Paleolimnology and Natural Versus Anthropogenic Influx During the Late Holocene from Vembanad Wetland, Ramsar Site, Kerala, India

Abstract

A multi-proxy study of diatoms, palynofacies, and grain size was conducted on a 100 cm core from Arookutty, Vembanad wetland, Kerala, India, to reconstruct paleolimnological changes during the late Holocene, with a focus on natural versus anthropogenic influences. Four distinct depositional phases, from ca. 500 BCE to ca. 400 CE, were identified, aligning with the Roman Warm Period (RWP). The period from ca. 500 BCE to ca. 450 BCE shows high freshwater and marine planktic diatoms, augmented by silicoflagellates and terrestrial organic matter, with a low dinocyst presence, suggesting a dynamic aquatic environment. The period from ca. 450 BCE to ca. 350 BCE is marked by a high sand content, indicating significant runoff and terrestrial influx, along with increased freshwater and marine planktic diatoms and evidence of human activity in the area. Similarly, the period from ca. 350 BCE to ca. 50 CE is characterized by high sand content and strong anthropogenic influences, with a rise in silicoflagellates, pointing to rising sea levels and high monsoonal precipitation. The period from ca. 50 CE to ca. 400 CE initially shows a decrease in sand and an increase in mud, reflecting a weakening southwest monsoon, likely due to solar variations. However, from ca. 300 CE to ca. 400 CE, sand content rises again, accompanied by high terrestrial influx and dinocysts, while silicoflagellates diminish completely. Thus, despite the dominance of the RWP, the coastal region experienced an extended period of reduced monsoonal activity for a particular span.

1. Introduction

The Holocene epoch, the most recent geological period, has experienced significant climate variability and relative sea-level fluctuations, marked by rapid changes, such as polar cooling, increased aridity, and shifts in atmospheric circulation [1,2]. These global climatic oscillations have been documented in various records, including the Greenland ice cores [3,4], deep sea cores from the North Atlantic [5,6], the Mediterranean [7,8], the Tropical Atlantic [9,10], and the Antarctic regions [11,12], as well as in lake sediments [13,14], peat bogs [15], speleothems [16,17], fossil pollen [18,19], and tree rings [20,21]. Holocene climate variability across different timescales has been linked to various climate forcing mechanisms, such as solar insolation variability, volcanic activity, land use/land cover, atmospheric greenhouse gas concentrations, and climate feedbacks [22,23,24]. The Holocene is currently divided into three stages based on climatic and environmental shifts: the Greenlandian (11.7–8.2 ka), Northgrippian (8.2–4.2 ka), and Meghalayan (4.2 ka–present) stages [25]. The late Holocene (Meghalayan) has been characterized by distinct spatial patterns, with warmer coastal regions and cooler inland areas [26]. However, inconsistencies in the global distribution, and the timing, amplitude, and causes of these fluctuations, highlight the need for more regional studies to better reconstruct and understand climate change [27,28,29,30].

The Meghalayan stage was a time of significant climate shifts, marked by increased variability and dramatic changes in weather patterns experiencing notable fluctuations, reflecting the Earth’s dynamic climate system. During approximately the last two kiloyears (ka) (the Common Era, CE), notable periods of widespread remarkable climate anomalies have included the Roman Warm Period (RWP-ca. 350 BCE–350 CE), the Dark Ages Cold Period (DACP-ca. 400–765 CE)/Late Antique Little Ice Age (LALIA) [31], the Medieval Warm Period (MWP-ca. 800–1250 CE), and the Little Ice Age (ca. 1250–1860 CE). We may acknowledge that the actual timing of these phases is subject to ongoing debates, and likely depends strongly on the study site/region. They are largely attributed to variations in solar activity, ocean–atmosphere interactions, volcanic activity, aerosols, changing sea ice, surface temperatures (SSTs), and greenhouse gas concentrations (GHG) [32,33,34,35,36]. The climatic fluctuations during the late Holocene are broadly categorized either into “cool/wet” or “warm/dry” phases [36,37].

At a global scale, variations in solar irradiance and volcanic activity have been identified as key drivers of natural climate variability on centennial to millennial time scales during the Holocene [22,32,38]. Regionally, preferred phases of internal climate variability patterns, such as the El Niño–Southern Oscillation (ENSO), the Indian Ocean Dipole (IOD), and the North Atlantic Oscillation (NAO), are recognized for their influence over decadal to centennial time scales [39,40] along with (multi)-decadal patterns, like Atlantic Multidecadal Variability (AMO) or Pacific Decadal Variability (PDO, IPO), exerting a notable influence via preconditioning interannual modes, like the ENSO and the NAO, to exhibit a preferred phase (or an elevated/reduced likelihood of their respective phases).

