Regression Models to Estimate Accumulation Capability of Six Metals by Two Macrophytes, Typha domingensis and Typha elephantina, Grown in an Arid Climate in the Mountainous Region of Taif, Saudi Arabia

Abstract

In this study, we explored the capacity for two promising macrophytes, Typha domingensis and Typha elephantina, to be used for the surveillance of contamination by six metals, i.e., Cu, Fe, Mn, Ni, Pb, and Zn, in the mountainous area of Taif City in Saudi Arabia. Regression models were generated in order to forecast the metal concentrations within the plants’ organs, i.e., the leaves, flowers, peduncles, rhizomes, and roots. The sediment mean values for pH and the six metals varied amongst the sampling locations for the respective macrophytes, indicating that similar life forms fail to indicate equivalent concentrations. For instance, dissimilar concentrations of the metals under investigation were observed within the organs of the two rooted macrophytes. The research demonstrated that the segregation of metals is a regular event in all the investigated species in which the metal concentrations vary amongst the different plant constituent types. In the current study, T. domingensis and T. elephantina varied in their capacity to absorb specific metals; the bioaccumulation of metals was greater within T. domingensis. The relationships between the observed and model-estimated metal levels, in combination with high R² and modest mean averaged errors, offered an appraisal of the goodness of fit of most of the generated models. The t-tests revealed no variations between the observed and model-estimated concentrations of the six metals under investigation within the organs of the two macrophytes, which emphasised the precision of the models. These models offer the ability to perform hazard appraisals within ecosystems and to determine the reference criteria for sediment metal concentration. Lastly, T. domingensis and T. elephantina exhibit the potential for bioaccumulation for the alleviation of contamination from metals.

1. Introduction

Metals are a specific class of elements that, in contrast to organic contaminants, are unable to be broken down via biological processes [1]. Owing to their adverse impact on the environment, the issues relating to metals have recently been highlighted [2]. Biological pathways essential for the viability of organisms can be inhibited by raised metal concentrations. The latter are especially hazardous in wetland ecosystems as once these elements are introduced and accrued within the system, they enter and climb the food chain. In the upper trophic strata, their presence becomes biomagnified and can give rise to a number of chronic pathologies in both humans and animals [3]. Exposure to toxic metals typically occurs through the ingestion of contaminated products within the diet, the immediate consumption of soil or polluted drinking water, and the inhalation of air particle suspensions [4].

There are two principal causes of metal pollution. Firstly, metals may be derived from natural geochemical mechanisms such as ultramafic rock erosion; secondly, anthropogenic activities may facilitate metal contamination [5]. With respect to the latter, metals may be derived from a number of substances that are allowed access to the environment, such as fertilisers and pesticides, wastewater from commercial enterprises, sludge from sewage processing, waste products from mining and electroplating, and contaminants present in the air [6]. The transfer and accumulation of these metals in animal and human bodies through the food chain cause DNA damage, carcinogenic effects, and the induction of mutations [7,8].

Wetlands are ecosystems which are particularly susceptible to the adverse consequences of contamination from various metals [9]. Macrophytes can accumulate metals and nutrients due to their well-developed root systems, vigorous growth, tolerance to toxicity, high above-ground biomass, and high capacity for metal accumulation [10]. Thus, they could form a tool for metal sequestration from the environment and, consequently, phytoremediation [11,12,13]. However, metals may affect the fragile ecological stability of wetlands, whose fundamental role in nutrient cycling and pollution control is widely recognised [14]. More advanced aquatic vegetation is an essential constituent of wetland ecosystems and a major actor in maintaining the ecological equilibrium and environmental biogeochemical metal cycle, predominantly owing to their superlative capacity for metal [15] and nutrient [12] uptake. Thus, in-depth comprehension of their metal accrual and translocation properties and their correlation with metal concentrations in water and sediment are key in understanding the translocation mechanisms of metals at the plant–abiotic constituent boundary and for the renewal, safeguarding, and surveillance of wetland ecosystems [16].

In order to diminish the adverse impact of metals and their entry into the food web, it is essential to gauge the influence of variable sediment parameters on plant metal bioavailability and uptake [17]. A straightforward appraisal of bioaccumulation factors (BAFs) provides an approximate guide into the metal uptake spectrum but fails to offer in-depth information on location-specific sediment properties [18]. Regression models have utility for the estimation of the metallic levels in plants according to a range of sediment measurements, such as metal concentrations, pH, and proportion of organic matter [19]. Plant uptake is primarily governed by metal solubility and bioavailability [20]. The models frequently incorporate pH as this can affect the latter [21,22]. pH plays an important role in metal accumulation within the various organs of T. domingensis [23]. Thus, pH is encompassed within the regression models in the current study. The principal benefit of regression models is that they are economical in both time and expense, diminishing the quantity of data necessary to recognise statistically relevant sediment characteristics [21]. Such models are of high functionality in the assessment of possible hazards to the environment arising from various metals, and they are commonly utilised to estimate wetland plant metal absorption and accumulation [24].

As frequently arising perennial and emergent macrophytes, the cattail species Typha domingensis Pers. and Typha elephantina Roxb. are globally prevalent in wetlands, lakes, rivers, freshwater lagoons and pools, and in brackish marshy regions within tropical and temperate climatic zones [25]. These species favour a muddy sediment and low-depth, slow, or stagnant aquatic environments. They have the capacity to rapidly take over vegetative populations in wetlands, establishing monotypic stands [26]. Typha species are herbaceous and rhizomatous plants that can grow to a height of over 3 metres [27]. They grow rapidly and establish a significant biomass, are able to withstand conditions containing metal pollutants, and take up high concentrations of metals. Several researchers have demonstrated that uptake, accumulation, and translocation of metals may differ significantly among plant species [28,29]. Knowing the differences in metal bioaccumulation between macrophytes is thus fundamental not only to assess the suitability of one species instead of another for phytoremediation uses, but also to implement appropriate actions of ecological restoration and management, thus making phytoremediation more sustainable [1,30]. Although there is a plethora of contemporary research pertaining to the metal bioaccumulation properties of these two cattail species, there are insufficient data to establish whether these two macrophytes exhibit comparable metal accretion and translocation characteristics.

