Long-term Experiment on Fennel-cluster Bean Cropping System as Affected by Production Systems: A Strategy for Improving Yield, Quality and Soil Fertility

The rise in unprecedented pressure of producing enough food for an ever-growing human population resulted in the use of considerable amounts of agrochemicals in crop production. (FAO, 2017). Although this approach can augment farm output and reduce yield gaps, yet has potential adverse impacts on the soil, water and air. Using annual pulse crops such as cluster beans (Cyamopsis tetragonoloba L.) in cropping systems is an alternative technique for enhancing output while minimising environmental impact. Cluster bean, is primarily grown in dry and semi-arid parts of India, Pakistan and the United States. Legume crops provide several distinctive and key ecosystem services, such as resource utilisation (Kumar et al., 2023), improved soil health and increased nitrogen use efficiency (Jangir et al., 2017b), reduce carbon footprint and positive effects on soil properties (Liu et al., 2020), water conservation (Ding et al., 2018), reduced N₂O emission, enhanced biodiversity (Guiguitant et al., 2020) and production sustainability (Sahruzaini et al., 2020). Based on studies conducted worldwide, crops cultivated after pulse crops yielded more than crops grown after non-pulse crops (Mann et al., 2020). Legume crops add large amount of organic residues through leaf fall and rhizodeposition and the intermediate acids produced during organic residue decomposition also solubilise fixed forms of nitrogen and phosphorus in soil resulting in increased available nitrogen and phosphorus (Lal et al., 2022). As a result, adding pulse crops to cropping systems may boost system stability and on-farm profitability (Nath et al., 2023). In India, fennel (Foeniculum vulgare Mill.) is commonly grown as an annual and spice crop which is mainly cultivated in the states of Gujarat and Rajasthan (almost 80% production of India). The main components of fennel seeds essential oil are (E)-anethole (increases prolactin hormone), methyl chavicol (anti-spasmodic properties), fenchone (treats gastrointestinal diseases) and limonene (anthelmintic effect), which make it feasible in the pharmaceutical, food and cosmetic industries (Abdellaoui et al., 2020). The demand for seed plants having medicinal and nutraceutical properties has been increased across the globe and consumers become now more aware about natural products and their health benefits. However, given the potential negative effects of excessive use of herbicides and chemical fertilisers on the profile of active ingredients in medicinal and aromatic plants, the majority of food, pharmaceutical and cosmeceutical companies prefer materials derived from sustainable and organic systems (Rezaei-Chiyaneh et al., 2020). Based on the numerous research outcomes, it has been determined that integrated use of organic and inorganic agro-inputs and fertilisers has proven to be a right source of nutrients (Kumar and Dhar 2010; Gezahegn, 2021). With the addition of pulse to the fennel crop cycle, soil fertility has enhanced, particularly the quantity of accessible nitrogen and soil organic carbon. The increased fennel output after leguminous crops may be due to the residual effects of legumes on biological N fixing and the addition of root biomass by legumes to successive fennel crops (Rathore et al., 2011; Jangir et al., 2017a).
       
In view of above, the key objective of the present study was to diversify fennel cropping systems with pulse crops and to assess the impact of different production systems on the quality, yield, yield attributes of fennel-cluster bean viz-a-viz soil quality over the study period.

