:b ratio clonal variation accounted for 24.5% of the total variation (Table 3).

12. Leaf Macronutrients

From the 2017 CO₂ × DRT experiment, species, CO₂ and water main effects were significant for leaf N accounting for 14.7, 13.1 and 20.6% of the total variation (

Table 5. 2017 CO₂ × soil moisture experiment: leaf macronutrients variance components and ANOVAs including source of variation, degrees of freedom (df), mean square values (MS), variance component (VC), p values, and coefficient of determination (R²). p values < 0.05 are in bold print.

Source of Variation	df	Leaf Nitrogen (%)			Leaf Phosphorus (%)			Leaf Potassium (%)
		MS	VC (%)	p-Value	MS	VC (%)	p-Value	MS	VC (%)	p-Value
Block	1	0.807	6.9	0.008	0.016	5.8	0.045	0.040	0.2	0.602
SP 1	3	0.571	14.7	0.003	0.002	2.2	0.686	1.194	19.2	<0.001
CO2	1	1.526	13.1	<0.001	0.003	1.1	0.352	0.019	0.1	0.718
Water	1	2.394	20.6	<0.001	0.037	13.3	0.003	6.788	36.5	<0.001
CO2 × SP	3	0.013	0.3	0.942	0.004	4.3	0.364	0.044	0.7	0.824
Water × SP	3	0.041	1.1	0.756	0.004	4.3	0.392	0.303	4.9	0.116
Water × CO2	1	0.003	0.0	0.872	0.004	1.4	0.293	0.216	1.2	0.231
Water × CO2 × SP	3	0.051	1.3	0.691	0.003	2.9	0.544	0.022	0.4	0.928
Error	47	0.104	41.9		0.004	64.8		0.146	36.9
R2				0.596			0.349			0.632
Source of variation	df	Leaf calcium (%)			Leaf magnesium (%)			Leaf sodium (%)
		MS	VC (%)	p-value	MS	VC (%)	p-value	MS	VC (%)	p-value
Block	1	0.111	1.4	0.303	0.018	5.5	0.002	0.00009	1.4	0.338
SP	3	0.260	10.2	0.067	0.065	59.5	<0.001	0.00046	21.0	0.006
CO2	1	0.166	2.2	0.209	0.004	1.2	0.110	0.00002	0.3	0.653
Water	1	0.588	7.7	0.020	0.018	5.5	0.002	0.00001	0.2	0.701
CO2 × SP	3	0.103	4.0	0.398	0.001	1.2	0.480	0.00001	0.5	0.963
Water × SP	3	0.109	4.3	0.374	0.001	1.2	0.505	0.00008	3.8	0.471
Water × CO2	1	0.517	6.8	0.029	0.002	0.6	0.285	0.00007	1.1	0.419
Water × CO2 × SP	3	0.016	0.6	0.922	0.001	1.2	0.522	0.00001	0.6	0.935
Error	47	0.102	62.8		0.002	23.9		0.00010	71.2
R2				0.373			0.763			0.293

¹ Species.

). INT had significantly greater leaf N than the other three species, by on average 22% (Figure 5a). All willow species showed a significant leaf N downregulation in response to eCO₂, which averaged 15.8%. All willow species showed on average a 24% leaf N upregulation in response to DRT (Figure 5b). From the 2018 CO₂ x FERT experiment, species, CO₂ and FERT main effects were significant for leaf N accounting for 2.6, 0.6 and 89.6% of the total variation (

Table 6. 2018 CO₂ x fertilizer experiment: leaf macronutrient variance components and ANOVAs including source of variation, degrees of freedom (df), mean square values (MS), variance component (VC), p values, and coefficient of determination (R²). p values < 0.05 are in bold print.

