1 Introduction

Worldwide, concerns about food security and food system sustainability and increasing dietary awareness suggest the need to advance new and alternative food sources. Naturally, most of the global agricultural production is concentrated in fertile regions. However, because those regions are limited in their capacity to increase food production, marginal areas, and even drylands, desert areas have the potential to enhance their contribution to meeting the growing demand for food.

Globally, drylands now cover about 40% of the Earth’s land surface and are home to approximately one-third of the world’s population (UN 2011). While most people tend to live in a very limited climate subset of the average temperature of about 13°C, up to 3 billion people will experience a much warmer climate over the coming decades. Therefore, alongside climate mitigation, the highest priority is to enhance human development in those regions (Xu et al. 2020).

Expanding agriculture into less suitable land may be technically possible in many instances. Nevertheless, it would likely imply lower yields, require additional inputs (e.g., fertilizers, irrigation), or necessitate additional investment in infrastructure that would increase production costs (FAO 2019a). Attempts to overcome water scarcity can result in a tradeoff between impact categories like GHG emissions (Page et al. 2012). Further, it may increase pressure on environmental structures and functions to undermine the system’s sustainability. It follows that the suitability and sustainability of various crops in arid conditions should be examined (e.g., Aune et al. 2017; Mupambwa et al. 2019; Xu et al. 2020), including an analysis of the environmental implications of existing systems, improving alternative and more efficient methods and crops with potential to take advantage of the desert’s benefits (e.g., Weselek et al. 2019), despite its limitations.

Most of those studies embracing a systematic approach using methods such as material flow analysis, life cycle assessment (LCA), and footprint analysis were conducted in Europe and North America, not in desert areas (Cerutti et al. 2014), although they are becoming increasingly common for a wide variety of products and processes. Also, such studies focusing on fruit are still quite limited (e.g., Ingrao et al. 2015; Ingwersen 2012), perhaps due to the various methodological challenges of such analyses (Basset-Mens et al. 2016; Bessou et al. 2013; Cerutti et al. 2014). The date-palm fruit (Phoenix dactylifera) has received limited attention from environmental researchers. Clune et al. (2017) reviewed a wide range of agriculture-related LCA studies published in recent decades and found only one concerning dates. To the best of our knowledge, none have yet focused on Medjool dates, a highly popular variety of palms (Housler et al. 2017; Robinson et al. 2012) on which this paper focuses.

Nevertheless, the number of studies focusing on environmental aspects of agriculture in arid and semi-arid regions is slowly increasing (e.g., Aldababseh et al. 2018; Aune et al. 2017; **e et al. 2018), yet very few are conducted in a hyper-arid climate (e.g., Mupambwa et al. 2019). Among the arid research, more date palm–related research is being published, usually focusing on the economic perspective (e.g., Kotagama et al. 2014) or different, specific environmental aspects, such as the use of water, fertilizers, pesticides, and implications on land and waste management. Here, too, most of them were done in arid and semi-arid climates (e.g., Al Hinai and Jayasuriya 2021; Al-Mulla and Al-Gheilani 2017; Al-Oqla and Sapuan 2014; Chandrasekaran and Bahkali 2013; Shabani et al. 2012; Zhong et al. 2020) and less in hyper-arid regions (e.g., Al-Muaini et al. 2019; Shani et al. 2005; Sperling et al. 2014; Tripler et al. 2007). However, exploring the sustainability of fruit in the context of its geographic location requires embracing a systematic approach that considers a range of direct and indirect environmental interactions throughout its life cycle, from the supply of raw materials, through the cultivation stage, sorting, and packing, until it reaches the point of consumption.

Recently, Hesampour et al. (2018) analyzed dates grown in the Khuzestan Province of Iran. Their findings match other fruit-plantation studies, emphasizing the high contribution of fertilizers (production, transportation, and application) and energy use to impact categories and, in hot climates, water (Al Hinai and Jayasuriya 2021; Al-Muaini et al. 2019; Al-Mulla and Al-Gheilani 2017; Bamber et al. 2020; Basset-Mens et al. 2016; Pergola et al. 2013). However, that research (Hesampour et al. 2018) has several limitations, including scope (focused only on the cultivation phase) and missing information (e.g., the ages of the palm trees, yields, and more). The current study is one of the first to systematically assess the cultivation of any kind of fruit in an arid area.

The date palm fruit is among the most popular crops native to arid areas and has been cultivated in the Middle East and North Africa for more than 5000 years (FAO 2019b; Tengberg 2012). It originated in ancient Mesopotamia (today’s Iraq) and have spread to many regions around the world, between latitudes 10–39° north and 5–51° south (Agoudjil et al. 2011; Tengberg 2012). The date palm is an evergreen palm adapted for hot climates and tolerant of saline water. These unique characteristics allow it to be a food source even in harsh environmental conditions, such as for rural communities in arid regions (Chao and Krueger 2007; FAO 2019b). Further, dates are an important part of a long heritage and the culture of many communities and religions across the globe (FAO 2019b; Tengberg 2012).