The field experiment was conducted under AI-Network Programme on Organic Farming (AI-NPOF) Project during 2016-17 to 2020-21 at the ICAR-National Research Centre on Seed Spices, Ajmer (Rajasthan), which is located in the Northwest of India (26.3661°N, 74.5933°E). The experiment was conducted in sandy soil having loamy sand texture. A two-factor RBD experiment (three replicates) with fennel (Rabi) - cluster bean (Kharif) crop sequence grown under (1^st factor) six production systems viz., (PS₁) 100% organic, (PS₂) 75% organic + 25% innovative practice (Compost extract, cattle urine), (PS₃) Integrated (50% organic + 50% in organic), (PS₄) Integrated (75% organic + 25% inorganic), (PS₅) 100% inorganic nutrient sources and (PS₆) State recommendation], the environment of five years in 2016-2020 was considered as (2^nd factor). The size of the plot was (10×6 m²) and spacing (Row to row, plant to plant) was kept 50×30 cm, 30×10 cm for fennel-cluster bean respectively. Observations on seed yield (kg ha^-1) and crop attributes such as days to initiation of seed germination, days to germination, plant height (cm) at different stages (45 and 90 DAS), number of primary and secondary branches plant^-1, number of umbel plant^-1, number of umbellate umbel^-1, days to flowering initiation, days to 50% flowering, biomass yield (kg ha^-1), harvest index and quality characteristics such as essential oil (%), seed nitrogen (%), seed phosphorus (%) and seed protein content (%) were recorded in Rabi fennel. Similarly, seed yields (kg ha^-1) and yield attributes such as plant height (cm), number of primary branches plants^-1, number of pods, number of seeds pods^-1 and biomass yield (kg ha^-1), seed index, seed nitrogen (%) and total protein content (%) were recorded in Kharif cluster bean. The environmental data for the year 2015-16 to 2020-21 have been recorded at meteorological unit of ICAR-NRCSS, Ajmer and their patterns over the period were analysed. The Kjeldahl method was used to calculate available nitrogen, which was then reported as percentage of protein content. The availability of P and K was assessed using the molybdate blue method and the flame photometry method, respectively, whereas the amount of soil organic matter was estimated using the dichromate-sulfuric acid (K₂CrO₇-H₂SO₄) oxidation method. The concentrations of soil-available N, P and K were measured using the methods explained by Olsen et al., (1954) and Page et al., (1982), respectively. Analysis of variance (ANOVA) was used to assess the study results compiled from 2016 to 2020 and Tukey’s test at a significance level of <0.05 was used to determine the significance of differences. Calculations were performed using R statistical program, version 4.2.2. The findings were subjected to a 2-way analysis of variance in a randomised block design and data synthesis was performed using the mixed model. The effect of the environment on seed yield and quality characteristics of fennel-cluster bean was examined year after year. Furthermore, the coefficient of determination (R²) was computed to evaluate dependencies and relationships between the selected parameters (Yield, yield components and soil parameters).

Environmental data
 
Analysis pertaining to various attributes of environmental data revealed that throughout the experiment period, decreasing pattern of total average rainfall (R=-0.057; p=0.85) was recorded from the years 2015-16 to 2019-20 (Fig 1). During the growing season of 2018, the fennel growing (Rabi) season received the least average rainfall (0 mm) while, the maximum average rainfall (741 mm) occurred during the cluster bean growing (Kharif) season of 2019. The increasing pattern of evaporation (R=0.16; p=0.59) was recorded (non-significant) in the years 2015-16 to 2019-20. In all experimental years, a maximum temperature was recorded during the cluster bean growing season (Kharif) of 2019 and the Rabi season of fennel at 39.2°C and 27.5°C, respectively. While, the minimum temperature recorded in 2018 for Rabi fennel and 2015 for the Kharif season were 3.5 and 28.3°C, respectively. The decreasing pattern of maximum and minimum temperature (R=-0.089; p=0.76 and R=-0.19; p=0.52, respectively) was recorded in the years 2015-16 to 2019-20.
 

 
Yield and yield attributes
 
Over the period, long-term integrated fertilization has significantly increased fennel and cluster bean yield compared to the solely organic fertilization however, in the year 2017 and 2018 reduced yield was recorded due to variations in climate variables such as temperature and rainfall. Fennel yield increased over the years, especially under the PS₄ and PS₆ treatments while, cluster bean yield showed irregular patterns due to unfavourable climatic conditions, particularly in the Kharif season. On an average across the years from 2015-16 to 2019-20, compared to the PS₅, fennel seed yield increased by 2.53%, 4.32%, 11.97%, 19.62% and 19.72%, respectively and cluster bean yield increased by 3.48%, 4.75%, -1.00%, 19.09% and 14.43%, respectively, under the PS₁, PS₂, PS₃, PS₄ and PS₆ production system, respectively (Fig 2a). In the year or temporal factor, the fennel seed yield in 2016–2020 ranged between 1148 (2017) and 2207 (2019) kg ha^-1. In the production system factor, the highest fennel grain yield (1971 kg ha^-1) was observed in PS6 completely at par with PS₄ (1970 kg ha^-1). However, the lowest seed yield (1646 kg ha^-1) was observed in PS5 completely at par with PS₁ (1688 kg ha^-1). Similarly, the highest cluster bean yield (1501 kg ha^-1) was observed in PS₄ completely at par with PS₆ (1443 kg ha^-1) (Fig 2c). The study showed that PS₄ was the most suitable production system for seed yield in both crops. The highest annual biological yields of fennel (6817 kg ha^-1) and cluster beans (6799 kg ha^-1) were observed in the years 2019 and 2017 respectively (Fig 2b and 2d).
 

       
The yield-year interaction analysis data was previously tested for normality using the Shapiro-Wilk test (W=0.983; p-value=0.3012, R²=0.995 and W=0.989; p-value=0.8383, R²=0.998) for fennel and cluster bean, respectively. The two-way analysis of variance (ANOVA) detected that the differences between the evaluated seed yield of fennel and cluster bean were highly significant (p<0.001), which proved that the production system used in this study had great differences in seed yield (Fig 2). High significance (p<0.001) was also observed between the seed yield and year (Y).
       