In the Indian context, monsoons are arguably the most critical climatic factor, profoundly influencing the region’s life, ecology, and economy. Agricultural practice in India is primarily dependent on the water provided by the southwest monsoon (SWM), with even winter crops relying on the residual moisture from the summer rains. In the basic sense, paleoclimatology in India is synonymous with the reconstruction of monsoon patterns. Recent studies on past climates have uncovered significant fluctuations in temperature and rainfall across various time scales, from decades to millennia, and across spatial scales, from localized regions to nearly the entire globe. Some of these past climate changes were both widespread and dramatic, far exceeding the variations recorded in historical climate records [41].

The climate fluctuations in India are closely linked to the dynamics of the seasonally reversing monsoon winds, which lead to significant variations in rainfall across regions and time periods. Monsoons play a crucial socio-economic role in India and South Asia, where civilizations have risen and fallen with these climate changes [42,43]. Additionally, the South Asian (SA) monsoons influence the hydrography of the inland, coastal, and ocean regions, and the input of land-derived materials into the oceanic realms. The southwest monsoon (SWM) is particularly important for rainfall and related oceanic processes, such as upwelling and productivity in the Arabian Sea [44]. Various proxies, including biological, sedimentological, geochemical, biogeochemical, and isotopic indicators, have been widely used in reconstructing the Indian monsoon patterns [45].

During the early Holocene, the SWM exhibited significant climatic variability across various regions. In the late Holocene, the SWM weakened, culminating in the large-scale drought around the 4.2 ka event. Both ENSO and IOD are interannual variability phenomena actually manifesting in intra-annual/interseasonal anomalies, while persistent anomalies (e.g., a “persistent El Nino”, are commonly not very likely due to the generally oscillatory nature of those coupled ocean–atmosphere phenomena [46,47,48]. Additionally, the aridification associated with the 4.2 ka event has been accompanied by a negative phase of the North Atlantic Oscillation (NAO), which has been connected to cooler North Atlantic sea surface temperatures (SST) and possibly ice-rafted debris (IRD) events. These factors, through teleconnections, contributed to the weakening of both the Indian and East Asian summer monsoons [25,49,50,51,52].

During the late Holocene, the Indian subcontinent experienced significant changes, evident both in its continental and coastal regions. Numerous studies have documented climate variability and amelioration in terrestrial environments [53,54,55,56,57,58,59]. India’s extensive coastline, stretching approximately 7500 km, has been a lifeline for coastal communities from historical to modern times. This coastline holds ecological, economic, cultural, and historical significance, with ancient ports and coastal towns serving as key centers of trade and cultural exchange throughout history. Sea level and climate changes have had a profound impact on coastal marine processes, as evidenced by various proxies derived from both terrestrial and marine inputs. However, only a limited number of studies have explored the entire coastal zone, particularly the interaction between biotic and abiotic proxies. Research along India’s coast primarily focuses on foraminiferal studies, palynology, sedimentology, geomorphology, and geochemical proxies [60,61,62,63,64,65,66,67,68,69,70,71,72].

Global and continental records reveal significant temporal and spatial climate variability, underscoring the importance of assessing region-specific changes over time. Understanding regional Holocene variability is crucial for contextualizing biotic and environmental responses to anthropogenic climate change. Given the significance of the coastal environments along the southwest coast of India, the Vembanad wetland, a designated Ramsar site in Kerala, was selected for this study. This research focuses on the climatic events that occurred between approximately 500 BCE and 400 CE during the late Holocene. This study employs a multiproxy approach, incorporating statistical analysis of diatom assemblages, palynofacies, and grain size distribution of coastal sediments. This comprehensive approach enabled the identification and characterization of key climate oscillations and regional changes over the past 4000 years, enhancing our understanding of natural climate variability. The primary objective of this study is to examine the limnological and depositional settings of this period, providing insights into the environmental dynamics of the region. The coastal regions of India, particularly along the southwest coast, have undergone significant environmental changes throughout history. These changes include fluctuations in sea levels, variations in sedimentation patterns, and shifts in climate. Such alterations have had profound effects on human settlements, influencing subsistence practices, cultural developments, and human interactions. Understanding these environmental shifts is crucial for reconstructing the historical landscape and its impact on the communities that thrived in this region during the late Holocene. This study of the Vembanad wetland aims to contribute valuable data on how these coastal processes shaped the human and natural history of the area.

2. Regional Setting

2.1. Study Area

Arookutty (AR) is located along the banks of Vembanad Lake, one of India’s largest and most ecologically significant wetland systems, in Kerala. As a designated Ramsar site, Vembanad Lake is part of an intricate network of lakes, canals, and rivers that are vital to the region’s hydrology and ecology. Positioned on the southwest coast of India, Arookutty lies within the expansive Vembanad wetland system, which extends across the districts of Alappuzha, Kottayam, and Ernakulam (Figure 1). The regional and closer view of the location are marked as ‘A’ and ‘B’. The lake spans over 96 kilometers in length and varies between 14 and 15 kilometers in width, making it one of the longest lakes in India [73].