Dates are considered an excellent source of essential nutritional components (Adeel et al. 2018; Al-Farsi and Lee 2008; Al-Shahib and Marshall 2003; FAO 2019b), and have rare, proven medical properties (FAO 2019b;Yasin et al. 2015) that are recognized by both traditional and modern medicine. Therefore, it is understandable that some consider them an ideal food (Adeel et al. 2018) and even a future “super-food” (Al-Shahib and Marshall 2003). Increased awareness of dates’ dietary and health advantages has contributed to a growing global demand for the fruit, leading to increased production and trade. The overall quantity of dates grown has increased by more than 66% in the last decade (Housler et al. 2017). More than 8 million tons of dates are grown annually on over 100 million palm trees, mostly in Egypt, Iran, Saudi Arabia, Morocco, and Algeria (FAO 2020).

This paper embraces a “footprint family” approach to advance a systematic biophysical analysis of Medjool dates grown in the Central Arava region in Israel. The footprint family has been described as “a set of indicators […] able to track human pressures on the surrounding environment, where pressure is defined as the appropriation of biological natural resources and CO2 uptake, emission of GHG’s, and consumption and pollution of global freshwater resources” (Galli et al. 2012). It is often used to identify and assess environmental loads associated with a process, product, or system and allows for examining potential biophysical tradeoffs from proposed policy and other measures (Galli et al. 2012; Giljum et al. 2011; Steen-Olsen et al. 2012).

The analysis covers various direct and indirect environmental interactions (e.g., materials, land, water, solid waste, and GHG emissions) of dates grown in the region studied. It quantifies and analyzes selected inputs and byproducts and a wide range of other factors, from the pre-cultivation stage until the dates are delivered to a European gateway port. It identifies hotspots in the local date production system as a step toward suggesting a primary baseline and potential starting point for making future improvements and moving toward a more sustainable system.

Israeli agricultural practices are well-regarded globally, thanks to many innovative solutions for irrigation, fertilizers, and plant protection, making them highly productive (Almukhambetova et al. 2017; Douthwaite and Hoffecker 2017). It follows those insights from the studied region that can be relevant to other arid regions.

The Arava region is the section of the Great Rift Valley stretching from the Dead Sea to the Red Sea, along the Israeli-Jordanian border. It is the driest region in Israel, with an average annual rainfall of only 42 mm, making it a hyper-arid zone. The region’s water supply system is based on wells drilled in a series of hydrological basins drawing on the regional aquifers (Israel Water Authority 2009). Despite its small size, Israel is among the top growers and a leading exporter of Medjool date (Ordines 2000); indeed, approximately 70% of the worlds Medjool date exports originate in Israel (Housler et al. 2017).

In 2019, for example, Israel produced roughly 43,000 tons of dates. In recent years, there has been a sharp increase in the number of date plantations in Israel generally, and specifically in the region under consideration. These mainly grow the Medjool variety, which is currently grown by some 80% of the date plantations in Israel. In the Central Arava, there are now more than 750 ha of Medjool date plantations, including about 224,500 palms of various ages, producing more than 6000 tons of fruit annually, which is about 14% of the country’s Medjool production (The Portal of Nature, Agriculture and Environment 2017). The Medjool date has become one of the region’s major crops.

This research focused on seven jointly owned date plantations, covering an area of 300 ha all together, with approximately 32,300 palm trees that produce about 2500 tons of Medjool fruit annually, about 40% of the studied region’s date production (Barkan 2017). While the cultivation methods are varied and influenced by specific physical conditions and decisions of each farmer, the plantations included in the research have several characteristics in common; a few main steps are presented in Fig. 1.

Fig. 1
figure 1

a–d Example of some of the main steps in date production. Photographs by the authors.

Full size image

The date plantations contain palm trees (males and females of different ages, see Fig. 3) planted about 9 m apart, usually 12 palms/1000 m2. Every age group requires different inputs and produces different yields; male palms and immature females produce no fruit, and yield increases with the age of the palms. The palm trees require a great deal of water, which is piped to each individually. Most of the palms, from the age of 4 onwards, are treated with a plant growth regulator hormone (Uniconazole) to slow their upward growth and reduce the need for hydraulic platform machines (Fig. 1a) that are used for most of the tasks required during the cultivation phase, such as picking and pruning. Pests and weeds are usually treated only in response to problems, not as a preventive action. Fertilizers are applied via fertigation (by the irrigation system). After the harvest, the palm bunches, thorns, and excessive fronds are trimmed and removed (Fig. 1b).

