Soilless Production of Lettuce (Lactuca sativa) in the Atacama Desert Using Fog Water: Water Quality and Produce Mineral Composition

Abstract

Soilless vegetable production in the Atacama Desert of Northern Chile is spreading since it is perceived as an alternative that requires much less water than open field soil production. However, strong competition between mining and urban use for human population consumption exists, forcing growers to use alternative water sources. Fog is commonly present in the coastal areas of Northern Chile; however, little information exists with regards to its chemical composition and the effect on nutrient quality of the produce. To address this knowledge gap, a set of experiments was carried out in Chañaral, a small town located in the Atacama Desert of Northern Chile. There, a 200 m² greenhouse equipped with twenty deep flow pools was used in two consecutive growing cycles. Water for the mixing of the nutrient solution was collected from the fog using fog-catchers and later stored in 2000-L tanks. Fog water quality (electrical conductivity, pH and mineral content) was monitored directly from the storage tanks. Two types of lettuce, green butterhead and red oak leaf, were compared on their yield and accumulation of nutrients and heavy metals. The results indicate that fog water is of good quality for soilless production, with an electrical conductivity value of 0.65 ± 0.18 and low content of heavy metals. Plants’ heavy metal accumulation is below the recommendation of Food and Agriculture Organization and World Health Organization. Fog water presents as a viable water source for soilless production in Northern Chile.

1. Introduction

The Atacama Desert in Northern Chile is considered one of the driest places in the world, with mean annual precipitation as low as 0.8 mm in locations such as Iquique [1]. The extreme aridity limits agricultural production to small valleys where surface or groundwater can be found; however, due to the reduction in precipitation in the Andes mountains during the last decade, water is becoming less available year after year, forcing the use of alternative water sources and innovative cultivation methods [2]. Every year, more growers are becoming attracted to soilless production, since they understand that considerably less water is required for vegetable production in comparison to open field soil production [3]. Growers in Northern Chile mostly rely on tap water for mixing the nutrient solution, but the electrical conductivity (EC) of this water commonly exceeds 1.5 dS m⁻¹, which impairs produce quality [4]. Using low-volume reverse osmosis systems is becoming the preferable method for water treatment in this area, with the consequent cost in energy and the disposal of brine.

An underused water resource is fog. Fog is commonly present in the coastal areas of Northern Chile, since it forms by the condensation of moisture within the cold marine layer that is trapped under the subsidence of warm air masses related to the Hadley cell [5]. The result is a low-altitude stratocumulus cloud that interacts with the coastal mountains next to the seashore [1]. Then, using fog catchers (low-cost devices built with a rigid frame that holds a 32 m² mesh through which fog flows capturing and condensing water droplets, Figure 1) an average of 100 L per day can be harvested [1]. These devices are in use in several locations across the globe, mainly in arid or semi-arid region like Oman, Namibia, South Africa, or Saudi Arabia [6].

Little information is available with regards to the chemical quality of fog water, especially on the EC values. However, most concerns arise with regards to the presence of heavy metals in this water source, since they can accumulate in the produce, becoming a threat for human health [7].

In a previous publication [1], our research group presented the results of water yield from fog collectors located along the Atacama Desert. However, knowledge gaps in relation to water quality remained unanswered. Here, we present the preliminary results of water quality analysis and heavy metal accumulation on two lettuce genotypes cultivated under greenhouse conditions in the Atacama Desert using only water collected from fog.

2. Materials and Methods

The study was conducted in Northern Chile in a location named Falda Verde (26°17′ S/70°37′ W). The location is known as a wasteland due to the severe accumulation of mining wastes on the banks of the Salado River, including high concentrations of lead, copper, and arsenic [8]. A 200 m² passive plastic greenhouse was set up where the ventilation was naturally occurring by opening the sides and upper vents (Figure 2). Ten independent deep flow hydroponic pools each with a surface of 2.7 m² were installed and connected to an air pump with a flow rate of 70 L min⁻¹. Energy for the air pumps was obtained from photovoltaic panels installed outside the greenhouse. On top of the hills, twelve 32 m² fog collectors were installed and they supplied water to a set of five 2000 L tanks. Water was drawn to the deep flow pools by means of hoses by gravity.