The area around Arookutty is characterized by low-lying plains with a blend of coastal and backwater landscapes. The proximity to the Arabian Sea significantly influences the region’s climate and hydrology, creating a unique environment where freshwater and saline water interact. Vembanad Lake is fed by several rivers, including the Periyar, Meenachil, Pamba, and Achankovil, contributing to its complex hydrodynamic system [73]. The lake’s water levels fluctuate with the seasons, driven by monsoonal rainfall and tidal patterns from the Arabian Sea. During the monsoon, the lake receives substantial freshwater inflow, while in the dry season, saline water from the sea can intrude into the lake.

2.2. Climate

Kerala experiences a tropical monsoon climate, characterized by distinct seasonal variations in its physical and hydrobiological features [74]. The region receives an average annual rainfall of approximately 3000 mm, with the southwest monsoon (June to September) contributing about 71% of this total, and the northeast monsoon (October to December) providing an additional 16%. The remaining rainfall is due to summer showers, resulting in 120–140 rainy days annually in Kerala. Historical precipitation data for the study area is available from 1901 to 2021 (Figure 2) [75]. The annual temperatures in the region typically range between 26 °C and 28.5 °C, with vapor pressure varying from 27 to 20 hPa. The Vembanad wetland in Kochi experiences mixed semi-diurnal tides, with a spring tide range of approximately 1 m [76]. This tidal range gradually decreases both southward and northward from the Cochin barmouth.

2.3. Flora

The region is renowned for its rich biodiversity and lush landscapes, supporting a wide variety of plant species, including tropical rainforests, mangroves, and aquatic vegetation. From coastal areas to high-altitude mountains, the flora includes notable species such as Dipterocarps, Tectona grandis (teak), Dalbergia latifolia (rosewood), and bamboo. Common mangroves in the area include Rhizophora mucronata and Avicennia officinalis. The region’s major plantation crops are Cocos nucifera (coconut) and Hevea brasiliensis (rubber), while spices such as Piper nigrum (black pepper), Elettaria cardamomum (cardamom), and Syzygium aromaticum (clove) are prevalent. Medicinal plants like Azadirachta indica (neem), Ocimum sanctum (tulsi), and Phyllanthus emblica (amla) are commonly found, along with ornamental plants like Bougainvillea, Hibiscus rosa-sinensis, and orchids. The region also hosts endemic species, including Baeolepis nervosa and Strobilanthes kunthiana (Neelakurinji) [77].

3. Materials and Methods

3.1. Field Work, Coring, Sampling Method and Lithological Details

A 150 cm core sample from Arookutty (AR) in the Vembanad wetland, Kerala was collected using a 2.5-inch diameter PVC pipe, with the help of a diver from 1 m deep water. The top 50 cm of the core was discarded due to the disturbance of sediment by human activity. Also, the sample yielded modern ¹⁴C AMS dates for the top 50 cm sediment core. The sediment type comprises sand, silty sand, and sandy silt composition. The remaining core was subsampled at 1 cm intervals, resulting in 100 samples that were analyzed for diatoms, palynofacies, total organic carbon (TOC), stable carbon isotopes, and grain size at the Birbal Sahni Institute of Palaeosciences (BSIP), Lucknow.

3.2. Radiocarbon (¹⁴C) Dating and Chronology

For the Arookutty core, which was raised from the Vembanad wetland, three carbon-rich samples were selected for AMS ¹⁴C radiocarbon dating. The selected samples were from depths of 18 cm (AR-16–-18), 60 cm (AR-58–-60), and 100 cm (AR-98–-100). The graphite preparation was conducted at the BSIP laboratory, where the final graphite targets were generated for radiocarbon dating and isotope characterization. The dating process was completed at the Physical Research Laboratory AMS facility, where the ¹⁴C concentrations (PRL-AURIS) were measured. The samples were assessed using international standards such as OX-II and IAEA-C3. Anthracite powder (C15H11O) was used to prepare blank graphite. The samples were calibrated using Calib 8.2 and IntCal 20 [78,79,80,81].

Table 1. Calibration of the AMS ¹⁴C dates, as well as extrapolation and interpolation of the dates at different depths of the AR sediment core, Arookutty, Kerala, India.

Sample No.	Radiocarbon Age	Cal Year BP (Median Age)	Cal Age Ranges (Cal yr BP)		Cal Age Ranges (BCE-CE)
			1 Sigma	2 Sigma	1 Sigma	2 Sigma
ARD-1	1839 ± 75	1747	1623–1864	1548–1926	89 CE–326 CE	24 CE–402 CE
ARD-3	2302 ± 90	2318	2150–2461	2059–2700	512 BCE–201 BCE	751 BCE–110 BCE
ARD-6	2400 ± 56	2460	2348–2671	2339–2703	722 BCE–399 BCE	754 BCE–390 BCE

presents the median ages along with the calibrated ages. To establish the age–depth relationship, a Poisson process deposition model was applied using Bayesian age–depth modeling [82]. The model indicated that all agreement and convergence indices exceeded the critical thresholds, allowing all three ¹⁴C samples to be retained in the analysis. The Bayesian age–depth model for the Arookutty core, developed using the R package R 4.4.2 rbacon [83], is illustrated in Figure 3. The entire section is dated within the Before Common Era/Common Era (BCE/CE).