We located only a single study [23] that has explored the capacity of T. domingensis to act as a bioaccumulator of the contamination of eight metals (Ag, Cd, Cu, Co, Fe, Pb, Ni, and Zn) in Lake Burullus, Egypt, through the construction of de novo regression models. To date, no research has been conducted in this manner for T. elephantina. Thus, the principal aim of the present study was to generate regression models that incorporated sediment characteristics as explanatory variables in order to estimate the respective levels of Cu, Fe, Mn, Ni, Pb, and Zn in both T. domingensis and T. elephantina, grown naturally in a mountainous area in a dry climate in Taif, Saudi Arabia. Additionally, the concentrations of the above metals were measured within the various organs of these two macrophytes; the heterogeneity of distribution was investigated according to the species. Further work was conducted in order to compare the metal concentrations within the plants and sediment; the data were utilised as the foundation for the appraisal of the potential of these species to function as bioaccumulators.

2. Materials and Methods

2.1. Study Area

Taif’s history can be traced back to an ancient era that existed over 2000 years ago. The site of the city and its nature as a primitive metropolis in the pre-Islam period earned Taif worldwide recognition; it became a centre of enterprise, enticing visitors from remote regions, e.g., Romans and Persians and people from Abyssinia, Yemen, and Syria [31]. Taif is situated in southeast Makka, between the latitudes and longitudes of 20–22 and 40–42 degrees, respectively. At an altitude of 1700 metres above sea level, the city sits on the eastern hills of the Al Sarawat Mountains within a characteristic warm desert climate. Taif covers approximately 1036 square kilometres, and 95.1 square kilometres form the city’s footprint. Approximately 58% of this is high-density buildings; the residual 42% is non-built-up territory [31]. The region chosen for the study was a natural wetland environment in proximity to the city (Figure 1), to which T. domingensis and T. elephantina are native, generating highly populated monospecies stands. The documented climatic data from the Taif meteorological station indicate that the spectrum of monthly mean minimum temperatures was 7.9 to 23.4 °C, the mean maximum range was between 22.9 °C and 36.3 °C, and the total monthly mean was 23.2 °C. The mean (± standard deviation) maximum and minimum temperatures were 30.6 ± 4.8 °C and 15.8 ± 5.5 °C, respectively; the mean monthly humidity was recorded as 40.6 ± 14.8%. The data from the last 10 years showed considerable inter-annual variation in the monthly amount (range 4.3 ± 5.7–294.1 ± 383.8 mm/month) and timing of rainfall. The monthly amount of rainfall ranges from 4.3 ± 5.7 mm/month in December to 294.1 ± 383.8 mm/month in September [32].

2.2. Sampling Design

Over the years, vast areas of virgin lands in Taif have turned into agriculture lands, resulting in the disappearance of many wild habitats, including wetlands [32]. Therefore, and due to the limitation of available sites, the T. domingensis and T. elephantina populations were identified only in six available sampling locations, three per individual species. Each site comprised stands of almost solely one of these two macrophytes; the remaining vegetation made up less than 5% of the entire standing biomass. Monthly specimen collection was performed between February and September in 2018. For each location, above-ground plant organs were gathered from three randomly assigned 0.5 × 0.5 metre quadrats. The below-ground plant systems, i.e., the roots and rhizomes, were dug from identical quadrats 0.5 metres below ground level, where over 90% of these systems are situated (see [33]). Contamination was circumvented by the utilisation of stainless-steel implements.

The plant organs were segregated into leaves, flowers, peduncles, rhizomes, and roots and were transported to the laboratory in polyethylene sacks. The specimens were rinsed in tap water; a 4 mm filter was utilised to prevent the loss of any parts. They were then oven-dried for 7 days at a temperature of 85 °C. The biomass was assessed using gram dry matter per square metre (g DM/m²) (data not stated). For each plant system, the substance was pulverised with the use of a plastic mill containing no metal parts (Philips HR2221/01, Philips, Shanghai, China). The particulate was then combined to create two composite specimens per sampling location and retained for analysis of metallic concentration. In total, 48 plant specimens per organ, i.e., leaves, roots, and rhizomes and 38 plant specimens per reproductive organ, i.e., peduncles and flowers, were utilised to measure the respective concentrations of Cu, Fe, Mn, Ni, Pb, and Zn within the two macrophytes. As, Cd, Cr, and Hg were also measured, but they were below the detection limits (data not presented).

2.3. Sediment Sampling and Analysis

In comparison to concentrations in water, metal concentrations in sediments are indicators of the long-term accumulation of metals in stagnant and running water bodies [34]] because water metal concentrations can be subject to seasonal variations and might not accurately reflect the actual contamination level [35]. Thus, for each sampling location, three sediment samples were gathered from the identical quadrats from which the vegetation specimens were obtained. The sediment samples were extracted using a stainless-steel manual sediment corer, of length and inner diameter of 50 cm and 70 mm, respectively, to provide a 0–50 cm profile. The sediment specimens were air-dried; small stones and debris were extracted by passing the samples through a 2 mm sieve. They were then combined to generate two composite samples for the individual sampling locations.

Sediment:water extracts (1:5, w/v) were used for the evaluation of pH, which was performed with a pH meter (Model 9107 BN, ORION type, Thermo Fisher Scientific, Waltham, MA, USA). A digestion technique was utilised for metal extraction from both the sediment and the specimens of studied macrophytes by employing a tri-acid admixture (HNO₃:H₂SO₄:HClO₄; 5:1:1, v/v/v) and a microwave sample preparation system (Perkin Elmer Titan MPS, Perkin Elmer Inc., Waltham, MA, USA). Inductively coupled plasma optical emission spectrometry (Thermo Scientific iCAP 7000 Plus Series; Thermo Fisher Scientific, Waltham, MA, USA) was used to measure the metal concentrations of Cu, Fe, Mn, Ni, Pb, and Zn. Every tenth sample, the blanks and quality control references were evaluated in order to identify any contamination or drift. The limits for metal recognition were as follows: Ni, 6.0 µg/kg; Cu, 2.0 µg/kg; Fe, Pb, and Zn, 1.0 µg/kg; and Mn, 3.0 µg/kg. The apparatus parameters and functional properties were in accordance with the vendors’ guidelines. The systems were calibrated using reference solutions with predetermined metal strengths [36]. The levels of metal in the two macrophytes and sediment specimens were computed with reference to dry weight.

2.4. Quality Assurance and Control

We validated the precision of the chemical analysis method using a certified standard, i.e., SRM 1573a, tomato leaves. Comparable techniques to those used for T. domingensis and T. elephantina were applied to this material. The metal digestions and analyses were conducted thrice. The precision was established by contrasting the experimentally determined value with the certified standard measure; the output was given as a percentage. The recovery rate spectrum was between 94% and 106%.