Water samples were taken before the start of the experiment and were sent to a commercial laboratory to determine pH, EC and mineral content. With the results, the amount of fertilizers required to mix a half-strength Hoagland solution (7.5 mM N, 0.5 mM P, 3 mM K, 2.5 mM Ca, 1 mM Mg, and 1 mM S) was determined. Nutrient solution pH was adjusted to 5.8 using phosphoric acid, and the initial EC value was 2.32 dS m⁻¹.

Two lettuce varieties, a green butterhead (Fabietto, Rijk Zwaan) and a red oak leaf (Mondaï, Rijk Zwaan), were sown in plastic trays containing peat and perlite at a 9:1 rate. The trays were irrigated daily only with fog water. Once the plants reached the four true leaves stage, they were transplanted into the floating systems at a density of 30 plants per m². Three weeks after the transplant, the fresh weight (FW) was recorded, and the leaf samples were taken to a commercial laboratory for the determination of mineral content. The results are expressed in a dry weight (DW) basis and the conversion to FW basis was calculated by dividing for the percentage of dry biomass. This experiment was replicated twice. The first experiment ran from 2 May 2024 to 23 May 2024, with average minimum and maximum temperatures of 12.61 ± 0.32 and 28.74 ± 0.80 °C, respectively. Mean daily light integral reached 11.31 ± 0.69 MJ m⁻². The second experiment took place between 13 June and 4 July 2024, reaching an average mininmum temperature of 10.53 ± 0.21 °C, an average maximum temperature of 2.56 ± 0.54 °C, and a mean daily light integral of 6.42 ± 0.17 MJ m⁻².

Differences in nutrients and heavy metals in leaf dry biomass between both genotypes (treatments) were analyzed by ANOVA and mean separation was carried out by LSD.

3. Results

Fog water is slightly acidic with a pH of 6.74 ± 0.54, and it presents a low content of dissolved solids expressed as an EC of 0.65 ± 0.18 (

Table 1. Chemical characterization of fog water before the addition of fertilizers. Values represent average ± standard error of four replicates.

Parameter	Units	Value	Recommended Maximum Value [8]
pH		6.74 ± 0.54	6.5–8.4
EC	dS m−1	0.65 ± 0.18	0.7
Calcium	mEq L−1	0.43 ± 0.07	<20.0 *
Magnesium	mEq L−1	0.91 ± 0.30	<5.0 *
Potassium	mEq L−1	0.08 ± 0.03	<0.05 *
Sodium	mEq L−1	3.60 ± 1.10	3.0
Chloride	mEq L−1	4.50 ± 1.30	3.0
Sulphate	mEq L−1	0.72 ± 0.20	<20.0 *
Bicarbonate	mEq L−1	0.28 ± 0.03	1.5
Ammonium	mg L−1	1.80 ± 0.20	<5.0 *
Nitrate	mg L−1	5.60 ± 0.90	<10.0 *
Aluminum	mg L−1	0.12 ± 0.02	5.0
Arsenic	mg L−1	0.04 ± 0.01	0.10
Boron	mg L−1	0.15 ± 0.08	0.7
Cadmium	mg L−1	<0.01	0.01
Copper	mg L−1	1.50 ± 0.08	0.20
Iron	mg L−1	0.09 ± 0.05	5.0
Manganese	mg L−1	0.04 ± 0.00	0.20
Mercury	mg L−1	<0.01	0.02 **
Molybdenum	mg L−1	<0.01	0.01
Lead	mg L−1	<0.01	5.0
Zinc	mg L−1	0.01	2.0

* Common values found in irrigation water [8]. ** Maximum recommendation value according to the Ministry of Environment, Canada [9].

