1 Introduction

Buckwheat (Fagopyrum sp.), being a pseudocereal, is renowned for its numerous medicinal and nutritional benefits. Buckwheat has many nutraceutical properties, like cardioprotective, anti-cancer, anti-tumour, hepatoprotective, anti-hypertension, anti-inflammatory, anti-diabetic, neuroprotection and cholesterol-lowering [1]. Unlike many common cereals, buckwheat protein boasts high quality and is notably rich in the essential amino acid lysine [2]. Furthermore, buckwheat proteins are gluten-free and exhibit a balanced amino acid and micronutrient composition, with elevated levels of health-promoting bioactive flavonoids, positioning it as a promising crop for the future [3]. However, enhancing buckwheat productivity remains a challenge, with seed characteristics such as yield, size, shattering, lodging resistance, and early maturity being focal points for breeders [4]. The bitterness of Tartary buckwheat (F. tataricum) seeds, attributed to high rutin levels, is a major factor limiting its broader consumption [5], compounded by issues of low production and incompatibility. Most of the research on buckwheat concentrated on the above-ground traits of the crop, thus neglecting the below-ground parts [6]. However, for efficient crop performance, it is important to deeply examine, understand, and select potential traits in RSA that would help in solving problems related to buckwheat production and productivity.

Roots, being vital underground components of plants, play crucial roles in supporting crop yield and growth by serving as the primary interface between plants and the soil environment. Through their intricate network, roots facilitate the uptake of water and essential nutrients vital for metabolic processes and crucial to plant growth and development. Roots provide structural support, anchoring plants securely in the soil and enabling optimal exposure of leaves to sunlight for photosynthesis [7]. Their symbiotic associations with beneficial microorganisms enhance nutrient acquisition, while their ability to explore soil resources contributes significantly to crop productivity [8]. Understanding and optimising root functions are essential for sustainable agriculture, ensuring robust crop performance and food security in diverse environmental conditions.

RSA refers to the spatial configuration of the root system and the arrangement of roots within the soil, encompassing various morphological and developmental patterns. The RSA demonstrates plasticity and responds to various environmental factors, including soil moisture, nutrients, temperature, pH, and microbial communities [9]. RSA is critical for maximising crop yield due to its influence on nutrient and water acquisition, as well as its impact on plant adaptation to environmental stresses. Moreover, the spatial distribution and depth of roots within the soil profile significantly affect water uptake and drought tolerance [10]. The architecture of the root system influences soil structure and fertility by promoting soil aggregation, organic matter decomposition, and nutrient cycling. Healthy soil conditions fostered by well-developed root systems create an optimal environment for root growth and function, ultimately enhancing crop yield and quality [11, 12]. Thus, understanding the complexity of RSA is essential for optimising crop growth, yield and enhancing agricultural sustainability.

Studying the entire root system in field presents challenges due to its underground nature and the fact that the available equipment is still in its primary stage. The Soil-based traditional methods such as shovelomics, soil coring, rhizolysimeters, and mini rhizotrons are commonly used, but they are time-consuming, labour-intensive, low-throughput, and not suitable for large-scale genotypic studies required for genome-wide association mapping, thus necessitating the development of minimally intrusive, non-destructive methods for root system evaluation. However, in recent years, the development of column-culture, soil-less media, hydroponics, and gel-based media has eased root analysis in various crops [13]. Moreover, current advancements in image-based systems such as X-ray computed tomography, laser, nuclear magnetic resonance, ground penetrating radar, infra-red, and near infrared imaging, along with automated analysis software, offers promising opportunities for studying roots in their natural environments.

Root architectural studies in various crops have provided useful insights into crop plasticity to abiotic stresses and resource acquisition, and there is substantial experimental evidence that roots will drive the resilience of farming systems [14,15,16]. There are no reports of root architectural variations in buckwheat and their relationship with the adaptive potential of this valuable crop. The Western Himalayan region is a rich repository of buckwheat variability, and in view of its projected importance as a future crop, we conducted a comprehensive study of RSA to understand species variability for RSA and its relationship with above-ground biomass using a diverse set of 117 lines.

2 Material and methods

2.1 Plant materials and seed treatment

A diverse set of 117 genotypes of buckwheat species (F. esculentum and F. tataricum) were collected from different regions of Jammu & Kashmir (Kishtwar, Machil, Kishtward, Kupwara, Minji, Gurez, Kargil, Leh, & Ladakh) and accessions procured from the National Bureau of Plant Genetic Resources (NBPGR), New Delhi, India (Supplementary Table 1). The seeds were surface sterilised using 1% NaOCl (sodium hypochlorite) and 0.1% HgCl2 (Mercuric chloride) solution. The final washing was done several times with the autoclaved distilled water to remove the left-over chemicals. The seeds were dried in an aseptic condition and stored until sowing.