The days to germination in fennel depended significantly on the production system used for the raising of the crop and the interaction between the year and the production system (Fig 3a). Significantly, the highest days to germination initiation were characteristic of the seed with significantly highest in the PS₄ production system (8 days) and the pattern was almost similar to days to complete germination (12 days) (Fig 3b) in the year 2016. While, plant height at different stages (45 DAS and 90 DAS) had observed lowest in the year 2016 and highest observed in the year 2020 in the production system PS₄ (43.24 cm and 118.56 cm, respectively) (Fig 3c, 3d).
 

       
In fennel, the number of primary and secondary branches (12 and 26, respectively) was highest in production system PS4 which was at par with the production system PS₆ in the year 2018 and 2016, respectively (Fig 4a, 4b). The number of umbels per plant and umbellate per umbel was significantly higher observed in the production system PS₄ which was at par with the production system PS₆. The number of umbels per plant was highest observed in the year 2020 and was on par with the year 2018. The umbellate per umbel was highest observed in the year 2018 at par with the year 2016 (Fig 4c, 4d). The days to flower initiation and 50% flowering were longest in the year 2020 which was significantly superior to other years. It was observed that the maximum days to flowering were recorded in production system PS₄ which was completely at par with the PS₆ (Fig 4e, 4f).
 

 
In cluster bean, the plant height and number of primary branches were observed significantly highest in production system PS₄ which was at par with the PS₆. Meanwhile, the plant height and number of primary branches were observed in decreasing patterns over the year (Fig 5a and 5b). It was also observed that the lowest plant height and number of primary branches were studied in the production system PS₁. The number of pods had significantly superior in the year 2020 over to the remaining years. The production system PS₄ had significantly superior in the number of pods and seeds per pod of cluster bean. The seed per pod was significantly superior in the year 2018 over to the remaining year (Fig 5c and 5d). The seed index of cluster bean was highest observed in the production system PS₄ and PS₆ (Fig 5e).
 

       
Higher yield might be a result of increased availability of nutrients from organic manure and favourable impacts from production system PS₄ (75% organic + 25% inorganic), which supplies nitrogen and phosphorus to support plant yield and growth. In contrast to organic fertilizers, which release nutrients gradually through microbial mineralization to assure nutrient availability in the seed developmental stages of crops and even in the succeeding crops, inorganic fertilisers supply nutrients immediately after application. When compared to the combined application of organic and inorganic fertilizers, it was revealed that the single use of inorganic fertilisers was ineffective at increasing crop output. These findings are consistent with the findings of Mevada et al., (2017); Singh et al., (2018) who observed significant effects of mixed organic manure and chemical fertiliser application on yield production and growth parameter of fennel. The addition of legumes in crop sequence during the long term experiment favoured improved nutrient availability and uptake, resulting in greater source accumulation and efficient translocation of photosynthates into a sink and therefore, a higher seed production. Similarly, according to research by Kumar et al., (2022), the application of both organic manures and inorganic inputs had a substantial impact on the fennel crop’s growth and yield characteristics metrics.
 
Seed quality attributes
 
In fennel, the highest seed nitrogen (2.92 and 5.08%, respectively) was observed in the production system PS₄ which was significantly superior to other production systems but at par with the PS6 in the cluster bean (Table 1). The study shows that fennel following cluster bean in the PS₄ system had 17.29 and 30.75% increment of protein as compared to PS₅ and PS₁ in fennel and cluster bean, respectively. The protein content in fennel, as well as cluster bean, was highest observed in the year 2020 due to the inclusion of pulses in cropping systems which helps in improved plant uptake. On an average across the years, compared to the PS₂, fennel essential oil increased by 5.51%, 2.29%, 0.75%, 2.29% and 2.29%, under the PS₁, PS₃, PS₄, PS₅ and PS₆ production systems, respectively. While, the cluster bean seed index compared to the PS4 increased by 4.05%, 2.66%, 3.70%, 4.76% and 0.98%, respectively, under the PS₁, PS₂, PS₃, PS₅ and PS₆ production systems. Therefore, diversified cropping systems that include pulses in the rotation can more consistently produce high seed and protein yields than a continuous fennel monoculture system and fallow systems in semiarid prairies, regardless of growing conditions.
 