). The highest mineral content found corresponds to chloride followed by sodium, where the sum of both elements represents 75% of the total dissolved solids present in the water samples. Small concentrations of arsenic were detected but the heavy metal in higher concentration was copper. The concentration of cadmium, mercury, molybdenum, and lead were below the equipment detection limit.

After three weeks of growth in the deep flow system, plants reached an average fresh weight of 68.18 ± 4.01 g plant⁻¹ and 55.25 ± 5.14 g plant⁻¹ for the green and red varieties, respectively. Both varieties presented a similar mineral nutrient composition, except for the content of potassium which was higher in the red oak leaf variety and calcium which was higher in the green butterhead genotype. Both genotypes presented a significant accumulation of chloride and sodium (

Table 2. Leaf mineral nutrient composition. Values represent average ± standard error of ten replicates. Different letters in the same row denote significant differences (p < 0.05) between both genotypes.

Nutrient	Green Leaf	Red Leaf
Nitrogen (%)	3.03 ± 1.01 a	3.25 ± 0.56 a
Phosphorus (%)	0.50 ± 0.04 a	0.45 ± 0.01 a
Potassium (%)	1.41 ± 0.09 b	2.08 ± 0.29 a
Calcium (%)	0.49 ± 0.02 a	0.36 ± 0.03 b
Magnesium (%)	0.27 ± 0.01 a	0.26 ± 0.01 a
Sulfur (%)	0.08 ± 0.00 a	0.08 ± 0.00 a
Chloride (%)	3.33 ± 0.14 a	2.83 ± 0.56 a
Sodium (%)	1.05 ± 0.23 a	0.96 ± 0.02 a
Iron (mg kg−1 DW)	129 ± 27 a	137 ± 33 a
Boron (mg kg−1 DW)	29 ± 7 a	44 ± 12 a
Manganese (mg kg−1 DW)	79 ± 48 a	76 ± 43 a
Molybdenum (mg kg−1 DW)	0.21 ± 0.06 a	0.15 ± 0.02 a

).

With regards to heavy metals, the largest content corresponds to manganese, zinc, and copper, with no differences between both varieties (

Table 3. Leaf heavy metal content. Values represent average ± standard error of five replicates. Different letters in the same row denote significant differences (p < 0.05) between genotypes analyzed on a dry weight basis.

Heavy Metal	Mineral Content (mg kg−1 DW)
	Green Leaf	Red Leaf
Zinc	49 ± 23 a	49 ± 24 a
Copper	43 ± 24 a	27 ± 7 a
Arsenic	0.40 ± 0.00 a	0.44 ± 0.03 a
Cadmium	0.10 ± 0.00 a	0.10 ± 0.01 a
Mercury	0.02 ± 0.00 a	0.02 ± 0.00 a
Lead	0.47 ± 0.02 a	0.31 ± 0.01 b

).

4. Discussion

Fog water collected in the experimental site presents an average EC value of 0.65 ± 0.18 dS m⁻¹, which is mainly attributed to the content of sodium and chloride (Table 1). Chloride accounts for 49% of total dissolved solids (TDS) present in the water samples while sodium represents 26% followed by sulphate, bicarbonate, magnesium, calcium, nitrate, potassium, and ammonium, which account for 11%, 5%, 3%, 3%, 2% 1%, and 1%, respectively. This is in agreement with the results presented in the literature regarding the study of aerosols formed from the sea surface as the result of water evaporating directly from the sea [9,10,11]. However, the concentration of these compounds may vary throughout the year, mainly because of dust deposition on the fog collectors, as reported by Shanyengana et al. [12], who described variations between 70 and 1000 mg TDS L⁻¹, equivalent to EC values of 0.1 and 1.5 dS m⁻¹, in fog water samples collected in the Namib Desert. In our first experiment, the water EC average value was 0.83 ± 0.11 dS m⁻¹, while in the second experiment, the value dropped to 0.48 ± 0.15. Therefore, further research is required to understand the variations in TDS throughout the year and make the required adjustments in the fertilizer dosage for the nutrient solution mix.

