Cultivation Practices Affect Biomass Yield and Quality of “Felina 32”, an Industrial Hemp Variety

Abstract

Hemp is a multipurpose crop that produces high amounts of lignocellulosic biomass. There are few studies dealing with hemp’s biomass production (lignocellulosic and fiber) under cropping strategies such as irrigation and nitrogen fertilizer. Therefore, the aim of the current study was to assess the effect of irrigation and N-fertilization on the lignocellulosic biomass of one of the most well-known industrial hemp (Cannabis sativa L.) varieties worldwide (Felina 32) under a typical Mediterranean climate. For the purposes of this study, a field experiment was conducted in central Greece (Thessaly region) over the cultivating years 2022 and 2023. We determined the produced biomass dry yield and the dry biomass yield vs. N-uptake relation, as well as quality characteristics for hemp (cv. Felina 32) under three irrigation (I1: 33%, I2: 66%, and I3: 100% ETo) and three N-fertilization levels (N1: 0, N2: 70, and N3: 140 kg ha⁻¹). A significant difference in the dry biomass yield was found, ranging between 10–11.2 t ha⁻¹ using I2 and/or I3 irrigation levels and remaining at 8.6 t ha⁻¹ with lower irrigation (I1). In terms of fertilization factor, it appeared that in samplings where statistically significant changes were observed, all degrees of fertilization differed from one another, with the N3 treatment exhibiting the highest production (11.4 t ha⁻¹). The average protein content varied between 10% in 2023 and 14% in 2023. A linear biomass yield–nutrient uptake relationship was found with high R², pointing to a nitrogen use efficiency of 55.15 kg kg⁻¹. Thus, it would seem that nitrogen fertilization and irrigation are both crucial factors of industrial hemp cultivation, helping to raise the crop’s overall yield of lignocellulosic biomass. The introduction of hemp into land-use systems necessitates thorough evaluation, as hemp shows considerable potential as a crop that can yield substantial quantities of above-ground biomass (lignocellulosic: stems and fibers). This is especially true in regions where irrigation is possible, and the application of nitrogen fertilizers can further enhance these yields.

1. Introduction

Hemp (Cannabis sativa L.), the only genus in the Cannabaceae family, has been a subject of debate over its number of species. Some suggest numerous polytypic species, while others suggest a monotypic, highly polymorphic species, Cannabis sativa L. [1,2]. The taxonomic classification of cannabis strains, particularly modern hybrids, is ongoing. Recent reports [3,4] accept a two- or three-species interpretation, while Small [2] suggests a one-species notion with discrete variants. Nonetheless, it is recommended that the basis for classifying cannabis be the cannabinoid and terpene profile [5,6].

It is reported that in the case of hemp, there are both monoecious and dioecious varieties, where there are several differences between the plants in their morphological characteristics. However, most hemp plants are dioecious and grow to a height of 1–6 m [7,8,9,10].

Hemp’s adaptability leads to varying growth and development requirements based on its intended use. In the literature, it is reported that for fiber production (fibers found in the hemp bark and used in many applications: ropes, fabrics, etc.), the typical nitrogen requirement ranges from 50 to 100 kg ha⁻¹, while for seed production, it increases to between 100 and 150 kg ha⁻¹. Notably, applying up to 200 kg per hectare of nitrogen has been found to be beneficial in both scenarios [11,12]. Although there are studies suggesting specific nitrogen fertilization rates, research on the phosphorus and potassium needs of hemp is limited. Jordan et al. [13] recommended a phosphorus application of 30 kg ha⁻¹ for fiber production, with higher amounts suggested for seed production. As for potassium, the crop does not have specific requirements, and it is advisable to replace the potassium removed from the biomass during the harvesting process. Since a significant portion of the absorbed potassium is found in the stem, typical harvests for various uses would primarily include the stem, leaving any potassium present in the roots and leaves in the field [9].