2.5. Data Analysis

The likelihood of metal accumulation within the roots of T. domingensis and T. elephantina was estimated using the BAF. The translocation factor (TF) was calculated to give an indication of the ability of T. domingensis and T. elephantina to translocate metals from the roots to the rhizomes and the above-ground biomass, i.e., the leaves, peduncles, and flowers. These parameters were computed following the method published by Eid and Shaltout [29]:

BAF = Metal concentration in the roots (mg/kg)/Metal concentration in the sediment (mg/kg)

(1)

TF_rhizome = Metal concentration in the rhizomes (mg/kg)/Metal concentration in the roots (mg/kg)

(2)

TF_leaf = Metal concentration in the leaves (mg/kg)/Metal concentration in the roots (mg/kg)

(3)

TF_peduncle = Metal concentration in the peduncles (mg/kg)/Metal concentration in the roots (mg/kg)

(4)

TF_flower = Metal concentration in the flowers (mg/kg)/Metal concentration in the roots (mg/kg)

(5)

The Student’s t-test was used to assess the differences in the sediment quality variables, BAFs, and TFs between the respective macrophyte locations. The relationships between the respective sediment and plant metal concentrations were appraised using Pearson’s simple linear correlation coefficient (r). No variations among the data (not given) for the three sampling locations relevant to each macrophyte were identified. In order to substantiate this, 12 random data sets were selected for each of the five plant organs. The remaining 36 and 24 data sets, for the roots, rhizomes, and leaves and the flowers and peduncles, respectively, were utilised to generate regression models in order to estimate the metal concentrations within the various organs of the two macrophytes. These were deemed to be the dependent variable for the various sediment elements, i.e., Cu, Fe, Mn, Ni, Pb, Zn, and pH. The latter were established as independent variables. The principal elements utilised to define the plant metal concentrations were the sediment metal concentrations and the pH [20]. The main model equation is:

C_plant = a + (b × C_sediment) + (c × pH)

(6)

where C_plant and C_sediment indicate the relevant metal level within the organs of T. domingensis or T. elephantina and within the sediment, respectively, and a, b, and c refer to the regression coefficients.

The standard of the model was established by computing model efficiency (ME), model strength (mean normalised average error: MNAE), and the coefficient of determination (R²) [23]. These parameters were computed according to the following equations [20]:

ME = 1 − {∑ (C_model − C_measured)²/∑ (C_measured − C_mean)²}

(7)

MNAE = {∑ (|C_model − C_measured|/C_measured)}/n

(8)

where C_model, C_measured, and C_mean indicate the model estimated and the observed and mean metal concentrations, respectively, and n represents the observation frequency.

The Student’s t-test was used in order to identify any variations between the model estimated and the observed metal concentration within the identical plant organs. The data set was tested for the presence of a normal distribution and variance homogeneity with the use of the Shapiro–Wilk and Levene tests, respectively, prior to conducting an analysis of variance (ANOVA). Where required, log transformation was applied. One-way ANOVA was used to evaluate the BAFs and TFs for each macrophyte for the respective metals under investigation. In order to appraise the potential impact of the two Typha species and their organs on the TFs for the respective metals, a two-way ANOVA was performed. The Statistical Package for the Social Sciences [37] was used for all the statistical techniques.

3. Results

3.1. Sediment Analysis

Variations within the sediments from the T. domingensis and T. elephantina locations were observed in terms of pH and the concentrations of the investigated metals. The chemical analysis indicated that the sediment was slightly alkaline, ranging between 7.80 and 8.70, with an average of 8.06–8.29 (

Table 1. Metal concentrations and pH of the sediment from six sampling sites (three sampling sites for each species) in the mountainous area of Taif City in Saudi Arabia supporting the growth of Typha domingensis and Typha elephantina populations during one growing season (February 2018 to September 2018).

Value	pH	Metal Concentration in Sediment (mg/kg)
		Cu	Fe	Mn	Ni	Pb	Zn
Typha domingensis
Minimum	7.80	1.88	2.22	2.92	0.09	0.12	0.46
Maximum	8.50	3.19	12.15	4.64	0.13	0.15	2.23
Mean (n = 48)	8.06	2.39	4.76	4.01	0.11	0.14	1.07
CV (%)	2.41	17.73	75.92	14.29	12.16	7.22	54.95
Typha elephantina
Minimum	7.80	2.01	2.28	3.35	0.10	0.13	0.72
Maximum	8.70	5.38	3.55	4.00	0.20	0.24	1.07
Mean (n = 48)	8.29	3.74	2.93	3.73	0.14	0.18	0.88
CV (%)	2.59	38.26	15.56	6.19	28.98	25.12	14.93
t-value	5.98	7.09	3.72	3.29	5.15	5.78	1.98
p	0.000	0.000	0.001	0.002	0.000	0.000	0.054

t-values represent Student’s t-test; CV: coefficient of variance.

). The sediment metal concentration spectrum from the T. domingensis sites was between 0.11 mg/kg (Ni) and 4.76 mg/kg (Fe). The sediment metal concentration range from the T. elephantina locations was between 0.14 mg/kg (Ni) and 3.74 mg/kg (Cu). The CV ranged from 7.2% for Pb to 75.9% for Fe and from 2.0% for Zn to 7.1% for Cu for T. domingensis and T. elephantina, respectively.

3.2. Plant Metals

Within T. domingensis, the mean metal concentration ranges were as follows (

Table 2. Metal concentrations in the roots, rhizomes, peduncles, flowers, and leaves of Typha domingensis and Typha elephantina populations grown in the mountainous area of Taif City in Saudi Arabia during one growing season (February 2018 to September 2018).