When the female palms bloom, a pollination machine is used to pollinate them with pollen grains produced by male palms. A few weeks later, the spikelets and fruit are thinned. The fruit bunches are then wrapped in plastic net bags (typically made of a flexible wire, 18 × 20 mesh, Fig. 1c) and fastened with plastic string. When the fruit ripens, it is picked, placed in multiple-use plastic crates, and transported to a packing house by tractors. The fruits are placed in refrigeration for 24–48 h and then machine-sorted by size and quality (Fig. 1d). The dates are packed in disposable cardboard crates. The packages are wrapped in shrink-wrap plastic, stacked on wooden pallets (1 ton per pallet), and covered top and bottom with cardboard covers, with additional cardboard sheets that are inserted between the crates to ease the pressure from their weight. They are stored in freezers at −18°C until transported, while refrigerated, by truck to the Port of Ashdod and then by sea, often to a European market.

The current study uses primary, original data gathered locally from farmers in the study region. The research enriches plantation study methodologies by using two functional units (FU) and by considering the various types and ages of the palms on each plantation. It also offers a spatial analysis of each palm age group across seven plantations and seeks the sources of differences and similarities. Therefore, the analysis presented in this paper facilitates identifying major hotspots in the date production system under consideration, which can be used as a starting point for future improvements toward a more sustainable system. It can also narrow the gap in knowledge about the date production system and be a primary foundation for future studies in other desert areas.

2 Material and methods

2.1 System’s boundaries

The study examines the materials (MF), land (LF), water (WFblue), and energy footprints, as well as solid waste (SWF) and carbon footprints (CF), throughout the life cycle of one metric ton of marketable Medjool dates (mass FU) and a single date palm (palm FU), from cradle to port in key export destinations.

The research includes three main stages: (1) The cultivation and the plantation stages; (2) sorting and packaging; and (3) transportation. As illustrated in Fig. 2, a series of inputs and byproducts were analyzed for each stage separately.

Fig. 2
figure 2

Research boundaries.

Full size image

Like previous studies (e.g., Giudice et al. 2013; Ingrao et al. 2015; Mohamad et al. 2014), this research did not consider human power and production of the date shoots. The latter is mainly an on-site process at the plantation itself, and the data available was insufficient to separate the mother-tree’s inputs from those used by the shoots. Also, like other fruit-related studies (Goossens et al. 2017; Milà i Canals et al. 2004; Sanjuán et al. 2004), this study did not differentiate between dates of differing quality, nor did it analyze the farm infrastructure, such as buildings, or production of the machinery.

2.2 Data collection

The primary data was site-specific data sourced from farmers and packinghouse managers. Other stakeholders in the region were consulted regarding plant protection, fertilizers, and irrigation methods. The following production operations were included in the analysis: fertilization, plant protection, irrigation, pruning, pollinating, harvesting, sorting, packing, and transportation. Primary information was collected via in-depth surveys using a specially designed checklist (as in Ingrao et al. 2015) concerning input material types and amounts. This data is regarded as high-quality data; it was taken directly from the stakeholders (Page et al. 2012). Also, for each operation considered, each product’s active ingredient, the amount of energy and fuel consumed, and the amount of waste produced were calculated and used to calculate the CF. The data included the quantities and types of water and main materials (plastic, PVC hose or pipes, pesticides, fertilization, wood, etc.; see Appendix Table 2 for details), the location of the suppliers, and the type of machines and their annual use of energy. The data also included the size of the plantation, quantities, and ages of palm trees, yield, solid waste, and waste management methods. The packing house managers reported on the type and size of the most common plastics and boxes (which were later measured and weighed), energy and water consumption, quantities of packed dates, and organic and inorganic waste. The information included transportation means and routes and the location of the main customers. Secondary data was collected from local authority publications and professional literature on the subject.

2.3 Impact assessment

The data collected was scaled into the footprints of 1 ton of dates. This facilitated estimating five impact categories: (1) The material footprint (kg/FU), which accounts for inputs over the life span of the material (the material input was divided into its years of usage, see Appendix Table 3). (2) The land footprint (m2/FU) was calculated based on yield, i.e., the area needed to produce 1 ton of marketable dates from the whole plantation, and young or mature palm trees. (3) The “water footprint” represents the direct and indirect volume of water needed to produce a good and can uncover hotspots associated with the water use (Hoekstra et al. 2011). Recently, Fridman et al. (2021) found that methodologies measuring WF typically relate to rainfall (green water) and fresh irrigation water (blue water) while ignoring other water sources. In a hyperarid climate, the negligible annual precipitation was irrelevant for the irrigation of crops (Al-Muaini et al. 2019). Therefore, in the current research, the term water footprint refers is WFblue, i.e., brackish and desalinated water (Fridman et al. 2021). WFblue (m3/FU) was measured not only to determine the water intake but also to calculate the amount of electricity used for water production (using information from the Israel Water Authority 2009) and estimate the amount of material needed to produce that energy, including coal, gas, and other fuels, as well as the related GHG emissions (Israel Energy Company 2021). Based on the input data required for calculating the above footprint components, the research also calculated (4) the carbon footprint (kgCO2eq/FU) throughout the life cycle of the materials (i.e., production, use, transportation, and disposal), and finally (5) the solid waste footprint (kg/FU).

