To assess the growth attributes, total height (leading culm) in all the plots was measured by using the Forestry pro (Nikon) and recorded in meters (m). Ten culms per clump were selected for biometric attributes measurements like internodal length and culm diameter. Culm diameter at the 5th internode was measured using a digital calliper, in two perpendicular directions and average recorded in millimetres (mm). Internodal length was determined by measuring the length of the 5th internode on each culm using a tailor tape and recording it in centimetres (cm). The total number of internodes of the leading culms was counted till the utmost visibility from selected clumps and average was recorded. Crown width was measured by projecting the crown onto the ground in two perpendicular directions (north-south and east- west) and recording the mean value in meters (m). Clump circumference was measured at breast height (1.37 m) using a tailor tape and recorded in meters (m). Finally, the total number of culms clump-1 was counted for selected clumps and recorded.

LAI was measured using a LAI 2000 plant canopy analyzer (LI-COR Inc., Lincoln, Nebraska, USA) at a height of 1.5 meters above ground. To obtain accurate LAI measurements, the LAI 2000 unit was first calibrated by taking a reading of sky brightness outside the plot in open conditions. Then, ten random locations within each plot were sampled to measure light transmission beneath the canopy. To ensure consistent measurements, the unit was always oriented in the same direction both inside and outside the stand. To avoid direct sunlight reaching the sensor, measurements were taken during periods of low solar elevation, such as just after sunrise or just before sunset. A 90-degree view restrictor was used to prevent direct sunlight and the observer from obstructing the sensor.

Installation of bamboo litter traps:

Litter collection was conducted using specially designed circular litter traps, following the methodology outlined by Hughes et al. (1987). Each trap consisted of four 210 cm long galvanized (2-3 mm) iron wires, arranged in a tripod configuration. A circular hoop, measuring 55 cm in diameter, was created using the remaining wire and secured horizontally to the tripod. A plastic grain bag with a tapered bottom was placed inside the hoop, forming a litter trap with a collection area of 0.25 m² (50 cm × 50 cm) and a depth of 30 cm. Three such traps were installed in each plot, totalling 24 traps across the eight replicates. The traps were positioned 1 meter away from the bamboo clumps and 50 cm above the ground level.

Sampling and weighing:

Litter was collected monthly from each trap for a period of 12 months, spanning from November 2021 to October 2022. The collected litter samples were transported to the laboratory and oven-dried at 70°C until they reached a constant weight (approximately 72 hours). Once cooled, the dry litter samples were weighed to the nearest milligram.

Bamboo biometric growth attributes were analysed by single factor -spacing- Randomized Block Design (RBD) whereas litter fall production was analysed by two factor -spacing + months- Randomized Block Design (RBD). Data were analysed for treatment comparisons at the p ≤ 0.05 level of significance using statistical software R 3.6.1 version and Microsoft Excel 2010 version.

The present finding showed that the biometric growth attributes particularly: culm height, number of culms per clump, crown width, and clump circumference (except culm diameter, number of internodes and length) of four-years-old D. stocksii plantation were significantly influenced by planting spacing/density (Table 1).

Table 1. Biometric performance of 4-year-old Dendrocalamus stocksii stand under differential spacing regime in Kerala, India.

(± values are standard error of the mean; values followed by the same superscript in a column do not differ significantly (p< 0.05); NS: Non-significant)

The culm height was reduced as the bamboo spacing expands. Culm height was highest at 8 × 4m (8.12 m) and decreased by 10% as the spacing was widened to 8 × 8m (7.27 m). The average annual increment of stand was 0.99, 1.1 and 1.04 m at 8 × 8, 8 × 6 and 8 × 4 m respectively. Height growth patterns in response to varying planting densities are well documented (Kittur et al., 2016; Kar et al., 2019; Rane et al., 2018). Similarly, in this study, spacing manifested rapid height growth at closer spacing, but lower densities encouraged greater crown spread and radial growth. In dense, even-aged plantations, competition for light can actually stimulate height growth. As a strong light demanding species, D. stocksii naturally showed taller at closer clump densities. Interestingly, the observed culm height in this study was relatively higher than that reported by Rane et al. (2018) for the same species, where height increased as stand spacing decreased from 6x6 m to 3 × 3 m. This trend was also reported for D. strictus by Kittur et al. (2016) in same geographic location and Dev et al. (2020) in central India

The culm diameter at the 5th and 6th internodes was statistically insignificant across all spacing treatments. Intriguingly, despite varying spacing, culm diameter remained relatively consistent, likely due to a combination of the plantation's young age and the site's high soil fertility. Radial growth between 5th and 6th internodal diameter was found moderate for D. stocksii in the present study which varied from

36.82 mm (8 × 4 m) to 38.08 mm (8 × 8 m) which is comparable to the mean values reported by Rane et al., (2018) from the Western Ghats. Patil and Mutanal (2017) on their study at Karnataka, reported higher culm diameter for D. stocksii at wider spacing (4 × 5m), compared to closer spacing (4 × 2m). Similar trend was also reported by Prasath et al. (2014) for B. vulgaris in the transition from 5 × 5m to 7 × 4m from Tamil Nadu.