Organ	Value	Metal Concentration (mg/kg)
		Cu	Fe	Mn	Ni	Pb	Zn
Roots	Typha domingensis
	Minimum	8.70	699.98	69.75	3.05	3.70	26.58
	Maximum	40.60	2003.05	183.10	282.85	304.40	84.40
	Mean (n = 48)	20.09	1209.05	118.16	141.97	149.42	52.54
	CV (%)	53.33	35.66	37.38	86.20	82.80	40.88
	Typha elephantina
	Minimum	8.77	551.97	35.93	5.57	5.13	11.17
	Maximum	39.40	2281.10	117.68	226.88	205.90	84.18
	Mean (n = 48)	20.40	1207.41	78.92	97.09	90.87	33.59
	CV (%)	53.85	49.38	40.99	84.53	82.71	70.96
	t-value	0.23	0.02	11.02	3.51	4.37	7.92
	p	0.817	0.988	0.000	0.001	0.000	0.000
Rhizomes	Typha domingensis
	Minimum	12.65	980.50	157.80	25.00	22.65	13.35
	Maximum	38.88	5555.03	642.50	148.75	144.05	41.35
	Mean (n = 48)	29.39	3154.89	289.38	108.29	102.46	27.89
	CV (%)	31.63	53.92	59.02	36.91	37.43	33.79
	Typha elephantina
	Minimum	56.60	1309.50	103.77	17.93	16.93	18.18
	Maximum	91.18	4248.80	279.33	127.60	118.23	52.00
	Mean (n = 48)	71.69	2517.53	218.55	72.92	68.06	27.61
	CV (%)	16.82	39.82	28.71	50.95	50.13	43.80
	t-value	26.96	1.85	2.22	10.85	9.95	0.12
	p	0.000	0.071	0.031	0.000	0.000	0.907
Peduncles	Typha domingensis
	Minimum	9.20	339.15	35.90	31.90	30.80	6.50
	Maximum	47.10	2362.10	149.00	304.70	310.70	52.55
	Mean (n = 36)	26.02	1276.01	94.61	156.65	147.07	17.53
	CV (%)	58.43	71.04	43.95	71.31	74.64	101.87
	Typha elephantina
	Minimum	3.40	102.11	44.43	1.37	2.07	2.33
	Maximum	100.95	1247.93	147.10	140.00	127.90	18.80
	Mean (n = 36)	39.19	652.32	79.37	77.02	71.62	10.62
	CV (%)	100.07	69.85	45.31	80.34	79.94	67.89
	t-value	2.17	3.49	1.69	3.89	3.65	2.88
	p	0.036	0.001	0.099	0.000	0.001	0.006
Flowers	Typha domingensis
	Minimum	10.65	390.00	34.90	41.50	41.10	14.85
	Maximum	39.80	1592.35	293.20	263.20	275.40	58.75
	Mean (n = 36)	20.53	786.72	181.31	123.07	127.96	27.19
	CV (%)	49.95	54.67	51.51	64.26	65.08	62.48
	Typha elephantina
	Minimum	2.40	144.70	55.23	1.10	3.23	10.07
	Maximum	40.97	926.00	231.30	96.83	92.63	15.60
	Mean (n = 36)	14.90	529.63	130.27	32.26	31.81	12.77
	CV (%)	99.74	59.53	55.35	106.53	101.69	15.89
	t-value	1.67	2.56	2.05	6.54	6.62	5.58
	p	0.103	0.015	0.047	0.000	0.000	0.000
Leaves	Typha domingensis
	Minimum	1.15	162.25	87.75	2.45	2.40	5.95
	Maximum	21.80	1181.75	691.15	177.55	172.80	81.85
	Mean (n = 48)	13.68	721.90	245.65	103.86	101.32	35.28
	CV (%)	49.53	45.63	84.39	60.69	61.09	80.08
	Typha elephantina
	Minimum	9.15	392.60	91.40	4.20	16.47	6.68
	Maximum	22.80	1058.85	187.07	234.10	214.30	70.46
	Mean (n = 48)	13.64	663.68	142.40	86.66	83.17	31.72
	CV (%)	34.40	34.39	22.55	85.49	78.36	66.39
	t-value	0.05	1.66	3.33	2.72	2.85	0.57
	p	0.961	0.103	0.002	0.009	0.007	0.573

t-values represent Student’s t-test; CV: coefficient of variance.

): Cu, 13.68–29.39 mg/kg; Fe, 721.90–3154.89 mg/kg; Mn, 94.61–289.38 mg/kg; Ni, 103.86–156.65 mg/kg; Pb, 101.32–149.42 mg/kg; and Zn, 17.53–52.54 mg/kg. Within T. elephantina, the mean metal concentration ranges were as follows: Cu, 13.64–71.69 mg/kg; Fe, 529.63–2517.53 mg/kg; Mn, 78.92–218.55 mg/kg; Ni, 32.26–97.09 mg/kg; Pb, 31.81–90.87 mg/kg; and Zn, 10.62–33.59 mg/kg. For T. domingensis, the CV ranged from 31.6% for Cu in the rhizomes to 101.9% for Zn in the peduncles; for T. elephantina, the CV ranged from 15.9% for Zn in the flowers to 106.5% for Ni in the same organ. For the two macrophytes, the results of the t-tests for the metal concentrations showed significant differences in the concentrations of Mn, Ni, Pb, and Zn in the roots; Cu, Mn, Ni, and Pb in the rhizomes; Cu, Fe, Ni, Pb, and Zn in the peduncles; Fe, Mn, Ni, Pb, and Zn in the flowers; and Mn, Ni, and Pb in the leaves (Table 2). In T. domingensis, greater concentrations of Cu, Fe, Mn, Pb, and Zn were present in the below-ground organs, and diminished concentrations of Cu, Fe, Ni, and Pb were identified within the leaves. Mn and Zn were in lower concentrations within the peduncles. In T. elephantina, all the investigated metals were found in increased concentrations in the below-ground organs, together with diminished concentrations of Fe, Ni, and Pb in the flowers. The leaves and peduncles exhibited diminished Cu and Zn levels, respectively. The greatest concentration of Cu (71.69 mg/kg) was noted in the rhizome system of T. elephantina. The rhizome samples from T. domingensis contained the maximum levels of Fe (3154.90 mg/kg) and Mn (289.38 mg/kg). The highest Ni concentration (156.65 mg/kg) was measured in the peduncles from T. domingensis. The roots of the latter macrophyte demonstrated the greatest accrual of Pb (149.42 mg/kg) and Zn (52.54 mg/kg).