Depending on each variety’s unique vegetative cycle, the estimated water requirements for the plant range from 250 to 400 mm (thus the reason Tang et al. [8] report an application in total of 60 mm to 155 mm of water in different growing years according to ET). Furthermore, Cosentino et al. [14] reported that lower requirements apply to early flowering cultivars, and Tang et al. [15] reported that lower irrigation leads to less LAI, which leads to lower yields.

This crop’s many advantageous agronomic characteristics, including its capacity to compete with weeds [16], its ability to successfully enter a crop rotation system [17], its relatively low input requirements for nitrogen and plant protection products (insecticides, fungicides, herbicides, etc.), and its capacity to produce biomass yields of over 20 t ha⁻¹ of dry shoots [18], are the reasons for the renewed interest.

Cannabis above-ground biomass yields vary significantly depending on the variety [19,20] (there are varieties that are higher than 2 m and some that are lower, one of the main reasons for biomass yield variation), cultivation practices, and climate of each region. Reports indicate that fresh biomass weight can vary from 18.73 to 45.5 tons per hectare [20,21]. For shoot dry weight, harvest yields can range from 6.2 to 11.7 tons per hectare [22,23], highlighting the advantages of more southerly climates for crop productivity. Additionally, the fiber content in the shoot can vary between 20.5% and 43.0% [18,24].

Hemp, based on its uses, can be classified into three distinct end-use categories (i) grain (food), (ii) fiber (fiber source and construction material), and (iii) cannabinoid (medicinal plant and for CBD oil output). Stem xylem contains shorter fibers with a higher concentration of lignin that is used to produce particleboard and insulating construction materials [25,26]. The total biomass of the plant can be used in processes for the production of bioenergy such as biodiesel, biogas, bioethanol, and solid biofuels [27]. Hemp pellets have a calorific value of 5.2 MWh/ton, which is greater than the 4.3 MWh/ton of wood [28]. Furthermore, it is important to note that each country maintains a list of hemp varieties that are approved for industrial purposes, with the primary regulatory criterion being the Δ9-THC (Tetrahydrocannabinol) content of the variety.

Finally, cannabis also adapts easily to different environments, making it possible to grow in a wide range of latitudes and climates. Limited research has been conducted on the production of lignocellulosic biomass and fiber under various cropping strategies, such as irrigation and nitrogen fertilizer. Therefore, the current study aims to explore the production of lignocellulosic biomass in the Felina 32 variety (one of the most known varieties worldwide) of industrial hemp (Cannabis sativa L.) in the Thessaly region under various agricultural techniques and a typical Mediterranean climate.

2. Materials and Methods

2.1. Experimental Site and Design

For the purposes of this study, a two-year field experiment was carried out to examine the impact of varying levels of irrigation and nitrogen fertilization on hemps’ growth and biomass yield within a typical Mediterranean soil and climatic setting. This study was carried out in central Greece, specifically in Eastern Thessaly, Velestino, Magnesia.

To achieve the research objectives, the Felina 32 variety was selected from the national list with a Δ9-THC (Tetra Hydro Cannabinol) content of less than or equal to 0.2% (a variety predominantly utilized by farmers in Greece, as well as more broadly throughout the Mediterranean region). A factorial split-plot design was utilized, with four replicates (blocks) and nine plots per replication. The main factor was the different irrigation levels: 33% ETo (I1), 66% ETo (I2), and 100% ETo (I3), using a drip irrigation system (irrigation schedule was determined according to the Class A evaporation pan method, and irrigation was applied weekly based on the A-Pan measurements [29]). Sub-factor was the different nitrogen fertilization levels, which included N1: 0, N2: 70, and N3: 140 kg ha⁻¹, using 46-0-0 fixed nitrogen fertilizer with urease inhibitor and nitrification inhibitor. The used (established) plant density was 120 plant/m² (20.85 kg ha⁻¹).