 
Soil properties
 
Soils under (PS₁) 100% organic and (PS₂) 75% organic + 25% innovative practice (compost extract, cattle urine) plots contained higher soil nitrogen at the depth of 0-20 cm, over the (PS₅) 100% inorganic nutrient sources in the year 2020 (Fig 6a). The mean estimate of N by the system was ~9% higher in the plots under (PS₁)-100% organic than that under (PS₅) - 100% inorganic nutrient sources. Soils under (PS₁) 100% organic plots contained higher soil phosphorus at the depth of 0-20 cm, over the remaining production system all year (Fig 6b). The mean estimate of P by the system was ~38.8% higher in the plots under (PS₁) - 100% organic than that under (PS₅)-100% inorganic nutrient sources. Over the years 2016 to 2020 it was observed that soil phosphorus had a significant increment with the application of different production systems. Soils under (PS₄) Integrated (75% organic + 25% inorganic) plots contained higher soil potassium in the 0-20 cm soil layer, which was at par with (PS₂) - 75% organic + 25% innovative practice (Compost extract, cattle urine) in all years (Fig 6c). The potassium concentration in the soil had not much variation in the production system. Over the years 2016 to 2020 it was observed that soil potassium had a non-significant increment with the application of different production systems.
 

 
The soil organic carbon (SOC) ranged from 0.27% to 0.34% for different production systems at the beginning of the experiment in 2016-2020 (Fig 6d). All the field plots were low in soil fertility as the organic carbon was less than 0.27%. All production systems had significant differences in SOC in year 2020. The SOC increased with the continuous application of organic and inorganic fertilizers for 5 years among all production systems. The soil fertility in 100% organic (PS₁) was the highest significant increment in SOC ranging between 0.27% and 0.34% at par with all other production systems except the PS₅ production system in the year 2020. The SOC in production system PS1 was completely at par with PS₂ (75% organic + 25% innovative practice (Compost extract, cattle urine) all year. The lowest SOC was observed in the production system PS₅ (100% inorganic nutrient sources) in all years of the experiment.
 
The quantitative relations between yield and soil parameters
 
The data included 30 yield data (fennel and cluster bean each) and 30 soil data of three soil variables (soil organic carbon, soil nitrogen and soil phosphorus each). A significant positive relationship was observed between fennel grain yield and soil organic carbon, soil nitrogen and soil phosphorus (Fig 7a-7c) and also the Shannon-Wiener indexes (R² = 0.75, P<0.01, R² = 0.83, P<0.01 and R² = 0.95, P<0.01). Similarly, in cluster bean, a significant positive relationship between cluster bean grain yield and soil organic carbon and soil phosphorus (Fig 7d and 7f) and also the Shannon-Wiener indexes (R² = 0.69, P<0.01 and R² = 0.71, P<0.01). However, the relationships between cluster bean grain yield and soil nitrogen were found not significant for the evenness index (R² =0.04, NS) (Fig 7e).
 

       
Under production system PS₁, the per cent increment of readily available N (~9%), P (~38.5%) and K (~non-significant) were noted over the years respectively, compared with PS₅. Long-term use of organic manures could effectively stop applied nitrogen from leaching and raise the N and SOC content of the ploughed layer (Meng et al., 2005). After 5 years, the rise in SOC was greatest (~25.2% increment) in production system PS₁ (100% organic) and lowest for production system PS₅ (100% inorganic nutrient sources). In a similar study, rice-wheat cropping systems based on pulse systems outperformed in terms of available nitrogen (8-29%), phosphorus (3-35%), potassium (6-15%) and sulphur (3-13%) at the surface soil (Nath et al., 2023). The pulse-based cropping system and organic manures both work wonders for improving the soil’s physical, chemical and biological qualities as well as its fertility gradient. These findings are consistent with the findings of (Brar et al., 2015), who studied continuous pulse farming systems, as well as combined effects of organic and inorganic fertilizers, boosted soil carbon sequestration and crop yields. Higher carbon input through organic supplements (farmyard manure and above-ground biomass) and biological nitrogen fixation resulted in increased SOC and accessible N under organic manure and pulse-based systems, respectively (Rob et al., 2022). The study resulted that the inclusion of pulse crops in the cropping pattern, the addition of organic amendments and a reduction in chemical fertiliser could gradually restore soil fertility and soil health according to higher SOC in the fennel-cluster bean (PS₁) than the fennel-cluster bean (PS₅).

From the study it could be inferred that seed yields and their attributes of the fennel and cluster bean were observed to augment over time with the production system PS₄ (75% organic + 25% inorganic). Even after many years of experimentation, the field trial’s soil carbon concentrations are not in a steady-state condition due to the changing environmental conditions. The production system (PS₁-100% organic) receiving manure had both the greatest overall SOC contents and the steepest growth throughout the experimental period. SOC, available N and P contents grew by 25.2%, 9% and 38.5% over the examined 5-year period.

Authors are thankful to ICAR- Indian Institute of Farming Systems Research, Modipuram, U.P. India for providing financial accreditation throughout the research programme under All India Network Programme on Organic Farming (AI-NPOF).