3.3. Bioaccumulation and Translocation of Metals

All the measured metals had a BAF > 1 within the roots of both macrophytes (Figure 2; Table S1). The BAFs for Cu, Mn, Ni, Pb, and Zn were dissimilar for T. domingensis and T. elephantina. A greater ability for biological metal accumulation was seen in T. domingensis. Nickel exhibited the largest BAF, i.e., 1405.35 in T. domingensis and 736.64 in T. elephantina. Copper displayed the lowest value BAF, i.e., 8.38 in T. domingensis and 5.76 in T. elephantina. For both macrophytes, the metal BAFs in descending order were Ni > Pb > Fe > Zn > Mn > Cu. The translocation factors for Cu, Fe, Ni, Pb, and Zn relating the roots to the rhizomes, leaves, peduncles, and flowers varied between the two cattail species. A diversity of TFs for all the metals under investigation was observed between the varying plant organs (Figure 3; Table S1). The two greatest TFs were seen for Fe (3.21) in T. domingensis and Cu (4.46) in T. elephantina; the two lowest values were for Zn (0.30) in T. domingensis and Ni (0.31) in T. elephantina. Overall, the BAFs and TFs varied for each of the metals measured within each plant organ (Table S1).

3.4. Plant–Sediment Correlations

Within both macrophytes, positive or negative associations were noted between the sediment and plant organ concentrations for most of the measured metals (Figure 4 and Figure 5; Tables S2 and S3). In both T. domingensis and T. elephantina, many metal levels within the varying organs exhibited a positive relationship with the identical metal in the sediment. In T. domingensis, positive associations were seen for both Cu (r = 0.624) and Fe (r = 0.814) within the rhizomes and for Pb in both the peduncles (r = 0.499) and the flowers (r = 0.696). In T. elephantina, positive correlations were noted for Cu (r = 0.349), Mn (r = 0.645), and Zn (r = 0.771) in the roots; for Cu (r = 0.888), Fe (r = 0.378), Mn (r = 0.514), and Zn (r = 0.531) in the rhizomes; for Ni (r = 0.621) and Pb (r = 0.599) in the peduncles; and for Ni (r = 0.293) and Pb (r = 0.398) in the leaves. In both T. domingensis and T. elephantina, several metal concentrations within the differing plant organs displayed a negative association with their respective sediment concentration. In T. domingensis, negative relationships were observed for Fe (r = −0.465), Ni (r = −0.848), Pb (r = −0.342), and Zn (r = −0.415) in the roots; for Mn (r = −0.752), Ni (r = −0.499), and Pb (r = −0.333) in the rhizomes; for Cu (r = −0.369), Fe (r = −0.344), Ni (r = −0.959), and Zn (r = −0.697) in the peduncles; for Mn (r = −0.621), Ni (r = −0.732), and Zn (r = −0.749) in the flowers; and for Fe (r = −0.743), Ni (r = −0.795), Pb (r = −0.294), and Zn (r = −0.764) in the leaves. In T. elephantina, negative relationships were demonstrated for Fe (r = −0.400) in the roots; for Ni (r = −0.716) and Pb (r = −0.690) in the rhizomes; for Cu (r = −0.958) and Fe (r = −0.533) in the peduncles; for Fe (r = −0.493) and Mn (r = −0.495) in the flowers; and for Cu (r = −0.493) and Fe (r = −0.396) in the leaves.

3.5. Prediction Models

In order to estimate the metal concentrations of the various plant organs from the two macrophytes, regression models were generated (

Table 3. Regression models between six metal concentrations in Typha domingensis organs (mg/kg) and sediment metals (mg/kg) and pH.

Equation	R2	ME	MNAE	Student’s t-Test
				t-Value	p
Roots
Cu = −85.76 + (5.08 × CuSediment) + (11.62 × pH)	0.111	0.445	0.173	1.517	0.163
Fe = 6962.87 − (72.92 × FeSediment) − (670.61 × pH)	0.286 **	0.798	0.102	1.155	0.273
Mn = −51.74 + (17.56 × MnSediment) + (12.35 × pH)	0.070	0.391	0.183	1.730	0.115
Ni = 38.52 − (7601.65 × NiSediment) + (118.12 × pH)	0.754 ***	0.979	0.028	0.190	0.853
Pb = 239.09 − (3783.22 × PbSediment) + (54.57 × pH)	0.123	0.462	0.155	1.496	0.164
Zn = −165.19 − (10.52 × ZnSediment) + (28.39 × pH)	0.222 **	0.691	0.119	1.360	0.201
Rhizomes
Cu = −168.31 + (10.60 × CuSediment) + (21.38 × pH)	0.569 ***	0.900	0.062	0.624	0.545
Fe = −21,294.84 + (455.45 × FeSediment) + (2763.73 × pH)	0.739 ***	0.945	0.033	0.282	0.783
Mn = 1696.59 − (210.24 × MnSediment) − (70.10 × pH)	0.570 ***	0.916	0.054	0.436	0.673
Ni = −604.81 − (1430.48 × NiSediment) + (108.26 × pH)	0.525 ***	0.876	0.068	0.898	0.388
Pb = −553.23 − (578.10 × PbSediment) + (91.36 × pH)	0.292 ***	0.811	0.081	1.065	0.310
Zn = −61.67 − (0.70 × ZnSediment) + (11.20 × pH)	0.065	0.353	0.267	1.743	0.112
Peduncles
Cu = 190.46 − (10.68 × CuSediment) − (17.07 × pH)	0.170 *	0.610	0.133	1.384	0.194
Fe = 7590.06 − (185.50 × FeSediment) − (703.10 × pH)	0.134	0.556	0.152	1.463	0.178
Mn = −84.28 + (3.89 × MnSediment) + (20.02 × pH)	0.008	0.100	0.346	1.939	0.079
Ni = 1256.03 − (7396.26 × NiSediment) − (35.22 × pH)	0.921 ***	0.982	0.008	0.164	0.872
Pb = 283.88 + (5222.91 × PbSediment) − (105.68 × pH)	0.275 **	0.788	0.105	1.247	0.244
Zn = −108.79 − (46.12 × ZnSediment) + (20.29 × pH)	0.522 ***	0.854	0.069	0.974	0.351
Flowers
Cu = 106.31 + (2.43 × CuSediment) − (11.30 × pH)	0.036	0.240	0.272	1.874	0.094
Fe = 3663.54 − (16.50 × FeSediment) − (347.84 × pH)	0.026	0.195	0.299	1.928	0.086
Mn = 1033.07 − (177.63 × MnSediment) − (12.55 × pH)	0.386 ***	0.814	0.081	1.000	0.343
Ni = 472.96 − (4068.78 × NiSediment) + (12.04 × pH)	0.536 ***	0.880	0.067	0.840	0.419
Pb = −567.01 + (5804.10 × PbSediment) − (13.06 × pH)	0.485 ***	0.825	0.074	0.977	0.350
Zn = −86.52 − (46.99 × ZnSediment) + (18.83 × pH)	0.595 ***	0.928	0.047	0.361	0.725
Leaves
Cu = −127.11 − (0.56 × CuSediment) + (17.63 × pH)	0.245 **	0.750	0.109	1.356	0.208
Fe = −1131.89 − (60.78 × FeSediment) + (265.80 × pH)	0.570 ***	0.901	0.062	0.622	0.550
Mn = 83.76 − (47.56 × MnSediment) + (43.71 × pH)	0.013	0.140	0.346	1.937	0.084
Ni = −406.67 − (3649.16 × NiSediment) + (113.86 × pH)	0.754 ***	0.967	0.032	0.247	0.809
Pb = −462.69 − (1131.34 × PbSediment) + (89.60 × pH)	0.154 *	0.578	0.139	1.445	0.182
Zn = 318.75 − (41.81 × ZnSediment) − (29.64 × pH)	0.616 ***	0.939	0.041	0.336	0.743