2.2. Soil Characteristics

The soil is classified as Calcixerollic xerochrept [30], and it is characterized as semi-fertile to unfertile (organic matter of 1.3% at a depth of 0–20 cm, 1.66% at 20–40 cm, 1.12% at 40–60 cm, and 0.54% at 60–80 cm). It is a clay loam soil with an alkaline nature (pH 8.1, 8.1, 8.3, and 8.3 at 0–20, 20–40, 40–60 and 60–80 cm depth, respectively). Additionally, the total nitrogen content (g/100 g) measured at various depths was recorded as 0.12, 0.11, 0.071, and 0.052 for the depths of 0–20, 20–40, 40–60, and 60–80 cm, respectively. Similarly, the electrical conductivity (μS/cm) values were 459, 473, 399, and 388 for the same depth intervals.

2.3. Experimental Management

Felina 32 was sown on 6 May 2022 and 9 May 2023, with a plant distance of 15 cm between lines and a plant density of 120 plants/m². Before hemp’s establishment, the field had remained uncultivated for a period of two years (2020–2021). Surface fertilization using fixed nitrogen fertilizer with urease inhibitor and nitrification inhibitor (46-0-0) took place on 3 July for both experimental years (2022 and 2023).

Plants were sampled four times during the growing season in July, August, and September in order to determine the reached biomass yield during crop growth. To minimize the border effect, plants from an inner plot with a surface area of 0.5 m² were cut 10 cm above ground level. During the initial sampling, the plants were observed at growth stage 55 (bulk appearance of the male flower buds), characterized by the prominent appearance of male flower buds. In the second sampling, they had progressed to growth stage 65 (full flowering), which is defined by full flowering. By the third sampling, the plants reached growth stage 79 (nearly all fruits have reached final size and coloration), where nearly all fruits had attained their final size and coloration. Finally, at the last sampling, the plants were at growth stage 99 (harvest product), indicating that they were ready for harvest, as outlined in the literature [31]. The plants were then weighed immediately in the field (the moisture content of the biomass during the first three samplings ranged from approximately 88% to 92%, whereas the final sampling recorded a moisture content of 30–35%.), and a sub-sample was taken for further laboratory measurements. The sub-samples were placed in a hot air dryer at a temperature of ≤59 °C until they reached a stable weight (dry weight with 4–5% moisture content) to ensure reliable measurements.

2.4. Yield Quality Characteristics Measures

Following air drying at a temperature of ≤59 °C, the desiccated samples (the entire plant consisting of stems, leaves, flowers, and seeds together) were crushed and chopped in preparation for laboratory examination. Using the DA 7250 NIR analyzer from Perten Instruments in Hägersten, Sweden, a near-infrared reflectance (NIR) spectroscopy approach was used to determine the ash, protein, neutral detergent fiber (NDF), and acid detergent fiber (ADF) in the analysis.

Plant quality traits NDF and ADF are correlated with the crop’s age and growth stage. The feed content of lignin, cellulose, hemicelluloses, and insoluble minerals is estimated by the NDF and ADF values. Protein content reflects the nitrogen biomass content and, thus, hemp’s N-uptake.

2.5. Meteorological Data & Statistical Analysis

Complete weather data (air temperature, radiation, air humidity, wind velocity, precipitation, and class-A pan evaporation rate) were recorded hourly in an automatic meteorological station, which was installed next to the experimental field.

The obtained data were submitted to an analysis of variance (ANOVA) within the sample timings for all measured and calculated variables using the statistical program GenStat (7th Edition). The differences in the means of the main and/or interaction effects were assessed using the LSD_0.05 test criteria [32]. A thorough review of the data was made possible by this statistical analysis, which also guaranteed that any differences between the investigated variables that were found were statistically significant and not the result of chance.