3.3. Diatoms

To conduct diatom analysis, 3 g of dry sediment were processed with 35% hydrochloric acid (HCl) and 30% hydrogen peroxide (H₂O₂) to eliminate carbonates and organic matter. The samples were then neutralized with distilled water. Following neutralization, 3 mL of the sample was spread and dried at 100 °C on a hot plate [84]. The dried diatom slides were mounted for microscopic examination, and taxonomic identification was carried out using authoritative literature [85,86,87,88,89,90,91,92,93,94,95,96]. Observations were made using an Olympus BH2 microscope with 600×/1000× magnification under oil immersion, and diatom images were captured using an Olympus DP-74 camera (Olympus, Tokyo, Japan). Depending on the abundance of diatoms, a minimum of 300 to 500 frustules were counted per sample.

3.4. Palynofacies

To isolate palynofacies components, a 5 g sediment sample was first treated with 10% hydrochloric acid (HCl) to remove carbonates. The sample was then repeatedly rinsed with distilled water before being treated with 40% hydrogen fluoride (HF) to dissolve silicates. No oxidizing agents such as KOH or HNO₃ were used for the palynofacies sample processing. After thorough washing and sieving, strewn slides were prepared using Polyvinyl alcohol, which were then mounted with Canada balsam for microscopic examination. The slides were observed under an Olympus BH-2 microscope, with images captured at 200×, 400×, and 1000× magnification using a DP-74 camera [97,98,99,100]. Organic particles were counted until 400–600 observations were made, and then converted into percentages. The recovered organic particles were classified into three main groups: phytoclasts (including degraded, opaque, and structured organic matter), palynomorphs (such as pollen, spores, dinoflagellates, algal and fungal remains, grass remains, and zoomorphs), and amorphous organic matter. These were identified based on the established literature [98,99,100,101,102,103,104,105,106].

3.5. Grain Size

Grain size distribution can indicate discharge energy and environmental characteristics [107,108,109,110,111]. For analysis, 1 g of the sample was treated with 10% hydrochloric acid (HCl) and 30% hydrogen peroxide (H₂O₂) to remove carbonates, organic matter, and soluble salts. The sample was then dispersed using an ultrasonic oscillator for 10 min, with sodium hexametaphosphate ((NaPO₃)₆) as a dispersant [112]. To avoid biases from biogenic silica, including diatoms, which can skew grain size distribution, standard procedure was followed to remove them [113,114]. The residue was treated with 2 M Na₂CO₃ at 50 °C for 24 h and then heated in a 75 °C water bath for 6 h to dissolve diatoms completely. Strewn slides prepared from the supernatant confirmed the absence of diatom traces. Grain size analysis was performed using a Beckman Coulter laser particle size analyzer (LS 13 320) covering a range of 0.004–2000 µm. Data analysis utilized the GRADISTAT program with the Folk and Ward method on the phi (ϕ) scale, providing outputs such as mean, sorting, skewness, kurtosis, and ternary plots [115,116]. Gradistat v8.0 is a software tool widely used in sedimentology to analyze grain size data and provide statistical summaries, including skewness and kurtosis, based on the phi (Φ) scale. As it is known that the Gradistat program plays a key role in analyzing grain size data, including converting micron-sized sediment measurements into the phi (Φ) scale and computing skewness and kurtosis using the Folk and Ward method. The program simplifies the complex process of converting micron grain size data into the phi scale, then applies the Folk and Ward method to estimate skewness and kurtosis. These analyses are critical for understanding sedimentary environments, transport mechanisms, and depositional energy. These metrics are essential in interpreting sedimentary environments. Based on the skewness values using the Gradistat program following the Folk and Ward method, the skewness is classified as very positively skewed: >0.30, positively skewed: 0.10 to 0.30, near symmetrical: −0.10 to 0.10, negatively skewed: −0.30 to −0.10, and very negatively skewed: <−0.30. Similarly for the kurtosis, the values are: very leptokurtic: >3.00, leptokurtic: 2.00 to 3.0, mesokurtic: 1.00 to 2.00, platykurtic: 0.67 to 1.00, and very platykurtic: <0.67. These values have been applied in the present study for the grain size analysis [116].

3.6. Statistical Analysis

Data analysis plays a pivotal role in multi-proxy studies. While simple spreadsheets and graphs are helpful, advanced software tools are often required for deeper interpretation. In this study, cluster analysis and multivariate statistical methods were employed, with a focus on Constrained Cluster Analysis by Sum Squares (CONISS) using Tilia ver. 1.7 [117,118]. This method classifies proxy groupings based on dominance and variations, helping to identify processes responsible for sediment deposition in sedimentary successions.

Among the various multivariate analysis methods, principal component analysis (PCA) was performed using CANOCO 5.0 [119]. PCA is a widely used ordination technique that transforms correlated variables into orthogonal, uncorrelated axes, known as principal components [120]. This technique is particularly valuable for analyzing relationships between variables, such as diatoms, palynofacies, and grain size relative abundances [121]. PCA is highly regarded for analyzing community composition data and performing gradient analysis, allowing ecologists to relate the abundance of plant community components to various environmental gradients like temperature, water availability, light, and sediment texture, or their closely correlated proxies [119,120,121,122,123].