R²: coefficient of determination; ME: model efficiency; MNAE: mean normalised average error; *: p < 0.05; **: p < 0.01; ***: p < 0.001. Student’s t-test was applied to determine the deviations of the estimated concentration of a metal in an organ from the measured metal in the same organ.

and

Table 4. Regression models between six metal concentrations in Typha elephantina organs (mg/kg) and sediment metals (mg/kg) and pH.

Equation	R2	ME	MNAE	Student’s t-Test
				t-Value	p
Roots
Cu = −141.88 + (3.56 × CuSediment) + (17.98 × pH)	0.232 **	0.770	0.078	1.204	0.254
Fe = 911.54 − (505.10 × FeSediment) + (214.12 × pH)	0.165 *	0.615	0.119	1.459	0.172
Mn = −448.87 + (86.00 × MnSediment) + (24.97 × pH)	0.443 ***	0.888	0.058	0.644	0.533
Ni = −1008.38 − (273.78 × NiSediment) + (138.09 × pH)	0.126 *	0.342	0.165	1.912	0.082
Pb = −892.57 − (148.37 × PbSediment) + (121.77 × pH)	0.115	0.221	0.201	2.136	0.061
Zn = −241.09 + (136.91 × ZnSediment) + (18.63 × pH)	0.623 ***	0.949	0.036	0.144	0.889
Rhizomes
Cu = 126.50 + (7.02 × CuSediment) − (9.78 × pH)	0.816 ***	0.970	0.034	0.137	0.893
Fe = 12,064.24 + (715.89 × FeSediment) − (1404.54 × pH)	0.231 **	0.675	0.084	1.211	0.251
Mn = −391.29 + (137.53 × MnSediment) + (11.69 × pH)	0.266 **	0.830	0.071	0.915	0.384
Ni = −134.81 − (685.53 × NiSediment) + (36.91 × pH)	0.556 ***	0.905	0.044	0.288	0.779
Pb = −118.83 − (574.13 × PbSediment) + (34.67 × pH)	0.521 ***	0.899	0.055	0.430	0.675
Zn = −147.20 + (46.16 × ZnSediment) + (16.20 × pH)	0.364 ***	0.886	0.061	0.670	0.520
Peduncles
Cu = −34.96 − (27.66 × CuSediment) + (22.61 × pH)	0.931 ***	0.996	0.029	0.019	0.985
Fe = −6254.05 − (397.58 × FeSediment) + (970.87 × pH)	0.470 ***	0.895	0.057	0.493	0.634
Mn = −196.19 − (17.98 × MnSediment) + (41.48 × pH)	0.060	0.100	0.457	2.227	0.053
Ni = −1103.38 + (938.76 × NiSediment) + (127.71 × pH)	0.577 ***	0.926	0.043	0.280	0.785
Pb = −1043.30 + (753.06 × PbSediment) + (119.85 × pH)	0.556 ***	0.900	0.051	0.420	0.683
Zn = −72.47 − (17.71 × ZnSediment) + (11.99 × pH)	0.165 *	0.576	0.148	1.663	0.125
Flowers
Cu = −239.94 + (0.21 × CuSediment) + (30.70 × pH)	0.193 *	0.647	0.106	1.454	0.180
Fe = −4581.89 − (127.53 × FeSediment) + (661.52 × pH)	0.287 **	0.880	0.065	0.797	0.446
Mn = 562.43 − (149.02 × MnSediment) + (15.42 × pH)	0.247 **	0.775	0.074	1.147	0.281
Ni = −420.22 + (124.09 × NiSediment) + (52.72 × pH)	0.140 *	0.391	0.159	1.845	0.092
Pb = −391.59 + (100.75 × PbSediment) + (49.18 × pH)	0.136 *	0.386	0.161	1.848	0.092
Zn = −24.82 − (4.45 × ZnSediment) + (5.03 × pH)	0.284 **	0.850	0.067	0.830	0.424
Leaves
Cu = −6.84 − (1.47 × CuSediment) + (3.13 × pH)	0.262 **	0.794	0.074	0.924	0.375
Fe = −810.57 − (178.38 × FeSediment) + (240.85 × pH)	0.206 **	0.675	0.087	1.283	0.226
Mn = −297.27 + (1.14 × MnSediment) + (52.53 × pH)	0.124 *	0.297	0.167	1.995	0.077
Ni = −140.59 + (499.08 × NiSediment) + (18.78 × pH)	0.088	0.129	0.359	2.141	0.056
Pb = −158.73 + (570.43 × PbSediment) + (17.14 × pH)	0.161 *	0.430	0.151	1.828	0.095
Zn = −242.09 + (1.50 × ZnSediment) + (32.87 × pH)	0.113	0.167	0.239	2.141	0.061

R²: coefficient of determination; ME: model efficiency; MNAE: mean normalised average error; *: p < 0.05; **: p < 0.01; ***: p < 0.001. Student’s t-test was applied to determine the deviations of the estimated concentration of a metal in an organ from the measured metal in the same organ.