Source of Variation	df	Leaf Nitrogen %			Leaf Phosphorus %			Leaf Potassium %
		MS	VC (%)	p-Value	MS	VC (%)	p-Value	MS	VC (%)	p-Value
Block	1	0.353	0.4	0.059	0.0008	0.3	0.617	0.522	4.0	0.020
SP 1	3	0.678	2.6	<0.001	0.0018	2.2	0.620	1.913	43.9	<0.001
CO2	1	0.481	0.6	0.028	0.0002	0.1	0.820	0.107	0.8	0.280
FERT 2	1	71.339	89.6	<0.001	0.0716	29.4	<0.001	0.956	7.3	0.002
CO2 × SP	3	0.019	0.1	0.892	0.0020	2.4	0.580	0.009	0.2	0.960
FERT × SP	3	0.244	0.9	0.063	0.0045	5.6	0.222	0.291	6.7	0.030
CO2 × FERT	1	0.066	0.1	0.408	0.0004	0.2	0.716	0.480	3.7	0.025
CO2 × FERT × SP	3	0.054	0.2	0.636	0.0018	2.2	0.622	0.049	1.1	0.651
Error	47	0.094	5.6		0.003	57.6		0.090	32.3
R2				0.945			0.424			0.677
Source of variation	df	Leaf calcium (%)			Leaf Magnesium (%)			Leaf sodium (%)
		MS	VC (%)	p-value	MS	VC (%)	p-value	MS	VC (%)	p-value
Block	1	0.075	0.5	0.412	0.00019	0.1	0.598	0.000564	17.0	<0.001
SP	3	1.045	22.8	<0.001	0.04318	68.3	<0.001	0.000281	25.3	<0.001
CO2	1	0.037	0.3	0.566	0.00069	0.4	0.316	0.000002	<0.1	0.824
FERT	1	3.866	28.1	<0.001	0.01351	7.1	<0.001	0.000039	1.2	0.268
CO2 × SP	3	0.248	5.4	0.093	0.00013	0.2	0.904	0.000014	1.3	0.717
FERT × SP	3	0.216	4.7	0.131	0.00344	5.4	0.004	0.000002	0.1	0.985
CO2 × FERT	1	0.066	0.5	0.443	0.00150	0.8	0.141	0.000127	3.8	0.049
CO2 × FERT × SP	3	0.010	0.2	0.966	0.00066	1.1	0.406	0.000081	7.3	0.063
Error	47	0.110	37.4		0.00067	16.6		0.000031	44.0
R2				0.626			0.834			0.560

¹ Species, ² Fertilizer treatment.

). INT had significantly greater leaf N than the other three species, by on average 20% (Figure 5c). All willow species showed a significant leaf N downregulation in response to eCO₂, which averaged 8.4%. All willow species showed on average a 326% leaf N upregulation in response to FERT (Figure 5d).

From the 2017 CO₂ × DRT experiment, leaf K was significant for species and DRT accounting for 19.2 and 36.4% of the total variation (Table 5). INT had significantly greater K than the other three species, by on average 27.6%. Leaf K declined for all species on average by 27.8 in response to DRT. From the 2018 CO₂ × FERT experiment, species and FERT effects and the species × FERT and CO₂ × FERT interactions were significant, accounting for 43.9, 7.3, 6.7 and 3.7% of the total variation (Table 6). Overall, INT and COR, with 2.11 and 1.95%, had greater K than DIS and ERI with 1.47 and 1.43% K, respectively. The species × FERT effect was a result of COR, DIS and ERI showing an increase in K in response to FERT, but INT showed no change. Overall, FERT and NO FERT treatments had1.86 and 1.62% K, respectively. The CO₂ × FERT interaction was the result of FERT mitigating the eCO₂ downregulation under NO FERT.

From the 2017 CO₂ × DRT experiment, leaf Ca was significant for water and CO₂ × water interaction accounting for 7.7 and 6.7% of the total variation (Table 5). Leaf Ca decreased under DRT by 9.6%. The CO₂ × water interaction was the result of increase in leaf Ca in response to DRT under eCO₂, but unchanged under aCO₂. From the 2018 CO₂ × FERT experiment, species and FERT accounted for 22.8 and 28.1% of the total variation (Table 6). COR and DIS were greater than ERI and INT with 2.04, 2.05, 1.76 and 1.52, respectively. FERT increased leaf Ca by 30.6%.