The farmers provided information about the type of commercial fertilizers, pesticides, herbicides, and fungicides they use and the amount they apply per palm/year or per 1000 m2/year. The active components in the products, namely nitrogen (N), phosphorus (P), and potassium (K), were determined using information published by the manufacturers, under the guidance of the local guide from the Israeli Minister of Agriculture (Fridman, personal communication, 24 June 2019). Emissions resulting from fertilizer use were calculated based on the amounts NPK for each FU, for both production and application of the material, according to the relevant emissions factor (see Appendix Table 3), and from the transportation of the input (by mass) to the plantation. Mass was also used to evaluate each FU’s input quantity (in kg). The amount of fuel required for land transport in Israel was calculated according to the data from the Israel Central Bureau of Statistics (2016), based on the average distance from the suppliers to the region and from the region to the Port of Ashdod. The estimations included the round trip of the empty trucks and whether the cargo was refrigerated or not. The fuel required for ship** to the destination ports was evaluated using data from the International Energy Agency (IEA 2017) and the distance to the destinations (Portworld 2016). Organic and inorganic waste was estimated by the weight of the remaining raw materials and biomass for each functional unit for 1 year. Organic waste includes the removed parts of the palms, especially the fronds: about 20 fronds are removed yearly from a FY palm, each weighing about 3 kg (Fig. 1a). The latter biomass is mulched and spread over the soil as a protective layer or disposed of in landfills. It also includes the unmarketable fruit; about half are being disposed of due to visual parameters, such as size, weight, and color. Others are not suitable for human consumption due to damages like the ones caused by pests or birds.

The inorganic waste is collected, according to Israeli regulations, in designated sites where it is sorted and later transported by an authorized, external body to be recycled or other processed using a permitted waste-management method. The research used the relevant emission factor (for recycling, mulching, or landfill disposal) to estimate the relevant impact of each method up to the gate of the recycling sites. The palms’ end-of-life destiny was not considered, mainly because the local plantations are relatively young (less than 50 years) and only very recently have procedures for removing some old or crooked palms begun. Currently, the removed palms are piling up on the plantations.

Great effort was invested in finding local, up-to-date emission factors. If they could not be obtained, the emission factors were taken from international guides, such as DEFRA (2016), or academic literature (specific conversion factors and data sources are presented in Appendix Table 3).

2.4 Analytical process

The methodological challenges of systematic environmental analysis in the fruit sector are acknowledged in the literature (e.g., Al Hinai and Jayasuriya 2021; Cerutti et al. 2014; Goossens et al. 2017) due to the unique characteristics of the plantations. Unlike annual crops, fruit trees (including dates) go through different stages of life: seedlings, young, and mature palms (e.g., Mohamad et al. 2014). Sometimes, as in the Central Arava, the plantation includes palms at various stages as it operates and evolves over many years. Palms in each stage have different yields (Fig. 3) and require different amounts of inputs per FU. The aging, mature palms in the region studied are cut down at around the age of 30–40 years because of their height mainly due to technical reasons and Health and Safety regulations, since the palms’ height prevents an easy treatment, such as pollination and harvesting (Robinson et al. 2012), and much more expensive machinery is then needed.

Fig. 3
figure 3

The studied palms by type, age, and yield range per palm. Photographs by the authors.

Full size image

In Oman, Kotagama et al. (2014) estimated that economically, the best time to remove the old palms and replant the dates is when the palms are 50–55 years old. In addition, every plantation has seedlings and male palms, which are both crucial for sustaining their long-term fertility. It follows that analyzing the environmental interactions of any unit of dates requires embracing a diversity of perspectives that reflect the system’s structure, as presented in Fig. 3.

Therefore, two FUs’ mass (ton) and a single palm were used. First, the analysis examined the production of 1 ton of dates from each plantation from a sustainability perspective that acknowledges the plantations as living, dynamic systems. For a plantation to be viable, it must have (or import) both young and male palm trees and fruit-bearing palms. Long-term fruit production necessitates the inclusion of all these components in the analysis. Hence, each studied plantation was examined by the mass FU concerning its complete mix of female and male palms of different ages (henceforth: “the whole plantation”). The analysis then continued to other parts of the life cycle of the dates, i.e., sorting, packing, and transportation, to obtain the complete estimations (from cradle to port in export destinations).

The second part of the analysis was based on an attempt to make the findings “uniform” by neutralizing the impact of the palm trees’ age on the result. This was done to facilitate comparison between palms in the same age group at all plantations (regardless of the unique distribution of every age group palm at each plantation) and to point out the typical environmental performance of palms in every age group.