Although the EC value is above the 0.5 dS m⁻¹ threshold recommended for hydroponics [4], fog water contains significant amounts of essential mineral elements for crops, such as magnesium, calcium, and sulfur (Table 1). Therefore, the supply of these elements needs to be adjusted in the fertilizer formula in order to keep the EC within an adequate range. Our results indicate that fog water supplies 8% of Ca, 45% of Mg, and 36% of S required in a half-strength Hoagland’s solution [13].

The boron content in fog water is remarkably low (0.15 ppm), which is an advantage in comparison to other water sources available in the Atacama Desert. For instance, Pincetti-Zuñiga et al. [14] reported concentrations of above 2.5 mg L⁻¹ in 64% of 90 samples of the surface and drinking water collected in Northern Chile (Azapa and Lluta valleys) and Figueroa et al. [15] reported concentrations of between 0.6 and 16 mg L⁻¹ in the irrigation water used for olive production in the same area. Therefore, boron is commonly found in water sources from Northern Chile beyond the recommended standard of 0.7 mg L⁻¹ [16]. Concentrations beyond this value result in yield reduction because of leaf damage and the generation of reactive oxygen species [17].

With regards to the content of heavy metals in the water, the most abundant element is copper with an average value around 1.5 mg L⁻¹. Thus, the copper content in fog water is above the recommendation of 0.2 mg L⁻¹ for irrigation water, according to the U.S. Environmental Protection Agency [18]. However, copper accumulation in the plant’s biomass depends upon plant species, and our results indicate that lettuce leaves present a concentration of between 27 and 43 mg kg⁻¹ DW, equivalent to 1.35–2.15 mg Cu per kg of fresh biomass. These values are below the maximum permissible value (73 mg Cu kg⁻¹ FW) recommended by FAO/WHO [19,20].

Although no cadmium, mercury or lead were detected in the water samples, they were present in the leaf biomass. Then, it is necessary to identify the source of these elements, since they could be present in the water at extremely low concentrations and accumulate in the leaves due to plant evapotranspiration [21] or can deposit on the leaves by air currents carrying them from the sediments on the coast [22]. Nonetheless, the content of cadmium in the biomass is below the FAO/WHO recommended threshold (4.0 mg kg⁻¹ DW), as well as lead (6.0 mg kg⁻¹ DW) [23].

The presence of arsenic in the leaves of our experiment is around 0.4 mg kg⁻¹ DW equivalent to 0.02 mg kg⁻¹ FW. This value is below the recommended maximum of 0.1 mg kg⁻¹ FW set by the European Union [24]. In the case of mercury, the only recommendation found in the literature comes from the China Food and Drug Administration, providing a maximum level of 0.01 mg kg⁻¹ FW, equivalent to 0.2 mg kg⁻¹ DW [25]. Then, the content found in our experiment is considered permissible.

Finally, the yield achieved in both cultivation cycles demonstrates the suitability of soilless production using fog water. Our results indicate fresh biomass accumulation of 68.18 ± 4.01 and 55.25 ± 5.14 g plant⁻¹ in the green and red varieties, respectively. These values are in agreement with those reported in the literature under similar temperature and radiation conditions [26,27]. Then, fog water presents as a viable water source that can be easily obtained in the coastal areas of the Atacama Desert at low cost.

5. Conclusions

Our results indicate that water collected from marine fog in the Atacama Desert is suitable for use in lettuce soilless production. It presents a low EC value and low concentration of heavy metals. Good yield can be achieved using this water source for hydroponic lettuce, and the accumulation of heavy metals is within the safe range defined by the international health authorities. Further studies are required to test its suitability for other vegetable crops grown under soilless methods. Overall, fog water collection presents as a viable alternative for agricultural production in arid lands.