3. Results

3.1. Meteorological Data

The Mediterranean climate of the research region is well known for its warm summers and moderate winters. For both experimental years, the average air temperature during the germination period (May) varied between 15.2 and 19.2 °C. The 2023 growing season experienced higher temperatures than in 2022. In 2022, the maximum average temperature during the crop’s growing season was 27.7 °C, with a minimum of 15.2 °C. In contrast, the 2023 growing season recorded a maximum average air temperature of 29.7 °C and a minimum of approximately 16.3 °C (Figure 1).

Another crucial environmental component that significantly affects crop growth and output is precipitation. In 2022, the area recorded 170.8 mm of precipitation during the growing season; however, the amounts of rainfall varied between the sampling periods, coming in at 93.4 mm, 0 mm, 1 mm, and 76.4 mm, respectively. Conversely, in 2023, the entire amount of precipitation measured in the region from seeding to the last measurement was 476.8 mm. In the interim between sampling periods, the recorded rainfall amounts were 54.2 mm, 0 mm, 6.3 mm, and 416, 3 mm, respectively. The aforementioned data emphasizes the presence of drought conditions in the summer, especially from mid-July to mid-August (Figure 1), whereas the extreme “Daniel” phenomenon that occurred in Thessaly is responsible for the significant amount of water that occurred in September 2023.

3.2. Hemp’s Biomass Yield

In the first experimentation year (2022), the first two samplings showed statistically significant differences in interaction, although overall dry biomass appeared to be significantly affected by both irrigation and nitrogen fertilization factors (Figure 2).

It seems from the very first sampling that there are differences between all irrigation levels, with the I3 treatment having the maximum production. The I1 treatment’s dry biomass yield remained lower (about 2 tons per hectare) than the I3 treatment’s (11.29 t ha⁻¹) dry biomass production.

Regarding the fertilization factor, it seems that all fertilization levels varied from one to another, with the N3 treatment having the maximum production, with the exception of the third sampling. Dry biomass production for the N1 treatment remained at low levels, roughly 2.6 t per hectare less than the dry biomass of the N3 treatment (11.48 t ha⁻¹).

Finally, the data illustrated in Figure 2 show that the average dry biomass production at the final sampling was approximately 10 tons ha⁻¹. The treatments (I2, I3) and (N2, N3) demonstrate similar yields across both years of the experiment.

The dry biomass of hemp appeared to have been statistically impacted by both irrigation and nitrogen fertilization factors during the past two samplings in the case of the upcoming experiment year (2023), with an additional effect identified in the interaction (Figure 2).

Although levels I2 and I3 of the dry biomass appeared to be the same in the case of the irrigation factor over the last two samplings, level I1 differed from both of these levels. At the final harvest, the I1 treatment’s dry biomass output was about 3 t ha⁻¹ lower than the I3 treatment’s dry biomass yield of 11.23 t ha⁻¹.

Regarding the fertilization factor, it seems that all fertilization levels differed from each other in the samplings where statistically significant variations were found, with the N3 treatment having the maximum production (Figure 2).

Thus, it would seem that nitrogen fertilization and irrigation are both crucial factors of industrial hemp cultivation, helping to increase the crops’ overall yield of lignocellulosic biomass.

3.3. Hemp’s Quality Characteristics

Farmers all around the European Union effectively cultivate and use the French industrial hemp variety Felina 32 for fiber production, biomass, and seeds. The initial sampling’s quality characteristic measurements (Figure 3) demonstrate that irrigation did not have a statistically significant impact on the quality characteristics, but nitrogen fertilization affected all the under-study quality characteristics. The average protein, ADF, NDF, and ash contents during the first sampling were 14.5%, 19.9%, 36.6%, and 6.7%, respectively.

According to Figure 3, it appears that the fertilization factor showed statistically significant differences in the percentages of protein and ash content quality characteristics, while the irrigation factor continued to not cause statistically significant differences in any of the quality characteristics under study during the second and third sampling.