An analysis was done for each palm tree (1 palm FU) according to the information from the growers. Then, following the approach of previous fruit-related studies (e.g., Cerutti et al. 2014; Goossens et al. 2017), the palms were divided into four groups: (1) FY—mature palms, in full production; (2) PY—young palms, partly productive; (3) NY—young, non-yielding female palms; and (4) M—male palms (Fig. 3). It should be noted that the M group contains palm trees of different ages as well but are considered here a homogenous group due to a lack of data. The analysis of the cultivation stage was conducted for each group separately, followed by a more detailed analysis of the yielding palm trees (Annaert et al. 2017; Ingrao et al. 2015; Milà i Canals et al. 2004). After that, the specific combination of each plantation was considered to check the contribution of each group to the plantation’s impact.

3 Results

The analysis presented in the following paragraphs includes both the average and range of values for the material flows and selected byproducts per unit of fruit for different palm trees and overall plantations based on palm composition.

3.1 The footprints of a unit of fruit: average and range

The research found that, on average, the supply of one metric ton of marketable dates in the region studied requires an average of 1520 kg of MF and has WFblue of 2450 m3 and LF 1530 m2 throughout its life cycle. That ton has an average SWF of 990 kg, and its CF is about 5350 kg CO2eq.

As presented in Fig. 4, for all the plantations, the cultivation stage had the highest share of materials and energy use and related emissions, followed by the sorting and packaging stage, and finally the transportation stage.

Fig. 4
figure 4

The average contribution of the stages to the footprints categories

Full size image

Next, each stage was considered separately to understand its contributions and the sources of its impact. Figures 5 and 6 present the breakdown of the main inputs and emissions related to each stage.

Fig. 5
figure 5

a–c The main sources of the material footprint in each cultivation stage.

Full size image
Fig. 6
figure 6

a–c The main sources of the carbon footprint in each cultivation stage.

Full size image

The cultivation stage

This stage uses an average of 1200 kg of materials (Fig. 5a) and 2360m3 of blue water for each ton of dates. It contributes roughly 4750 kg CO2eq of GHG to the CF (Fig. 6a) and leaves behind 830 kg of solid waste. The main use of the material is attributed to producing electricity for water extraction, which requires on average 760 kg of material and emits 2100 kg CO2eq of GHG for each ton of dates. Fertilizers are also significant, requiring approximately 340 kg of material and emitting 2200 kg CO2eq of GHG.

The sorting and packaging stage

On average, 1 ton of dates requires approximately 220 kg (range: 180–376 kg) of packaging materials, including wood (14 kg), paper and cardboard (55 kg), and plastics and nylons (85 kg). An additional 65 kg of material is needed for producing the electricity used in this stage (Fig. 5b). Processes in the sorting and packaging stage emit, on average, about 300 kg of CO2eq (Fig. 6b). The dates are packed in various package sizes, usually 5-kg, 2-kg, 1-kg, and 2-lb boxes. Each package size requires a different quantity of materials, based not only on its size but also on market demands (e.g., an extra sheet of nylon or plastic tongs that are added to certain packages). The 5-kg package (Fig. 1d, which uses the smallest amount of material, approximately 77 kg per ton) requires approximately 60% less material per ton than the 2-kg package (which uses the highest amount, about 125 kg/ton).

Furthermore, while plastic packaging materials comprise roughly 35% of material inputs, they contribute almost half of the GHG emissions. Conversely, paper packaging accounts for 55% of the material while producing less than half of the emissions at this stage.

As presented in Fig. 4, both the cultivation and packing stages generate significant amounts of solid waste. The production of 1 ton of dates in the plantations studied has an estimated SWF of 565–1920 kg. Approximately 80% of the waste is related to the cultivation stage and is almost entirely organic. The packing stage generates almost 160 kg to the SWF for each ton of dates, mostly non-organic (94%).

The transportation stage

On average, transportation-related inputs and emissions are estimated to be 125 kg of materials (Fig. 5c) and 345 kg CO2eq (Fig. 6c). The research identified three main components of ship**: transportation of inputs (e.g., fertilizers, packaging materials) to the plantations, transportation of fruit from the plantations to the port by truck, and marine ship** of fruit to ports in Europe. Most of the energy inputs (diesel, fuel) and related emissions are connected to refrigerated trucking of the fruit overland to the Port of Ashdod (about 180 km), followed by marine transportation of roughly 3700 km to European ports.

While all the plantations studied are located in the same region, the research found significant differences (Fig. 7). For example, their water inputs range from 1380 to 3480m3 per ton of dates. The LF varies from 800 to 3100 m2. While the MF throughout the life cycle of the dates varies between the plantations, the relative share of the main components in each plantation varies less. Similar patterns can be identified in the CF, ranging from 2300 to 9900 kg CO2eq per ton. The production of 1 ton of dates generates 560–1950 kg of solid waste.