2.2 Experimental design setup

The experiment for root phenotyping was performed in the green house facility at the Wadura campus of SKUAST-K. Polyvinyl chloride (PVC) pipes with 1.4 m height and 20 cm internal diameter were used in the column culture technique following completely randomised design (CRD). The columns were filled with an equal proportion of soil and sand to ensure the field conditions. Initially, four seeds were sown in each column, but finally only two plants per column were retained. The columns were regularly irrigated to avoid any water stress. The shoot length and biomass was recorded just before harvesting. The harvesting of roots was done after complete flowering (45 days after sowing) by cutting the stem at the first inter-node and taking out the whole soil with roots from the column ensuring no root fraction was left behind. The harvested roots were first washed with tap water, followed by washing with mild detergent, and then again rinsed with tap water to remove all soil particles. The washed roots were shade dried. Gross morphological traits were recorded on shade-dried roots, followed by storage in zip-lock bags until further analysis through root scanning (Fig. 1).

Figure1
figure 1

The Steps followed during analysis of RSA in Buckwheat

Full size image

2.3 Root phenotyping and scanning

The phenotypic data of roots were recorded for all 117 genotypes. Roots were scanned with the Epson Perfection V700 (Epson Perfection V700 with 6400 dpi resolution, Epson, Long Beach, CA) photo scanner and the images were analyzed with RhizoVision Explorer software [17]. The root parameters that were taken into consideration includes: root length (mm), root biomass, root tips, branch points, branching frequency (mm), average diameter (mm), perimeter (mm), volume (mm3), surface area (mm2), fine root length (mm), medium root length (mm), large root length (mm), fine root surface area (mm2), medium root surface area (mm2), large root surface area (mm2), fine root volume (mm3), medium root volume (mm3) and large root volume (mm3).The root images were acquired on a Grayscale to a resolution of 700 dots per inch (dpi). The analysis was done on the root morphology by setting the rough edge and noise removal to a higher level.

2.4 Statistical analysis

Genetic diversity was assessed through Principal Component Analysis (PCA) and Euclidean clustering of yield and quality traits using R Studio. The selection of principal components for further analysis was based on their Eigen values [18]. Eigen values greater than one were considered indicative of reliable principal component weights [19]. The graphical representation of variation patterns was achieved using a biplot. XLSTAT version 2021.2.2 (Lumivero Inc. Denver USA) were used for the correlation analysis.

3 Results and discussion

3.1 Variability for root traits

Substantial variability was recorded for gross root morphological traits as well as RSA features extracted through root scanning. The maximum root length was found in the case of F. tataricum (BWM-2; 108 cm) and the shortest length in the case of F. esculentum (BWZ-63; 8.67cm). The wide range of variation was recorded for other root characteristics including; root tips (130–4938), branch points (188–7679), branching frequency (0.51–2.67 mm), average diameter (0.87–0.21 mm), perimeter (343.6–11967 mm), volume (29.14–10,017.03mm3), surface area(237 mm2-20,462.33mm2), fine root length (112.33–2876.42 mm), fine root surface area (65.238mm2-1707.55mm2), medium root surface area (15.65 mm2-7248.09 mm2), large root surface area (1.23 mm2-4017.9mm2), fine root volume (3.8 mm3-104.33 mm3), medium root volume (9.44 mm3-925.07 mm3), large root volume (2.22 mm3-6208.03 mm3), root depth (8.67–108 mm), root biomass (0.23gm -21.23gm), shoot length (7–193 cm), shoot biomass (0.31gm-204.67gm), Root/shoot ratio (0.77–3.8)showed a lot of variation within two species of Buckwheat.

Comparative analysis of interspecific variability of root traits revealed that F. tataricum have better root system than F. esculentum in terms of root tips, root length (mm), root volume (mm3), root perimeter (mm), branching frequency (mm), surface area (mm2), root volume (mm3), root biomass (gm), root depth (mm), and root surface area (mm2). However, shoot biomass (gm), shoot length (cm), and average diameter of root (mm) are more in F. esculentum (Figs. 2, 3).

Fig. 2
figure 2

Different Root system from two species of buckwheat

Full size image
Fig. 3
figure 3

The average values of different traits (root tip, root length, perimeter, branch point, branching frequency, root volume, root surface area, root biomass, shoot length and shoot biomass) in case of F. tataricum and F. esculentum

Full size image

Better root length indicates total root system size, potential for absorption of nutrients and water from soil, an indicator of baseline soil microbial activity, especially Arbuscular mycorrhizal fungi (AMF), activity in soil, and microbial functioning [20]. Root volume helps in probable limitations to soil nutrients and water exploitation. Root diameter has a significant role in mycorrhizal development, regulation of water stress, potential for apoplast-symplast exchange, growth potential, and influencing and responding to soil physical condition. Root depth shows the physical stability or anchorage, the depth of soil exploited, the potential for resource use, and the estimation of available nutrient and water resources [21, 22]. Branching frequency reveals the relative intensity of soil exploitation, mean root longevity, soil exploitation strategy, hormone production (e.g., cytokinin) potential, meristematic activity, and presence in volume of soil. The ratio of root and shoot shows the relative allocation strategy, used against a benchmark of plant nutrient status [23].