Lastly, statistically significant variations from the fertilization effect for protein and ash were also noted during the fourth and final sampling (Figure 3). During this sampling, the mean contents of protein, ADF, NDF, and ash were 15.0%, 24.2%, 43.6%, and 8.7%, respectively. As a result, it seems that, as plants mature, their levels of protein, ADF, NDF, and ash rise.

Figure 3 displays the findings of the 2023 initial sampling’s measurements of the quality characteristics. As can be shown, neither fertilization nor irrigation had any impact on quality features. The average protein, ADF, NDF, and ash levels during the initial sampling were 11.2%, 21.2%, 39.1%, and 5.9%, respectively.

The ADF and NDF concentrations were the only ones affected by fertilization in the second sampling (Figure 3). The average protein, ADF, NDF, and ash levels were 10.3%, 20.7%, 38.0%, and 5.1%, respectively.

During the third and fourth sampling, as can be shown in Figure 3, neither irrigation nor fertilizer altered any quality characteristics. It was discovered that the average levels of protein, ADF, NDF, and ash were, respectively, 9.9%, 19.2%, 36.3%, and 5.7%.

In conclusion, it seems that irrigation and fertilization did not significantly contribute to the differentiation of the quality parameters in 2023, as evidenced by the average values of these attributes slightly declining to stable values. Protein, ADF, NDF, and ash had mean levels ranging from 9.9–11.2%, 19.2–21.2%, 36.3–39.1%, and 5.1%–5.9%, in that order.

3.4. Yield–Nitrogen Uptake Relation

The relationship between yield and nitrogen absorption is shown in Figure 4 by plotting dry biomass against nitrogen uptake (which is determined as ∑Ni × Di, where N is the amount of nitrogen in the plant tissue and D is the amount of dry biomass per plant, for each growing year in the left graph and for both growing years 2022 and 2023 in the right).

It is evident that each year influenced the nitrogen concentration, and given that all other variables (such as soil, variety, and treatments) remained constant, the variations appear to stem from environmental factors, particularly meteorological conditions, that differ from year to year. The period from 2022 until the end of June experienced warmer temperatures, with May being notably hotter than in 2023. This climatic difference impacts the protein concentration in both seeds and overall biomass. The elevated temperatures during the initial two months of the vegetative phase likely contributed to an increase in the protein concentration of the plants.

It is important to note that the yield–uptake relation’s slope represents the nitrogen use efficiency (NUE that has been obtained from the experiments conducted in both 2022 and 2023), which in our situation takes a significant value—roughly 51.5 kg kg⁻¹ (Figure 4).

4. Discussion

The plant density set in this study was 120 plants per square meter, a level that the literature suggests can help suppress weed growth [16]. Furthermore, this density is considered optimal based on previous research [33,34,35], as it can lead to increased yields of both dry and fresh above-ground biomass.

The current study’s dry biomass yield for this particular variety appears to be completely consistent with the literature, which states that the dry biomass yield of Felina 32 varies from 9.83 to 11.00 t ha⁻¹ depending on soil, climate, and cultivation practices [36]. The data presented in the tables indicate a notable increase in the weight of hemp corresponding to higher irrigation doses. This observation aligns with the existing literature, which suggests that restricted irrigation results in reduced biomass across various plant components, including shoots and leaves. This reduction is likely attributed to constraints on cell expansion [37,38]. Furthermore, studies have shown that drought conditions lead to a decrease in the total dry matter in hemp [39,40], and hemp needs supplemental irrigation in order to achieve optimum yields [14,41]. Furthermore, it has been reported that the yield of this variety is also influenced by varying sowing densities and fertilization levels for the above-ground dry biomass at the stage of seed maturation (same stage as the final sampling of the current research) [22,42,43].

Numerous investigations have examined the effects of nitrogen fertilization on hemp cultivation [8,44]. The results suggest that in fertile soils, biomass yield exhibits a limited response to nitrogen fertilization [45], while a notable increase is seen in soils with lower fertility [9]. These findings align with the current study, which demonstrated that biomass yield increased with higher levels of fertilization in semi-fertile soil.