Fig. 7
figure 7

The plantations’ footprint distributions.

Full size image

3.2 Analyzing the range of inputs and emissions at the plantation level

The differences identified between the plantations, as shown in Fig. 7, are mostly found in the cultivation stage (Fig. 4). Overall, the gaps approached 20% for material inputs, 65% for water, 60% for solid waste, and up to 70% for GHG emissions. Two possible reasons that might explain these differences are (1) the unique composition, i.e., the proportions of different palm trees (FY, PY, NY, M) in each plantation and (2) functioning (the human factor, practices implemented by farmers) of each production system. Both reasons are analyzed further in the following paragraphs.

Figure 8 (right hand side) presents the palms’ composition of each plantation studied. Some of the plantations include palms of all four ages, and others include only some. The yields of each plantation are different, and generally, mature plantations will have higher yields than those with more male and young female palms. Therefore, the palm composition of mature plantations is favorable in terms of material flows and emissions per unit of fruit, i.e., the higher the yields, the lower the share of inputs and emissions per unit of fruit (Fig. 8, left hand side).

Fig. 8
figure 8

The plantation footprints and age-group distribution.

Full size image

Table 1 presents the average inputs and emissions per both FU for every palm tree category in each of the plantations studied. It was found that the older the female palms are, the higher the rates of inputs and emissions are on all plantations. The mature palms in the FY group were found to have an average of 15% higher environmental impact per tree (apart from the solid waste). However, due to their one-third higher yield, producing 1 ton of fruit from FY palms had about 40% less impact than PY palms.

Table 1 Impact categories by palm type and plantation (results presented in units per palm and in brackets per ton of fruit).
Full size table

Nevertheless, as illustrated in Fig. 7 and 8, the link between the palm composition of each plantation and the overall environmental impact seems to be quite limited. Therefore, this cannot be the only explanation for a plantation’s impact rate. For example, while plantation no. 4, which has 98% FY palms, shows a low impact per unit of fruit in all examined categories, its impact is almost the same as plantations no. 2 and no. 6 (which have 43% and 40% FY palms, respectively) in some categories and even higher rates in others. Furthermore, while plantation no. 5 has the second-highest share of FY palms (about 83%), it has the highest GHG emissions per ton. Also, plantations 2 and 3, which have similar distributions of palms, present very different environmental impacts; plantation 2 has the lowest values in all the categories, but plantation 3 has the highest.

As Table 1 shows, there are clear differences between each palm type, the inputs used, and waste on each plantation in all categories. It follows that farmers’ practices may be significant for explaining the range of results. The highest variation was found among the NY group, other than the solid waste that was practically identical (very few palm bunches are removed from young palms, and there is no fruit waste), followed by the PY group. The highest variation was found in the GHG emissions category in all three groups, and the second highest was water intake and solid waste. While there is little difference in the water intake for FY palms, there is much less consistency in the amount of water used for the PY and NY palms. The great variation in the materials and carbon footprints can be traced back almost entirely to fertilizers and electricity for water production for all groups. For FY and PY palms, organic solid waste is an important factor, too. Most of the differences are human-made and can be attributed to the varying management methods used by the farmers.

4 Discussion

Medjool dates are becoming a popular fruit worldwide due to their taste, cultural importance, and nutritional values. The fruit thrives in arid climates; therefore, many farms in arid regions are growing Medjool for local consumption or shipment to consumers in overseas markets. By embracing a systematic approach to analyzing the footprint of production and supply of dates throughout their life cycle, we can identify both positive traits and problematic hotspots that could be used to advance a more sustainable system.

The research presented in this paper is one of the first to systematically assess the direct and indirect environmental interactions of any kind of fruit cultivation in an arid area, especially for Medjool dates. Some of its findings correlate with studies on other fruits, highlighting global factors that affect the food system’s sustainability. For example, it reconfirmed the high contribution of the cultivation stage to most impact categories (e.g., Bessou et al. 2013; Giudice et al. 2013; Hesampour et al. 2018; Ingwersen 2012; Page et al. 2012). In this stage, the fertilizers and the production of irrigation water (the fertigation systems) make the main contribution to the footprint indicators in agreement with other fruit-related studies (e.g., Bamber et al. 2020; Ingrao et al. 2015; Mohamad et al. 2014; Pergola et al. 2013).

As illustrated in Fig. 7 and 8, in most plantations, high WFblue reflects in high CF and vice versa, which suggests a water-energy tradeoff (Page et al. 2012).