4. Results

4.1. Description of Diatom Diagram

The study identifies four distinct zones for diatoms, categorized as freshwater planktic, marine planktic, freshwater benthic, and marine benthic with 55 genera and 107 species. The diatom assemblage is illustrated in Figure 4 (DZ-I (ca. 500 BCE to 450 BCE), DZ-II (ca. 450 BCE to 350 BCE), DZ-III (ca. 350 BCE to 50 CE), and DZ-IV (ca. 50 CE to 400 CE)) while Figure 5 displays the composite assemblage of diatoms, silicoflagellates, and ascidian spicules. Each zone boundary has a range of uncertainty (

Table 2. Starting and ending age of each zone which are the median of calibrated radiocarbon age. The values in the parentheses are the inter-decile range (2.5–97.5%) of ages which indicate the uncertainty in the zonal boundaries.

Zone	Start	End
Zone-IV	134 CE (62 BCE–332 CE)	428 CE (196 CE–684 CE)
Zone-III	151 BCE (339 BCE–28 CE)	134 CE (62 BCE–332 CE)
Zone-II	361 BCE (557 BCE–166 BCE)	151 BCE (339 BCE–28 CE)
Zone-I	535 BCE (694 BCE–341 BCE)	361 BCE (557 BCE–166 BCE)

) that were calculated using the 2.5 percentile age and 97.5 percentile age. The diatom genera are mainly shown in Figure 4 and in the text, but a table showing the species in the present study is also provided (Supplementary Table S1). The depositional history from ca. 500 BCE to 400 CE reveals a complex diatom assemblage across four distinct zones (DZ-I to DZ-IV), each reflecting variations in freshwater and marine environments.

The period ca. 500 BCE to 450 BCE records freshwater diatoms ranging between 140 and 592 valves, primarily Cyclotella, with occasional Aulacoseira. Marine planktic diatoms vary from 167 to 626 valves, with Thalassiosira as the dominant species (118 to 458 cells), while Podosira, Triceratium, Campylodiscus, Actinocyclus, and Biddulphia are also present in lower frequencies. Freshwater benthic diatoms spanned from 46 to 242 valves, with Nitzschia, Gyrosigma, Cocconeis, and Navicula as the most common taxa. Marine benthic diatoms ranged between 32 and 170 valves, with Diploneis as the most prevalent. Silicoflagellates (4–35 cells) and ascidian spicules appeared occasionally.

The ca. 450 BCE to 350 BCE period showed an increase in diatom abundance. Freshwater diatoms ranged from 156 to 682 valves, dominated by Cyclotella, while Aulacoseira and Melosira were rare. Marine planktic diatoms, predominantly Thalassiosira, ranged from 330 to 840 valves, while Actinocyclus, Podosira, and Triceratium also appeared. Freshwater benthic diatoms ranged from 102 to 235 valves, primarily in Nitzschia, Amphora, Gyrosigma, and Cocconeis. Marine benthic diatoms showed a slight increase, varying from 72 to 172 valves, with Diploneis as the most frequent. Silicoflagellates were present in lower frequencies (3–34 cells).

The period ca. 350 BCE to 50 CE was marked by a continued increase in diatom abundance. Freshwater planktic diatoms ranged from 148 to 698 valves, with Cyclotella still dominant. Marine planktic diatoms were abundant, varying between 196 and 996 valves, with Thalassiosira as the foremost taxon. The zone showed a moderate presence of Actinocyclus, Podosira, Triceratium, and Biddulphia. Freshwater benthic diatoms ranged from 28 to 208 valves, with Nitzschia, Gyrosigma, and Achnanthidium as key taxa. Marine benthic diatoms ranged from 60 to 170 valves, with Diploneis dominating. Silicoflagellates varied between 3 and 33 counts, but were absent in some samples.

Diatom assemblages during ca. 50 CE to 400 CE period were marked by significant variation. Freshwater planktic diatoms ranged from 150 to 598 valves, dominated by Cyclotella, while Aulacoseira and Melosira appeared sporadically. Marine planktic diatoms ranged from 100 to 431 valves, with Thalassiosira most prevalent, followed by Actinocyclus and Podosira. Freshwater benthic diatoms ranged from 32 to 236 valves, with Ulnaria ulna and Gyrosigma as the primary taxa. Marine benthic diatoms showed a lower frequency (14 to 124 valves), with Diploneis decreasing compared to previous zones. Silicoflagellates appeared in the early part but later disappeared entirely.

The results also mark the presence of diatoms which are sporadic, and these include Asteromphalus, Achanthidium, Pinnularia, Bacillaria, Gomphonema, Fragilaria, Rhopalodia, Raphoneis, Craticula, Cymbella, Cymbopleura, Licmophora, Lyrella lyra, Tryblionella, and Entomoneis during the entire period.