From the 2017 CO₂ × experiment, leaf Mg was significant for species and water effects accounting for 59.5 and 5.5% of the total variation (Table 5). COR had the greatest leaf Mg followed by ERI and INT and DIS with the lowest values at 0.376, 0.308, 0.245 and 0.240, respectively (Figure 6a). DRT increased leaf Mg for all species on average by 12.4% (Figure 6b). From the 2018 CO₂ × FERT experiment, species and FERT effects and the species x FERT interaction were significant (Table 6). Overall, COR had the greatest leaf Mg followed by INT and ERI and with DIS at the lowest with 0.283, 0.203, 0.188, 0.150%, respectively (Figure 6c). The species x FERT interaction was the result of an increase in leaf Mg for DIS and ERI, but unchanged for COR and INT (Figure 6d). Overall FERT and NOFERT treatments had 1.86 and 1.62% Mg, respectively.

From the 2017 CO₂ × DRT experiment, leaf Na was only significant for species accounting for 21% of the total variation (Table 5). INT was significantly greater than the other three willows with 0.016 and 0.005% Na, respectively. From the 2018 CO₂ × FERT experiment, species and the CO₂ × FERT interaction were significant, accounting for 25.3 and 3.8% of the total variation (Table 6). INT was significantly greater than the other three willows with 0.012 and 0.004% Na, respectively.

13. Interrelationships

Mean willow species TCC to leaf N for CO₂ treatments appear in Figure 7a. There is a substantial INT separation (circled) from the other three willow species. The regression line for the other three willow species is not significant (broken line) but shows the CO₂ treatment downregulation trend. Mean willow species TCC to leaf N for soil moisture treatments also show a substantial INT separation from the other three willow species (Figure 7b). There is a significant soil-moisture regression line showing the upregulation of both in response to drought. Mean willow species TCC to leaf N for NOFERT and FERT treatments show two groupings based on FERT treatments (Figure 7c). There is a significant regression line showing an upregulation of both traits in response to FERT.

Covariate analysis of 2017 V_cmax in relation to leaf N for CO₂ treatments showed no significant CO₂ × leaf N interaction (p = 0.455) nor CO₂ effect (p = 0.371). As a result, there is one single positive line (Figure 8a). Covariate analysis of 2017 V_cmax in relation to leaf N for soil moisture treatments showed a significant soil moisture x leaf N interaction (p = 0.092), significant soil moisture effect (p = 0.053) and leaf N (p = 0.019), thus there are two separate lines for soil moisture treatments with different slopes (Figure 8b). Covariate analysis of 2018 V_cmax in relation to leaf N for FERT treatments showed a significant FERT × leaf N interaction (p = 0.062). As a result, there are two separate lines with different slopes (Figure 8c).

14. Discussion

There are several consistent findings for chlorophyll pigment and macronutrient parameters between the two years and two experiments. First, chlorophyll pigments and leaf N downregulated in response to eCO₂ in both years for all four willow species. After an extensive review, there is not much in the scientific literature on willow regarding chlorophyll pigment and leaf N responses to eCO₂ ([30]: used a European cross/genotype Salix x dasyclados under CO₂ and N treatments). There are however, some publications on willow responses to eCO₂: biochemical efficiencies [31] and gas exchange [30,31,32,33]. Photosynthesis appears to be downregulated in response to eCO₂ [30,31], but [32,33] reported increased photosynthesis in response to eCO₂. This discrepancy in the literature is probably related to a measurement date effect. We know that the short-term response to an increase in CO₂ stimulates assimilation (A) [7,40]. However, there can also be a corresponding assimilation downregulation (A_dr), expressed as a decrease in photosynthesis and related traits in response to eCO₂ with time [9,10,11,12]. We found that all four willows downregulate from the maximum rate of carboxylation (V_cmax) and assimilation at 600 ppm (A₆₀₀) [31], consistent with chlorophyll pigment downregulation. The chlorophyll downregulation is underlying the performance based V_cmax downregulation.

The corresponding leaf N downregulation, the only macronutrient to downregulate in eCO₂, reflects the chlorophyll pigment downregulation. However, there is a strong species component for both traits reflected by the large component of variance for these traits. This is primarily due to differential INT response compared to the other three willow species for chlorophyll pigment and leaf N concentrations (Figure 1, Figure 5 and Figure 7). At first, these results may appear contradictory; INT has greater leaf N, but a lower chlorophyll pigment concentration than the other three willows. However, this may simply indicate a unique species response (more on this later). In a similar study illustrating leaf N levels with a strong species effect, despite large plant size differences, equal amounts of fertilizer demonstrated that red pine (Pinus resinosa) and pitch pine (P. rigida) had similar needle N concentrations while being at the opposite ends of the biomass size scale [38]. Red pine was the smallest pine and the second smallest of the eight pine and spruce species tested, whereas pitch pine was overall the largest of these eight species.