A photovoltaic (PV) solar energy system, engaging two of the greatest advantages of the desert—the abundance of open space and the high solar radiation almost throughout the whole year—might be the obvious step to reduce the negative impact of the fossil-energy use in all the plantations. It is already in use in the research area, and according to the literature (e.g., Weselek et al. 2019; **e et al. 2018), positive results were found while implementing it. Furthermore, since open space is not endless, even in the desert, in this arid climate, mounting a PV system over agricultural land (Weselek et al. 2019) not only can provide cleaner energy and save space but may also benefit the crop due to the extra shade, and benefit the PV system thanks to some cooling effect over the PV boards (Weselek et al. 2019).

The finding of this study may not be comparable to other date studies since their number is very limited. Yet, the yield and the water footprint are two of the rare components studied in peer studies, so some comparison is possible; although various units and calculation methods were used, and not all the information is available.

The current study showed, as did previous studies (e.g., Al Hinai and Jayasuriya 2021; Al-Muaini et al. 2019; Kotagama et al. 2014; Robinson et al. 2012) that the yield differs between different palm-age groups (Fig. 3), cultivation methods, and variety. Therefore, accurate system analysis requires integrating that complexity; otherwise, a comparison is limited, and the results may vary greatly. For example, Aldababseh et al. (2018) estimated the date yield in the Emirate of Abu Dhabi to be about 40 kg/1000 m2 and Al-Mulla and Al-Gheilani (2017) found 607 kg/1000m2 of date palms in Oman (of different varieties), lower than the average marketable yield found in this study (780 kg/1000 m2, from the whole plantations), and almost half of the yield reported from the average Arava FY palms (1300 kg/1000 m2). That may be due to the date varieties: a factor neutralized in this study that focused only on Medjool dates, a first-rate variety, almost the sole one in the Central-Arava. Indeed, some previous studies suggested that low-yield and low-quality varieties should be replaced by better ones (e.g., Al-Mulla and Al-Gheilani 2017).

Commercial demand is another factor that can explain the differences in the yield. In the studied region, farmers tend to focus on the quality of the fruit—rather than only its quantity—since almost all the date fruit are exported (Housler et al. 2017). Therefore, farmers tend to deliberately remove many fruits in an early stage to gain bigger, “first-class” fruits, sometimes resulting in lessened total yield. This is different from other countries, like Oman, in which about 54% of the date fruit are consumed locally, 24% is used for industry, 19% is fed to animals, and only 3% is exported (Al-Mulla and Al-Gheilani 2017, as reported at Al-Mulla and Al-Gheilani 2017). It can be assumed that in cases like that, the growers wish to maximize the yield quantity.

Yet even under the same conditions, the same variety, and the same age group of palms, as presented in Table 1, palm trees showed a wide range of footprints between the studied plantations, especially in the cultivation stage, with the greatest differences being for the NY and the PY groups. When focusing only on yielding palms of the same age group, it became apparent that some plantations’ MF and WFblue footprints are higher by 34 to 44% to produce 1 ton of marketable fruit, and some methods result in 65 to 70% higher CF and SWF than others. Also, farmers who use the most inputs do not always have the highest yields. There is less agreement between growers concerning the optimal methods at those stages due to a lack of knowledge and/or trust in the experts’ recommendations and/or willingness to change their practices. Implementing the cultivation method of the best performing plantations, mainly regarding the water and the fertilizers factors (with necessary adaptations suggested by experts), can reduce the environmental pressure and even increase the yield.

A recent study (Al Hinai and Jayasuriya 2021) presented the water use and productivity of date cultivation within Middle-eastern and Northern African countries, ranging from 0.15 to 2.8 kg/m3 and about 160m3/palm (range of 43–350 m3/palm). In Oman, the WF of date cultivation was estimated to be 2852 m3/ton (Al Hinai and Jayasuriya 2021), Al-Mulla and Al-Gheilani (2017) estimated that the production of 1 ton of dates needs 1755m3 water, and in Nevada, USA, it was found to be 79–118 m3/palm (Robinson et al. 2012). In the United Arab Emirates (hyper-arid climate), Al-Muaini et al. (2019) found a surprisingly low annual water intake and suggested a possible explanation: the adaptability of date palms to the desert environment (Al-Muaini et al. 2019). The current study results (Table 1) are well between the abovementioned ranges, higher than the other hyper-desert study and, at the same time, lower than some other findings from areas with milder temperatures. That may imply that the previous studies considered varied ages of palms. It may also suggest that it is possible to reduce the volume of irrigation water currently given to the palms in the Arava. This was previously suggested by local scholars, like Sperling et al. (2014), that showed a new approach to calculate the recommended irrigation intervals for dates in the studied region, leading to about 20% savings potential. The local date-growers are familiar with the findings of this study (and others with similar conclusions), yet very few, if any, implement them. One concludes that the human factor, the farmers’ decision-making process, plays a part here, rather than a lack of information.