In addition, based on the quality characteristics, it is possible that the increased level of protein observed up to the final sampling is a result of the increased production and maturation of the seeds at this stage, combined with the fact that the hemp seed has a very high content in protein [12]. The cellulose- and lignin-based cell walls of biomass are referred to as ADF, as previously mentioned. ADF values are crucial for generating various calculated metrics in forage reports, as forage digestibility typically declines with increasing ADF levels. It has been reported [46] that the ADF content in most cereal biomass (hay) ranges from 36% to 42%; however, in this study, the harvested biomass was found to be below 20%. The NDF value encompasses the entire cell wall, which includes both ADF and hemicellulose components. Within the rumen, cellulose and hemicellulose undergo partial digestion, while lignin remains indigestible [47,48]. Additionally, this study observed that NDF levels were also lower compared to biomass (hay) from other crop sources. The ash content of hemp biomass was found to be nearly consistent when compared to other lignocellulosic crops, including switchgrass [49].

The literature indicates that the protein content remains consistent across treatments within the same experimental field, rather than varying by the same variety across different fields, implying a climatic influence on plant responses [50]. Increased precipitation and reduced temperatures during the growing season may have adversely impacted protein levels. These findings align with other experimental data. A comparison of hemp seeds harvested in different years showed that seeds from the year with the least rainfall had a higher protein content. Additionally, in soybeans, protein levels in seeds were positively associated with temperature and negatively associated with rainfall during the seed development phase [50,51,52].

In conclusion, Felina 32’s NUE (in this research) was found to be approximately 51.5 kg kg⁻¹, which is almost the same as the reported NUE of industrial hemp (60 kg kg⁻¹; [45]), significantly less than switchgrass cultivation at the same site (e.g., switchgrass 240 kg kg⁻¹) [53], and nearly equal to other lignocellulosic biomass crops like kenaf (72–76 kg kg⁻¹), corn (60–70 kg kg⁻¹), and sunflower (25–30 kg kg⁻¹), all of which have been reported for the Thessaly area [54,55]. Apparently, “luxurious growth” had not yet been attained, whereby further intake increase is instead attributable to a rise in nutrient concentration in plant tissue and a consequent decline in NUE. In closer proximity to the crop’s production potential, higher biomass yields could be expected.

5. Conclusions

It appears that the initial year of experiments showed statistically significant differences in interaction and the overall dry biomass was significantly affected by both irrigation and nitrogen fertilization factors. The I3 treatment had the maximum dry biomass production, which was almost the same in both cultivating years (above 11 t ha⁻¹), compared to other irrigation levels. In the case of fertilization, the N3 treatment had the maximum biomass production (11.4 t ha⁻¹) compared to the rest fertilization levels. However, irrigation and fertilization did not significantly contribute to the differentiation of quality parameters in 2023, characteristics that are likely influenced by genetic factors and may be impacted under soil–climate circumstances.

The relationship between yield and nitrogen absorption showed a linear relationship, indicating a nitrogen use efficiency of roughly 51.5 kg kg⁻¹. The dry biomass yield of the Felina 32 variety is consistent with the literature, and the increased protein level observed may be due to increased seed production and maturation. The nitrogen use efficiency of Felina 32, as noted above, is comparable to that of other annual lignocellulosic biomass crops, including kenaf, corn, and sunflower. However, it is considerably lower than that observed in perennial crops such as switchgrass. Higher biomass yields could be expected with closer proximity to the crop’s production potential.

Considering all the above, it could be said that the cultivation of hemp, and especially the variety Felina 32, can produce a high amount of lignocellulosic biomass in Mediterranean climates, contributing to the strengthening of agricultural income and giving a possible solution to the replacement of traditional crops with new ones. Therefore, the introduction of hemp into land-use systems should be carefully considered, as it appears to be a promising crop for the production of lignocellulosic biomass.