TCC appears negatively related to leaf N within a treatment (Figure 7), but this is not consistent with the literature [38,41]. In fact, INT has shown a number of unique and differential species responses under saline and drought stress [28,38]. In a saline stress experiment under control (CTL) and medium and high salinity treatments (MST and HST, target EC = 1.5 and 3.0 mS cm⁻¹, respectively), all DIS and ERI plants expired under HST, but 33% of INT plants survived [34]. For DIS and ERI, total aboveground (AG) dry mass decreased from CTL to MST, whereas AG dry mass for INT increased slightly in MST and increased further in HST. In addition, INT had greater V_cmax than DIS and ERI and increased V_cmax under saline stress treatments [29]. In a drought response test, V_cmax was lower under DRT for COR, DIS and ERI, but 16.5% greater for INT under DRT compared to WW [31]. Drought response has a similar physiological effect as saline response in that both conditions induce moisture stress in plants, and both require cellular osmotic adjustment to withstand the stress. The greater drought and saline tolerance of INT may reflect its evolutionary origins in the arid Southwest USA and Mexico [42,43,44] where high evapotranspiration may increase drought and soil salinity (see [45] and references therein) and result in a natural selection for increased drought and saline tolerance. The exact cause of the greater INT drought tolerance could be due to changes to the hydraulic architecture of the xylem and thus greater hydraulic conductivity [46] or as a result of greater osmotic adjustment [47,48].

The TCC and leaf N downregulation in response to eCO₂ is a response to a greater supply of a resource (CO₂). The TCC downregulation and the thus the leaf N downregulation is probably in response to the accumulation of nonstructural carbohydrates in an eCO₂ environment. This is often associated with, and commonly interpreted as, evidence of a lack of sink activity [49] This supports the theory of sink regulation (biomass growth or lack thereof) of photosynthetic traits, in this case TCC [38,49].

Our preliminary biomass data showed no CO₂ effect on willow total aboveground dry mass after the first year of growth in 2017 (p = 0.214, Major and Mosseler unpublished). [50] found that fertilization of eCO₂ plots restored eCO₂ growth in forest species, and [10] found fertilization reversed photosynthetic downregulation. Thus, we decided to add a fertilizer x CO₂ experimental component in 2018 on a subset of fully irrigated treatments to see if we could observe a CO₂ effect under FERT. This change resulted in a mitigation of the biochemical downregulation found in 2017 in response to eCO₂ [31]. This was doubtless driven by the FERT effect on biomass accumulation which resulted in a significant (p = 0.049) increase in stem dry mass and CO₂ x fertilizer interaction (Major and Mosseler unpublished). As in 2017, there was no stem dry mass difference between aCO₂ and eCO₂ under NOFERT in 2018 with 11.8 and 10.7 g, respectively. However, under the FERT treatment, stem dry mass was 20.0 and 25.3 g under aCO₂ and eCO₂, respectively. Thus, nutrient limitation caused the lack of eCO₂ effect on willow growth in 2017.

TCC and leaf N increased, in response to FERT by 92 and 326%, respectively, for all four willow species. Note that the soil nutrients were the same or just slightly higher under FERT compared to UNFERT, hence most newly available nutrients were absorbed by the willows. In a companion study, FERT increased V_cmax, A₆₀₀, by 86 and 91%, respectively [31]. This seems to be a direct reflection of the increase in TCC. As found for Phoebe boumei seedlings FERT alleviates photosynthetic decline in response to eCO₂ [12].