Significant factors were found in the other production stages, such as packaging material (like Wikström et al. 2014)) and especially the plastic components (e.g., Giudice et al. 2013). Customer demand for a lower-impact packaging containing more fruit and/or recycled materials can reduce the fruit CF, MF, and SWF footprints. However, sometimes re-used packages are involved in additional procedures that bare an additional environmental price tag (e.g., Ingwersen 2012) and packaging with a lower environmental impact may result in higher food waste (Wikström et al. 2014).

In the transportation stage, the use of diesel has a negative environmental impact (e.g., Hesampour et al. 2018; Page et al. 2012), especially the refrigerated road transportation (Ingwersen 2012), solely used in the area because of the lack of other means, such as rails. The hydraulic platforms’ high diesel intake is one of the reasons for removing the old palms, along with health, safety, and financial considerations. Generally, the higher the working height of machines is, the more expensive and intimidating for the workers, and more specialized training is required. However, regarding the high inclusive impact of younger palms compared to older palms (although it was not found as the main source for variation) and the higher yield of the latter (Table 1), a further examination of the issue should be done (see: Kotagama et al. 2014). Such examination should consider more factors, like the extremely expensive machinery, the avoided deforestation and product if the logs are re-used (Ingrao et al. 2015), the palms’ carbon storage, and avoided solid waste. Solid waste can also be spared by mulching and spreading the fronds over the soil (Aune et al. 2017), a method used by some local growers. Others claim that pests thrive in the biomass, and the dry climate prevents the decomposing of the unevenly spread fronds, which presents a physical hazard to the safe passage of heavy machinery.

These notions are examples of the need to evaluate every solution according to the specific system’s features and other factors (e.g., Ingwersen 2012; Page et al. 2012), such as the workers’ safety, economic perspective, regulations, and the tradeoff between an improvement on one category and enhancing the pressure in another. It also seems that the government policy is involved in the date-production system, apart from the physical conditions and the farmers (e.g., Mupambwa et al. 2019; **e et al. 2018). There are options for producing the same amount of fruit with lower environmental impact at the hotspots above in all three production stages. The above examples illustrate why alternative scenarios’ overall environmental price tag should be assessed before implementing suggested improvements.

The results indicate that local physical factors, such as climate, soil, variety, and the development stage of the palm and more, affect the environmental footprints of a food system, as has been reported in the literature (e.g., Ingwersen 2012; Milà i Canals et al. 2004; Page et al. 2012).

Another key factor consistent with previous work (e.g., Annaert et al. 2017; Al Hinai and Jayasuriya 2021; Ingwersen 2012; Milà i Canals et al. 2004) is the human factor. Farm-specific practices, determined by the farmers’ knowledge, belief, trust in experts, experience, financial consideration, and more, significantly influence the total environmental footprint: the amount and type of materials, water, and energy used/produced throughout the product’s life cycle. As illustrated here, significant differences were found between the seven plantations studied, attributable to the types and age of palm trees of each plantation and farming practices. Based on the results, most of the differences are of human origin.

5 Conclusions

The production of date fruit throughout its life cycle generates significant environmental footprints. As illustrated in the analysis presented in this paper, it demands great amounts of inputs (especially water and energy) due to the desert’s unique climate conditions. However, significant differences were found between the plantations that were traced back to the palms’ age and human factors: the agriculture methods and decision-making.

Therefore, improving the system’s sustainability can primarily be accomplished by influencing those three factors. Some improvements are not feasible currently due to a lack of infrastructure (i.e., rails) or regulations (organic waste management), hence the importance of government intervention and policy. That policy should develop and provide professional knowledge to the growers, for example, regarding the palm-removal procedures, whether fronds mulching is recommended, and advised PY and NY palm cultivation methods. Yet, as for the last point, the problem is not necessarily the absence of information but rather the reluctance of the farmers to comply. Therefore, further improvement depends on the human factor and can be achieved by convincing farmers to adopt better alternative practices, including using less water, different fertilizers, and cooperative transport. The human factor also includes the retailer and customers, which can be encouraged toward more sustainable consumption patterns through education, and other government interventions.

The environmental footprint analysis presented in this paper attributed to the dates highlights challenges that should be addressed in advancing sustainable practices in food systems. Making the results of this and similar studies accessible to farmers, followed by the application of best practices, could promote the sustainability of agriculture in the region. The findings of this research can also be used in other arid regions where dates are cultivated. It can support the understanding of policy-makers and the scientific community of the desert-food system sustainability challenges and strengths.

Further research should be conducted to analyze alternatives for the hotspots identified and estimate their effect over the presented baseline. Such a study could consider current methods used by some farmers, which have been proven to have the least impact and explore other alternatives not presently used in the area.

As environmental change continues, especially climate change and desertification, more regions around the world may face the challenges that the Central Arava faces today, meaning that the present research insights might contribute to the prosperity and the livelihood of additional farming communities and support the development of sustainable food systems in arid regions.