3.2 Trait association for root and shoot traits

Simple and comprehensive data analysis software XLSTAT has given the precise and conclusive result for correlation of different traits of the buckwheat, which shows that the root tips had a negative correlation with plant height and average diameter of root and positive correlation with branch point and root length. Branch points and root length are negatively correlated with plant height and shoot biomass and while branch points are positively correlated with the root tip and root length and vice versa. Branching frequency is negatively correlated with the average diameter and positively correlated with the average diameter and long root surface area. The average diameter is negatively correlated with root depth and branching frequency. The root parameter is negatively correlated with plant height and average diameter. Root volume is negatively correlated with plant height. Surface area is negatively correlated with plant height and root depth and positively correlated with the surface area. Route depth is negative correlated with average diameter. Root biomass is negatively correlated with average diameter. Plant height is mostly negatively correlated with maximum parameters like: fine root length, fine root surface area, and fine root volume, etc. Shoot biomass is negatively correlated with average diameter and positively correlated with the plant height and shoot biomass (Fig. 4).

Fig. 4
figure 4

Heatmap showing correlation matrix of different root traits in Buckwheat

Full size image

3.3 PCA and cluster analysis

Principle component analysis (PCA) for root traits in buckwheat germplasm was performed to simplify the complexity in high-dimensional data while retaining trends and patterns. The complete variance is described by sixth PC’s component (Table 1). In PCA, eigen value less than unity are not considered, which occurred after the 6th PC value. Notable PC’s values are considered only where the initial two PCs are taken for the construction of a biplot (Fig. 5), which accounted for 58.466% of the variance (PC1: 42.448% and PC2: 16.018%). The first PC accounted for 42.488% of variance (mainly contributed by surface area), medium root length, and medium root volume, while PC2 accounted for 16.018% of variance mainly contributing branching frequency, root depth, and average diameter (Supplementary Table 2).

Table 1 Eigen value representing variance between different principal components
Full size table
Fig. 5
figure 5

Biplot analysis of different root traits. Aberration are given as RD- root depth, BF- branching frequency, RB- root biomass, PE- perimeter, RL- root length, RT- root tips, BP- branch point, SA- surface area, FRL- fine root length, FRSA- fine root surface area, FRV- fine root volume, VOL- volume, MRL- medium root length, MRSA- medium root surface area, MRV-medium root volume, LRL- large root length, LRSA- large root surface area, LRV- large root volume, AD- average diameter and PH- is plant height

Full size image

A biplot was constructed from the first two PCs for the root traits and revealed a strong correlation between root length, branching frequency, and root tips, while medium root length, medium root volume, and medium root volume also showed strong correlation. The length of the line shows the magnitude of the traits, while the angle shows the correlation. Acute angles forming between the lines showed a positive correlation between the two traits, obtuse angles revealed a negative correlation, and perpendiculars demonstrated no correlation (Fig. 5). Branching frequency and the ratio of root and shoot are negatively correlated, and the contribution of branching frequency in PC2 is maximum (14.050%) and the ratio of root and shoot is minimum (0.201%). Root tips are strongly correlated with branch points and root length. Branch points are highly correlated with root tips and root length. Root lengths are highly correlated with root tips and branch point. The parameter is highly correlated with root length and root tips.

The genotypes were also grouped on the basis of their RSA and formed four clusters (Supplementary Table 3). Cluster 1 (C1) has 75 genotypes, C2 has 6, C3 has 12 and C4 has 24 genotypes. It is interesting to note that the clusters contain genotypes from both species showing intra-species similarities but only C2 have F. tataricum species and considered as the cluster of best of 4 clusters as F. tataricum species have better root system (Fig. 6).

Fig. 6
figure 6

Cluster diagram where we found 4 clusters representing 117 genotypes of buckwheat

Full size image

The analysed buckwheat RSAs show their ex-situ state topologies. This phenotyping tool identifies the phonologies and fingerprints of soil-root interactions. Parameterization of crop RSA and illustration with time-series presentations provides a useful means of phenotyping the geometric, topological, and distribution behaviours of buckwheat RSAs. High root distribution in the upper layer of soil is favourable for optimum use of surface-applied nutrients [24]. Roots found in depth are not genetically governed, while they are more favoured by environmental factors. The placement of soil and roots in three-dimensional spaces provides new insights into RSA in soil and the interaction of roots with the environment. High root density provides potential for crop yield [25]. The phenotyping protocol for Buckwheat RSA offers an effective way to demonstrate how roots, soil, and the surrounding environment interact, benefiting precision agriculture and enhancing crop management.

4 Conclusion

Buckwheat is a highly nutritious crop grown mainly in the higher Western Himalayan, regions including the Kashmir Valley. In the present study, we undertook the first comprehensive study on root architectural variability in two buckwheat species grown in the Himalayan region. The study revealed wide variation in root architectural traits and provided insights about the relationship of RSA with above ground biomass. The multivariate analysis identified key traits contributing to towards variability, and we identified potential genotypes with vigorous root systems that will enable the breeders to develop resilient varieties of buckwheat that can be cultivated, especially in low-input marginal areas prone to low water stress.