Chlorophyll pigments, leaf N and Mg increased (the only macronutrients to increase) in response to drought for all four willow species. Each chlorophyll a and b molecule require four N and one Mg. There was a greater slope of TCC to needle N than Mg which reflects the four to one ratio difference in the chlorophyll requirement [38]. In a multi-species, inter-specific, study of shrub willows under drought treatment, leaf Soil Plant Analysis Development (SPAD) measurements, which measures chlorophyll as a proxy for leaf N, the willows were unaffected by water stress treatments [51]. Using 200 willow genotypes, ref. [52] found that greater leaf N in response to drought was frequently observed in drought-exposed plants and may be seen as an acclimation to water stress [53,54,55]. This acclimation enables plants to make better use of the available N resources when leaf area and stomatal conductance are greatly reduced [55,56]. Contrary to the above, Salix nigra in response to periodic soil moisture stress had a significant reduction in leaf chlorophyll content by 15%, which was not as severe as the photosynthetic and stomatal conductance reduction with 32 and 55% reduction, respectively, observed by [57]. In conifers, it appears that chlorophyll pigments either remain the same or decline depending on the severity of drought [38,58,59].

For green mass and stem number for seven willow species including two willow tree species, species effect accounted for 30 and 39% of the variation, and clone accounted for 12 and 11% of the variation, respectively [60]. In addition, population accounted for 12 and 8% of the variation for green mass and stem number, respectively. For chlorophyll and leaf nutrient, species was almost always significant accounting for a large part of the total variation However, clonal nested within species had about the same or less amount of total variation as morphological traits but was not always significant. Note, in Expt. 1, species and clones were the only significant sources of variation for CHLa:b ratio, accounting for just over 60% of variation, demonstrating a very strong genetic control. INT had the greatest CHLa:b ratio compared to the other three willow species. There were no water or CO₂ treatment effects. In experiment 2, species and FERT had significant effects, with INT once again having the greatest CHLa:b ratio. There have been reports that in response to eCO₂ spruces show a decline in CHLa:b ratio [38,61]. All four willows showed a decline in TCC:CAR ratio in response to eCO₂; a result of a greater TCC decline than CAR decline in response to eCO₂. TCC:CAR ratio increased in response to DRT and FERT due to a greater increase in TCC than in CAR.

Similar to our findings, needle nutrient analyses of Pinus sylvestris showed that N significantly declined in response to eCO₂ while K, Ca and Mg remained the same [62]. As discussed above, TCC, leaf N and Mg upregulated in response to DRT. In addition, leaf P, K and Ca downregulated in response to DRT. This often happens with these macronutrients as feedback to water shortage and leads to stomatal regulation and reduction in the transpiration—driven water flow required for most nutrient uptake [63]. These macronutrient responses to DRT can be very genus or species dependent [38,63,64]. Interestingly, leaf Na uptake was greater in INT than the other three willow species. This was also found in another salinity study compared with only DIS and ERI [28]. In a review of the physiological mechanism of saline tolerance, the absorption of Na⁺ into cellular vacuoles may increase osmotic potential in a saline environment [65]. The subsequent soil analysis showed very low levels of nutrients under NOFERT. In addition, soil analysis of the FERT treatments in 2018 showed no difference between FERT and NOFERT, particularly for soil N. Thus, everything supplied by the FERT treatment was taken up by the fertilized willows.

15. Application

Shrub willow species show rapid early growth, particularly after coppicing, and have become a promising group of species for biomass production and as energy crops [26,27,66]. Some willows do have a symbiotic relationship with diazotrophic endophytes in the stem that provide N to the plant in exchange for carbohydrates [67,68]. In the present study, we provided only two pulses of fertilizer early in the first year, assuming that endophytic bacteria would be present under both CO₂ treatments. However, it is now apparent that N fixation was either too low or that not enough other essential soil nutrients were present to allow the plants to take advantage of the eCO₂ treatment. With increasing atmospheric CO₂, if a site is nutrient-poor, or that plants lack the necessary endophyte associations, the addition of soil nutrients or endophytes can result in a synergistic effect to sequester more available carbon dioxide due to the effects of N limitation.

In addition, INT appears to be more drought and saline tolerant compared to the three other willows. INT and its closest relatives in the taxonomic section Longifoliae (6 or 7 species) are also unique among willow species for their ability to easily form new stems via stem sprouts from a network of shallow roots, resulting in multi-stemmed colonies [27,69]. The rapid growth, high biomass yields, and ease of vegetative (clonal) propagation of INT from unrooted stem sections may also be of interest for establishing biomass feedstock for bioenergy, chemicals, and materials industries, especially on marginal land with salinity issues or prone to periods of drought.