4.2. Palynofacies Analysis

The palynofacies from ca. 500 BCE to 450 BCE exhibit notable shifts in organic matter composition, palynomorph abundance, and dinocyst diversity, reflecting changes in environmental conditions (Figure 6).

The period ca. 450 BCE to 350 BCE shows phytoclasts, primarily degraded brown organic matter, range from 15% to 52%, with opaque phytoclasts contributing 5% to 15% and structured organic matter (OM) from 3% to 14%. Oxidized grass remains are minimal (1–4%). Palynomorphs constitute 17% to 60%, with spores more abundant than pollen, accounting for 3% to 6%. Botryococcus algae appear in high percentages (up to 31%) but are absent in many samples. Fungal remains (2–5%), microforaminiferal linings (up to 4%), and copepod egg envelopes (1–9%) are also present. Framboidal pyrite is minimal (<4%), while amorphous organic matter (AOM) varies from 20% to 43%. Dinoflagellate cysts are mostly Gonyaulacoid (6–72 cells), with the dominant taxa being Bitectatodinium spongium, along with Spiniferites, Selenopemphix, and Lejeunecysta.

Among the organic particles during the period ca. 450 BCE to 350 BCE, phytoclasts make up 27% to 50%, with degraded brown OM (12–26%) and opaque phytoclasts (5–14%) as significant components. Structured OM ranges from 4% to 15%, with oxidized grass remaining under 5%. Palynomorphs range from 16% to 44%, dominated by spores and pollen (2–10%). Botryococcus algae vary between 2% and 23% but are absent in many samples, while fungal remains range from 2% to 7%. Marine palynomorphs (up to 8%) and framboidal pyrite (~3%) are present, with AOM content between 24% and 52%. Gonyaulacoid dinocysts dominate (6–42 cells), with sporadic occurrences of Peridinoid dinocysts (2–6 counts) in specific samples.

The period ca. 350 BCE to 50 CE shows high phytoclasts levels, between 21% and 64%, with degraded brown OM (10–30%) and opaque phytoclasts (6–13%). Structured OM ranges from 4% to 18%. Palynomorphs account for 8–36%, with spores, pollen, Thecamoebians, Cyanobacteria, Botryococcus, and fungal remains each contributing up to 10%. Marine palynomorphs (up to 10%) include microforaminiferal linings, copepod egg envelopes, and scolecodonts. Framboidal pyrite is minimal (<2%), and AOM ranges from 9% to 38%. Gonyaulacoid dinocysts dominate (3–64 counts), with Peridinoids less frequent (2–10 counts). Bitectatodinium spongium remains prevalent.

The period ca. 50 CE to 400 CE has significant variation in phytoclast content (28–60%), with degraded brown OM (6–26%) and opaque phytoclasts (4–15%). Structured OM ranges from 1% to 13%, with tracheid and cuticle fragments noted. Oxidized OM varies from 4% to 25%, appearing intermittently. Palynomorphs, predominantly pollen/spores (1–9%) and cyanobacteria (~2%), make up 19–69%. Botryococcus is absent after 299 CE, but ranges from 9% to 48% before that. Marine palynomorphs contribute up to 10%, with AOM ranging from 1% to 33%. Gonyaulacoid dinocysts show high counts (3–102), with Peridinoids present in fewer samples (2–20 counts). Dominant dinocysts include Bitectatodinium spongium, Lejeunecysta, Selenopemphix, and Spiniferites.

4.3. Grain Size Analysis

The grain size analysis across different sedimentary zones from ca. 500 BCE to 400 CE reveals variations in clay, silt, and sand content (Figure 7), as well as sorting, skewness, and kurtosis, reflecting changes in depositional environments over time (Figure 8).

The period ca. 500 BCE to 450 BCE is characterized by low clay content (<5%) and silt content ranging from 5% to 44%, with fine-to-medium silt predominating. Sand is the primary component, making up 51% to 95%, dominated by medium and fine sand, with coarse sand rarely exceeding 10% except for a peak of 14% in sample AR-97-98. Very fine sand content is up to 10%. The grain size distribution shows a mix of unimodal and bimodal patterns, with mean sizes between 1ϕ and 3ϕ. Sorting varies between 0.89 and 1.78, indicating mostly poor sorting with occasional moderate sorting. Skewness is generally fine-skewed to symmetrical (0.08 to 0.46), except for a sample with negative skewness (AR-84-85). Kurtosis ranges from 0.61 to 1.58, with sediments being very leptokurtic to very platykurtic.

The grain size during the period ca. 450 BCE to 350 BCE shows clay content below 2% and silt content between 4% and 32%, which is primarily fine-to-medium silt. Sand dominates with 73% to 95%, largely fine and medium, though coarse sand occasionally reaches 15%. Grain size is generally bimodal, though unimodal instances occur, with mean grain sizes from 1ϕ to 3ϕ. Sorting ranges from 0.58 to 1.61, indicating predominantly poor sorting but with some moderately sorted samples (AR-61-62, AR-66-67). Skewness is positive (0.15 to 0.46), suggesting very fine to fine-skewed distributions. Kurtosis varies from 0.63 to 1.49, with distributions ranging from leptokurtic to very platykurtic.