). A high R² was noted from several metals, together with a high ME and a low MNAE; these parameters were indicative of the goodness of the models’ fit. An additional model optimal performance indicator was the lack of statistical variation between the observed and estimated metal concentrations. The following data refer to the significant models. In T. domingensis, the spectrum of R² was between 0.154 and 0.921, i.e., for the Pb within the leaves and the Ni within the peduncles, respectively. In T. elephantina, the R² range lay between 0.124 and 0.931, i.e., for the Mn within the leaves and the Cu within the peduncles, respectively. In T. domingensis, the ME ranged from 0.578 to 0.982, i.e., for the Pb within the leaves and the Ni within the peduncles, respectively. The equivalent MNAE values were 0.139 and 0.008, respectively. In T. elephantina, the spectrum of ME was between 0.297 and 0.996, i.e., for the Mn within the leaves and the Cu within the peduncles, respectively. The comparable MNAE values were 0.167 and 0.029, respectively.

4. Discussion

4.1. Sediment Analysis

The chemical analysis of the sediment samples from the study area showed that the pH was slightly alkaline, which may limit metal solubility and availability [38]. It is worth noting that most of the investigated metals in the sediments are within the range of concentrations considered safe in soils by Allen [36], Kabata-Pendias [6], and Chiroma et al. [39].

4.2. Plant Metals

Within the two macrophytes investigated, the mean metal concentrations varied according to the organ and metal concerned. Previous publications have indicated that the metal levels within different plant species are diverse, even when they are grown in identical sediment conditions in a similar ecosystem [6,22]. These differences in the metal concentrations could depend on the growing stage of the plant, the plant’s physiological factors, and the plant’s regulation mechanisms for controlling the sequestration of metals in plant cells [40,41]. In the current work, in T. domingensis and T. elephantina, the metals tested were preferentially accrued in the below-ground biomass rather than in the above-ground plant organs. This is in keeping with observations of numerous emergent macrophytes, e.g., Phragmites australis [42], T. domingensis [12], Echinochloa stagnina [1,43], Arundo donax [44], and Vossia cuspidata [24,45]. The aquatic macrophyte is recognised as amassing significant concentrations of metals in its below-ground organs owing to its intrinsic ability to detoxify [46]. These data may reflect the formation of complexes between the metals and the sulfhydryl groups, which diminishes the transfer of metals into the growing shoots [47]. Additionally, phytochelatin synthesis has been demonstrated to entrap metals, thus giving rise to their below-ground organ levels [48]. An additional reason for the high levels of metal in the below-ground organs is that these are the initial areas of plant metal exposure (see [29]). Significant differences were recorded for the metal concentrations throughout the growing season of T. domingensis and T. elephantina (data not presented). Thus, in the current study, the higher CV values of some metals in the organs of two macrophytes may have been due to the fact that the samples were collected over one growing season, from February to September 2018, and any diversity in the concentrations of the metals became merged.

The mean Cu and Mn concentrations recorded for T. domingensis organs in this study were within the phytotoxic ranges. The mean Zn concentration was lower than the phytotoxic range, while the mean Fe, Ni, and Pb concentrations were higher than the phytotoxic range (Cu: 20–100 mg/kg; Ni: 10–100 mg/kg; Zn: 100–400 [6]; Mn: 50–500 [49]; Fe: 5–200 mg/kg; and Pb: 5 mg/kg [50]). The mean Cu, Mn, and Ni concentrations recorded for the T. elephantina organs were within the phytotoxic ranges. The mean Zn concentration was lower than the phytotoxic range, while the mean Fe and Pb concentrations were higher than the phytotoxic range. These data support earlier research that demonstrated the ability of Typha species to withstand growing conditions polluted with metals [51,52]. Remarkably high Pb concentrations were observed at the study sites (Taif is a popular summer resort) because they were exposed to high Pb pollution related to vehicle traffic [40,53]. Elevated Pb concentrations have also been documented by Rzymski et al. [35] in emergent macrophytes growing in a metropolitan pond in proximity to busy traffic routes. The elevated concentrations of Pb within the above-ground biomass of both T. domingensis and T. elephantina imply that the route of plant uptake is through particles alighting on the surface of the leaves [35].

4.3. Bioaccumulation and Translocation of Metals

The accumulation of metals in plants relies on convoluted mechanisms, which are reliant on a range of elements, e.g., the physicochemical attributes of the sediment, the plant species, the phenology and physiology, the biochemical make-up of the rhizosphere, the climate parameters, or the chelation properties of additional metals and metallic chemical speciation [54]. Computing the BAF is a straightforward quantitative technique which can be used to evaluate metal transfer from sediment to plant [55]. The parameter reflects the metal uptake efficacy of a particular species and its ability to accrue the metal within its parts. The higher the BAF, the greater the plant’s capacity for bioaccumulation [56]. A plant that could behave as a metal hyperaccumulator is potentially recognised by a BAF > 1.0 [57]. In the current research, a BAF > 1.0 for all six metals under investigation was obtained for both T. domingensis and T. elephantina. It is well established that the metal uptake capacities amongst different plant species are variable [58]. Comparison between varying species can add to the current comprehension of metal dynamics within wetland ecosystems. This research has demonstrated that metal accrual between the sediment and the plant roots was heterogeneous between all six metals and the two macrophytes under examination; this is likely to be a consequence of the varied uptake approaches and abilities to withstand various metals [59]. The transfer of metals between the various plant organs was also diverse between the two species, but within a less broad spectrum for the different metals. These data imply that the movement of metals from sediment to roots is predominantly related to the particular species and element, whereas transfer within the plant itself is more pertinent to the individual species [16]. Overall, Ni exhibited the largest BAF, followed by Pb, Fe, Zn, Mn, and Cu. This observation suggests that the uptake of the varying metals within the sediment is neither homogeneous nor concentration-dependent for all elements [60]. These findings are in keeping with results obtained from T. domingensis grown in Lake Burullus, Egypt, where the highest BAFs were recorded for Fe, Zn, and Pb, and the lowest BAFs were recorded for Cu, Co, and Ag [23].