In the present study observed that when compared to wider spacing (8 × 8 m), closer spacing (8 × 4 m) clearly demonstrated a reduced growth rate in diameter and crown width. The rapid expansion of the crown driven by the wide spacing might have triggered the diameter growth at the detriment of the height growth (Jiang et al., 2007). Bahru and Ding (2020), in their study from southwestern China, found that a higher density stand (597 culms ha-1) of D. brandisii exhibited a denser canopy and larger leaf area, leading to a substantial reduction in culm diameter compared to a lower density stand (532 culms ha-1).

Internodal length is a key growth characteristic that significantly impacts the commercial value of bamboo. In the present study, internodal length and numbers were observed non-significant across all spacing treatments. Interestingly, at spacing (8 × 4 m), we found that taller culms had longer internodes but fewer nodes, suggesting a direct relation between height and internodal length and an inverse relation between height and node number. The opposite trend was observed at wider spacing (8 × 8 m). Internodal length in bamboo can vary significantly depending on soil and climatic conditions, even within the same species. For instance, D. stocksii in our study exhibited substantially longer internodes at all spacing regimes compared to previous reports from the Himalayas (26.6 cm) (Kaushal et al., 2021). Previous studies from Kerala, reported significant correlations between height and number of nodes (r = 0.480) and internodal length and number of internodes (r = 0.435) (Jijeesh, 2013). However, in our study it was not much notable, may be due relatively young age, site specific factor or difference in growth rate

The culm number per clump is a key determinant of bamboo growth and productivity. To optimize bamboo production, regulating spacing essential for clumps with an ideal number of culms and desirable culm dimensions. In this study, regardless of the non- significant variations in culm diameter, the number of culms per clump was significantly influenced by spacing treatments, showing a decreasing trend as spacing decreased. The highest count of culms/culms was at widest spacing (58.00) and this count drastically declined by 41.46 % at closest spacing. On a hectare basis: 9048, 10,156 and 12,792 number of

culms per hectare was recorded under 8 × 8, 8 × 6 and 8 × 4 m bamboo spacings respectively. The increased number of culms in wider spacing could be attributed to reduced crown competition and more available above-ground space for culm growth. Kittur et al. (2016) found that as spacing increased from 4 × 4 m to 12 × 12 m, the number of culms also increased from 46.66 to 130. Compared to Kittur et al. (2016), the number of culms per clump was lower in this study, possibly because of the younger age and different species. Notably, Rane et al. (2018) and Patil and Mutanal (2017) reported significantly lower culms per hectare than ours for D. stocksii in the Konkan region and Dharwad, Karnataka respectively, suggesting that agro-climatic conditions play a key role in the growth and productivity of bamboo.

The crown width was statistically significant (p < 0.05) among the spacing treatments. In the present study, it was observed that with increase in spacings, crown expanded 28.89 % more at 8 × 8m spacing compared to 8 × 4 m (6.58 m). The average crown width varies depending on the age of the plantation and the level of competition the tree undergoes. It is typical to expect a broader crown in widely spaced trees. This trend was also reported from Kerala, were crown width with increases as spacing increased from 4 × 4 to 12 × 12 m for D. strictus (Kittur et al., 2016). Kar et al. (2019) in their study reported a decrease in height and increase in diameter of tree along with increasing crown spread as moving from closer (8 × 1 m) to wider (8 × 4 m) spacing for Grewia optiva stand. The crown width of bamboo plays a critical role in manifesting PAR (Photosynthetically Active Radiation) and LAI (Leaf Area Index), which in turn affects understorey productivity in agroforestry systems (Kittur et al., 2016). A wider crown increases shade, reducing PAR transmission to the understorey, while a narrower crown allows more light penetration. Achieving an optimal shade level is essential for maximizing understorey crop yield without suppressing growth. In intercropping systems, maintaining the right balance of shade ensures that understorey crops receive sufficient light for photosynthesis while benefiting f rom the microclimate created by the bamboo canopy. Understory productivity may benefit from the increased PAR availability in wider stands. To mitigate the effects of overshading, canopy management techniques such as harvesting or pruning can help improve light penetration and stimulate understory growth and productivity.