The sediment is deposited between ca. 350 BCE to 50 CE, this unit maintains low clay content (<5%) and variable silt content (up to 45%). Sand remains high (50% to 98%), with fine and medium sand comprising the bulk (>65%). Coarse sand is less than 13%, and very fine sand is under 15%. The sediment is either unimodal or bimodal, with grain sizes primarily between 1ϕ and 4ϕ. Sorting is generally poor (0.81 to 1.72), but includes some moderately sorted samples. Skewness is mostly positive, though one sample (AR-29-30) exhibits negative skewness. Kurtosis ranges from 0.67 to 1.5, indicating very leptokurtic to very platykurtic distributions.

The final unit spans from ca. 50 CE to 400 CE, and shows clay content up to 7%. Silt content rises initially (ca. 50 CE to 300 CE) but decreases to 3% afterward, while sand content varies widely from 16% to 97%, with a notable increase above 95% after ca. 300 CE. Fine, medium, and coarse sands become dominant in the latter part of this period. Sediment distributions are both unimodal and bimodal, with mean grain sizes from 0ϕ to 5ϕ. Sorting ranges from 0.8 to 1.59, indicating poor sorting between ca. 50 CE to 300 CE, and moderate sorting from ca. 300 CE to 400 CE. Skewness varies from −0.49 to +0.19, reflecting a transition from very coarse to fine and symmetrical distributions, while kurtosis values between 0.65 and 1.42 suggest very platykurtic to leptokurtic distributions.

4.4. Principal Component Analysis (PCA)

4.4.1. Palynofacies and Grain Size

In the present analysis, the data are expressed in proportions or percentages. In environmental studies, this might include species composition, sediment grain sizes, or chemical compositions, where each component (e.g., percentage of a particular diatom species or grain size class) represents a part of a complete sample. In essence, the gradient length is the variation in the dataset along an ecological or environmental gradient. In ordination methods like detrended correspondence analysis (DCA), gradient length is measured in terms of standard deviation (SD) units, representing the turnover in species composition or the change in sample composition along the gradient. A gradient length of 1.1 SD units is relatively short, suggesting that the change in composition along the gradient is moderate rather than extreme. In community ecology, shorter gradient lengths (e.g., under 2 SD units) indicate that a linear ordination method, like principal component analysis (PCA), might be appropriate. This is because the data do not exhibit the high turnover seen over longer gradients (e.g., 4+ SD units), where unimodal (non-linear) methods would typically perform better.

The analyzed data are compositional, rather than any log-ratio transform data, with a gradient length of 1.1 standard deviation (SD) units, which suggests that a linear method, such as principal component analysis (PCA), is recommended, according to the CANOCO advisor. In the current PCA, the first four principal components account for approximately 81% of the variance, with the first two components (PC1 and PC2) alone explaining 65% of the total variance. The pseudo-canonical correlation for PC1 and PC2 is about 82% when analyzing the relationship between palynofacies and grain size (Figure 9A).

In the present study, palynofacies components are highlighted with black arrows, while grain size compositions are marked in red. The arrowheads indicate the steepest increases and highest correlations (when the angle is less than 90° for the palynofacies and grain size) among these components. In Figure 9A, it is observed that opaque phytoclasts align with coarse sand, suggesting that during high-energy conditions, their transport to distant settings is enhanced. Similarly, zoo remains and oxidized grass remains show strong correlations with fine and medium sand. Very fine sand, fine-to-medium silt, medium-to-coarse silt, and clay exhibit a synchronous relationship with microforaminiferal linings and cyanobacteria. Structured organic matter (OM), spores/pollen, and brown degraded OM correlate well with fine and medium sand, as indicated by the angles between them, which are less than 90°. Additionally, marine palynomorphs, such as microforaminiferal linings, are negatively correlated with amorphous OM (AOM), zoo remains, and Botryococcus, evidenced by angles greater than 90°, indicating a negative correlation according to the PCA. Palynofacies components located near the origin show zero or low correlation, as indicated by the arrowheads of framboidal pyrite, scolecodonts, tintinnids, and copepod egg envelopes. Thus, the intricate relation between palynofacies and grain size is well established and correlated [103,124,125].

4.4.2. Diatom and Grain Size

The response data are compositional and have a gradient length of 0.7 standard deviation (SD) units, indicating that a linear method, such as principal component analysis (PCA), is appropriate as per the CANOCO advisor. In the PCA conducted, the first four principal components explained about 75% of the variance, with the first two components (PC1 and PC2) accounting for 53% of the total variance. The pseudo-canonical correlation between PC1 and PC2, when examining the relationship between diatom and grain size, is approximately 52% (Figure 9B).