The TFs gave an indication of the dissemination of metals between the below-ground and the above-ground organs of the two macrophytes investigated. As stated above, intrinsic metal transfer is essentially individual to the species [16]. Metal translocation from roots to shoots is influenced by their type, the physiological mechanisms within the plant, the water transfer, and the species [61]. Additional factors include the overall plant metal concentration, the relevant metal’s biochemical characteristics and the plant itself [62]. Earlier studies have demonstrated that above-ground metal accrual is heavily reliant on the sediment metal bioavailability, the individual species’ absorption capability, the translocation ability, and the functional utility of certain metals within enzyme complexes and biological and metabolic pathways [13,63]. Within the current study, compartmentalisation was noted. Specifically, the metal transfer between the roots and shoots could be limited with no evidence of a toxic impact; this may reflect the intrinsic ability for detoxification [6]. These data reflect a plethora of research indicating that higher metal concentrations are amassed in the below-ground biomass rather than in the plant shoots [12,24]. The movement of the metals under investigation within the macrophytes were notably diverse with respect to the different plant organs. The largest TFs for most of the metals were observed between the roots and rhizomes, while the lowest were between the roots and the reproductive organs. In the current study, Fe and Cu evidenced higher TFs as they are key plant micronutrients [64], and Ni demonstrated a low TF. The latter may reflect the low Ni concentration of the below-ground plant organs compared with the above-ground biomass [65]. Variations between TFs could be associated with engagements between the metals, reflecting opposing or mutually potentiating mechanisms [66]. These could influence metal uptake efficacy and their dissemination [67]. Bello et al. [68] have proposed that vegetation with an elevated BAF and a small TF would be suitable for phytostabilisation; plants appropriate for phytoextraction would exhibit BAF > 1.0 and TF > 1.0. Thus, phytostabilisation of a number of metals, and the phytoextraction of others, could potentially be performed by T. domingensis and T. elephantina.

4.4. Plant–Sediment Correlations

The linear correlation analysis demonstrated that the sediment levels and the concentrations within the various plant organs for most of the individual metals were associated in either a positive or a negative manner for both macrophytes, indicating the cumulative effect of T. domingensis and T. elephantina on sediment contamination and implying their possible utility for the bioaccumulation of these metals. Moreover, plants with metal concentrations strongly correlated with those in the sediment have been considered potential indicators of metal availability (see [29]). These data are comparable with additional studies which have noted that macrophyte metal concentration is enhanced with rising metal levels within the environment [12,23,48,53], but this relationship is not simple and likely differs with both species and metal (see [12]). Similar data related to the association between the metal concentration of the sediment and T. domingensis have been published in previous studies [12,23]. Moreover, this study shows that there are some insignificant correlations between the metal concentrations in the sediment and the relevant concentrations in the plants, thereby indicating that T. domingensis and T. elephantina did not universally absorb all metals. Thus, there is no concentration-reliant absorption for all metals [60]. Furthermore, the relationships between sediment and roots may infer the key part played by the sediment as the principal origin of metals for plants within wetland ecosystems. This is supported by earlier publications which demonstrated that the source of the rooted macrophyte metal uptake is predominantly the sediment [69,70].

4.5. Prediction Models

In order to establish whether the generated model has a good fit with the relevant application, it is important to conduct verification. Useful mathematical tools for this purpose include R², ME, MANE, and p-values [20,71]. The greater the R², the higher the quality of the model [72]. All the regression models presented in the current work were found to have a good fit for most of the metals examined within the various plant organs of the two investigated macrophytes. The following data refer to the significant models. In T. domingensis and T. elephantina, the respective percentage variations were as follows: Cu, 17.0–56.9% and 19.3–93.1%; Fe, 28.6–73.9% and 16.5–47.0%; Mn, 38.6–57.0% and 12.4–44.3%; Ni, 52.5–92.1 and 12.6–55.6%; Pb, 15.4–48.5% and 13.6–55.6%; and Zn, 22.2–61.6% and 16.5–62.3%. The ME values for T. domingensis and T. elephantina were 0.578–0.982 and 0.297–0.996, respectively. The equivalent MNAE values for T. domingensis and T. elephantina were low, i.e., 0.008–0.139 and 0.029–0.167, respectively. MNAE ≤ 0.50 indicates comparability between the predicted and the observed metal concentrations, indicating that a good fit is offered by the models [20].

According to our best knowledge, no prior study has been conducted to develop regression models for the prediction of metal accumulation in T. domingensis and T. elephantina grown in an arid climate. Comparing the data was therefore only feasible with previous studies pertaining to equivalent macrophytes (Table S4). In T. domingensis, located in Lake Burullus on Egypt’s Mediterranean coast, the R² for Pb was computed to be between 20.5% and 58.5% [23]. This was similar to that calculated for P. australis, i.e., 17.2–50.5%, also grown in Lake Burullus [42]. For V. cuspidata, situated in the River Nile in Egypt, the R² for Pb was found to be between 9.1% and 57.8% [24]. The R² in the native T. domingensis lay between 12.3% and 48.5%; for T. elephantina, the values were between 11.5% and 55.6%. Possible reasons for the different R² values in the metal models in our study and previous studies could be the unequal numbers of observations across studies; the number of parameters used in the model construction; the different analytical methods for measuring metal concentrations in plant organs; the different methods for pH measurement; a large variety of environmental conditions for plant growth; the different stages of plant growth; the pollution levels; the physico-chemical characteristics of the soil at the sampling sites; the soil texture; the soil microbial activity; and the influence of soil type [6,40,73].

5. Conclusions

With the present study, we showed that the accumulation of metals within the studied species was relatively high and varied in the different organs of T. domingensis and T. elephantina. On the one hand, we showed that T. domingensis and T. elephantina share some general characteristics (e.g., metal compartmentalisation between organs and bulk metal concentrations in the roots), and on the other hand, we found that these species tended to have a specific response in terms of the translocation and bioaccumulation of metals in sediments. We identified a significant correlation between the concentrations of certain metals in T. domingensis and T. elephantina and the concentrations in the sediment; thus, T. domingensis and T. elephantina potentially represent a valuable tool for the bioaccumulation of metal contamination. Based on the comparative analysis of metal concentration, BAF, and TF in two studied species, it can be seen that T. domingensis and T. elephantina have sufficient phytoextraction ability for some metals and showed remarkable phytostabilisation abilities for other metals. Future studies are necessary to investigate the mechanisms of the phytostabilisation and phytoextraction of these metals. We developed new linear regression models to predict several sediment–plant metal transfers. The main advantage of these models is that they are time- and cost-effective, reducing the size of the data sets needed to identify the statistically significant sediment properties. While we endeavored to ensure that the model input data was of the highest quality possible, there undoubtedly remain uncertain results from the few sites selected. Due to the limitation of available sites, the T. domingensis and T. elephantina populations were identified only in six available sampling locations, three per individual species. For these three sampling locations for each macrophyte to make a community that is representative of 1036 square kilometres poses a particular challenge. Thus, the careful use of these models within a range of site-specific parameters is necessary. The quality of models depends on the degree of variability explained. Therefore, our recommendation is to use models with an R² higher than 40%.