Clump circumference showed a consistent increase as increase in the spacings. The maximum clump circumference was observed for widest spacing (4.06

m) and there was reduction of 16.33% in closest spacing (3.49 m). It can be linked as number of culms per clump increases, the average clump cover also increased. The average annual rate of increment was highest in 8 × 4m spacing (1.09 m) and lowest was in 8 × 8m spacing, declined to 0.89 m. A similar pattern of spatial expansion was observed when spacing increased from 4 × 4 to 12 × 12 m in, as reported by Kittur et al. (2016).

The spatial extent of bamboo clumps is a key consideration when integrating them into agroforestry systems. For agroforestry, bamboos with compact growth and lower clump circumference are always desired. Compared to the widely cultivated B. bambos in Kerala, which has reported clump circumference of

10.76 meters (Kumar et al., 2005) and values reported by Amani et al. (2021), D. stocksii in this study exhibited a more compact and less spreading growth habit. This implies that it would be more suitable for integrating in multitier agroforestry systems, like Kerala’s home gardens.

The impact of planting density (spacing) on litterfall in a four-year-old D. stocksii stand is evident in Figure 2. The litterfall rates across spacing and months were very significant (p<.005), while interaction effect of spacing × month was insignificant. With increasingspacing, litterfall production declined significantly (p < 0.05). Mean annual litter production was ranged from 7.86 Mg ha–1in the widest spacing (8 × 8m: 156 clumps ha–1) to 10.11 Mg ha–1in the closest spacing (8 × 4 m: 320 clumps ha–1). These values were relatively higher than reported by Lubina et al., (2021) in moist semi-arid areas, Karnataka and Kaushal et al. (2020, 2021) in the Himalayan foothills for the same species. The increased litter fall may be attributed to favorable climatic conditions and higher soil fertility, which promoted bamboo growth and denser canopies, leading to greater litter production year-round. The highest mean monthly litterfall (841.49 kg ha–1) was observed in the closest spacing, while the lowest (654.74 kg ha–1) was seen in the widest spacing. Leaf litter contributed dominantly to total litterfall (> 90%) across the spacing and seasonal variations (based on visual observation). Compared to wider spacing stand, closer spacing stand may produce more litter due to the presence of a larger number of non-functional crowns associated with them (Kunhamu et al., 2009).Seasonal variations in litterfall followed a bimodal pattern where the major peak observed during dry months (January to February) and a subsidiary minor peak observed from May to June. Seasonal variations in leaf litterfall are well-reported for many tropical species, as evidenced by studies in both natural and planted ecosystems (Jamaludheen and Kumar, 1999; Jijeesh and Seethalkshmi, 2015). Meteorological variables like low relative humidity, high wind speed, mean maximum monthly temperature coupled with lesser rainy days (Fig. 1) in the dry season (January- March) might have led to higher litterfall than in other seasons (Jijeesh and Seethalkshmi, 2015). Additionally, the planting density did not significantly alter the seasonal trend of litterfall, with Nov’2021 to Mar’ 2022 accounted for > 55% of the total.

Quantity of litterfall might also be regulated by factors such as LAI, crown spread, and crown structure. Studies reported that compared to low density stand,

D. brandisii at high density stand with high LAI and compact canopy structure exerted in ample foliar biomass which enhanced higher litter production (Bahru and Ding, 2020). Our finding implies that spacing, biometric and climatic factors had the greatest impact on the quantity and periodic fluctuations of litterfall year-round.

LAI, a species-specific measure of plant, is a critical factor in understanding plant systems. As a fundamental canopy structural parameter, LAI significantly impacts various eco-physiological processes, such as evapotranspiration, rainfall interception, photosynthesis, respiration, and soil organic matter contribution (Yan et al., 2019). From the fig. 3, it is evident that LAI and spacings were inversely related. LAI was highest at 8 × 4m (4.35 m2 m-2) and declined by 46.90 % as spacing widened to 8x8m. Decreasing trend of LAI with respect to spacing was in order of: 8 × 4 > 8 × 6 > 8 × 8m. Despite lower clump girth and crown width, the higher LAI in closely spaced stands could be attributed to the more number of clumps per unit area (Table 1). The similar trend was also observed by Kittur et al., (2016) for Dendrocalamus strictus within the similar ecoclimatic conditions and geographic location of the present study. High density stand (8 × 4 m) had overlapping crowns, led to reduced canopy gaps and consequently higher LAI values and vice versa. For instance, Bura and Ding (2020) reported a higher LAI value at high density stand of D. brandisii, favoured by largest canopy leaf area (5.63 ± 0.34 m2) and least canopy spread area (9.92 ±1.73 m2), while at lower density both factors altered and LAI eventually reduced. For successful integration of D. stocksii into agroforestry, regulating spacing will be a critical factor in maintaining the quantitative relationship between planting density, LAI, and understorey PAR transmittance, with the ultimate goal of optimizing understorey production (Kittur et al., 2016; Kumar et al., 2001).