For the present PCA, the raw diatom counts were converted to percentages. Only diatom data present at a minimum of 5% in at least one sample were included, to reduce the influence of numerous zero values and avoid redundancy. In the analysis, grain size is represented by red arrows, while blue arrows denote diatoms. Coarse sand shows a positive correlation with Cocconeis spp., Campylodiscus, and Cyclotella, but a negative correlation with Thalassiosira spp. and Trachyneis sp. Fine and medium sand exhibit a strong correlation with Amphora spp., Diploneis spp., and Nitzschia spp. Very fine sand, silt, and clay are prominently correlated with Actinocyclus, Gyrosigma, and Podosira taxa. The PCA reveals a clear relationship between hydrologic characteristics, flow conditions, sediment size, and the distribution patterns of diatoms [91,94,126].

5. Synthesis of the Proxy Record (Palynofacies, Diatoms, and Grain Size Analysis) for Paleolimnology and Natural Versus Anthropogenic Influx

To assess process-driven changes, a ternary plot of phytoclasts, palynomorphs, and AOM percentages was constructed for the Arookutty core (Figure 10). The diagram shows that most samples from the depositional period fall within the heterolithic oxic region, indicating a mix of terrestrial and marine inputs [125]. A substantial number of samples also correspond to a mud-dominated oxic coastal region, likely reflecting a period of sea-level rise during warmer conditions, when both terrestrial and marine components were abundant. This suggests that high levels of terrestrial organic matter reached the sea, enhancing the overall organic content, while finer marine sediments were deposited on land, increasing the presence of both terrestrial and marine palynomorphs in this environment.

Integrating diatom assemblages, palynofacies, and grain size analysis provides insights into monsoonal variability along the Indian coast. Diatoms indicate changes in salinity, nutrients, and turbidity, with freshwater diatoms peaking during monsoon seasons and marine diatoms dominating in drier periods. Palynofacies analysis differentiates between terrestrial and marine organic matter, with strong monsoons bringing more terrestrial material to coastal waters, while weaker monsoons lead to increased marine input. The grain size distribution, shaped by monsoonal intensity, shows coarser sediments during high-energy monsoons and finer sediments during low-energy periods, reflecting the strength and variability of monsoonal activity.

The period from ca. 500 BCE to 400 CE in Arookutty reveals four distinct phases of deposition driven by climatic changes during the late Holocene. This entire period is characterized by a high presence of both freshwater and marine planktic diatoms, with freshwater diatoms being particularly dominant in the first phase (ca. 500 BCE to 450 BCE). During this phase, both freshwater and marine benthic diatoms are moderately represented, while the presence of silicoflagellates indicates ongoing marine incursions, likely due to fluctuating interactions between marine and freshwater, influenced by varying rainy days during the southwest monsoon (SWM). These fluctuations are tied to changing pressure, temperature, and moisture conditions. Palynofacies analysis during this time, especially the presence of opaque phytoclasts and degraded brown organic matter, reflects a strong terrestrial influence. High terrestrial input, evidenced by the abundance of Botryococcus, other pollen, spores, and fungal remains, suggests a dominance of terrestrial over marine influx. Grain size analysis also shows high but variable sand content, indicating changes in transport and deposition dynamics. Low dinocyst abundance points to reduced salinity, likely due to significant freshwater input from terrestrial sources, diluting marine salinity. This period corresponds to the early phase of the Roman Warm Period (RWP) noted in European history, with evidence of a similar warming episode recorded in the coastal settings of Kerala during this time [36,73,127].

The phase from ca. 450 BCE to 350 BCE is marked by a high sand content and a low mud content, indicating significant terrestrial influx during this period. The palynofacies data align with grain size analysis, showing a substantial transport of phytoclasts from terrestrial regions. This phase also features a similar but increased presence of freshwater and marine planktic diatoms. Environmentally sensitive diatom species such as Cyclotella menenghiana, Nitzschia palea, Amphora spp., Achnanthidium spp., and Navicula spp., which are indicative of human activity near the Vembanad wetland, are also observed [128,129,130,131]. Additionally, there is a rise in freshwater and marine benthic diatom assemblages, along with a moderate increase in silicoflagellates. The silicoflagellates suggest warm water conditions during this period [132]. The palynofacies data reveal a high proportion of phytoclasts and palynomorphs, accounting for more than 70% of the total content. The increase in Glomus sp. and Alternaria sp. supports evidence of local erosion, while the presence of coprophilous fungi indicates anthropogenic activity in the suburbs of Arookutty and the Vembanad wetland. These findings are consistent with other records from the Core Monsoon Zone (CMZ), the Saurashtra coast, and the Kerala coast [49,50,71,133], and align with the Roman Warm Period (RWP) and a period of high monsoonal activity along the Indian coast [70,71,127].

Phase III, spanning from ca. 350 BCE to 50 CE, is characterized by high sand content, indicating high-energy conditions during deposition, as revealed by grain size analysis. This period aligns with the peak of the Roman Warm Period (RWP), a warming phase recorded across different parts of India [47,48]. The high sand content suggests increased terrestrial matter transport and a rising relative sea level, evidenced by the abundance of Gonyaulacoid dinoflagellate cysts, which indicate high primary productivity along the coast [70].

The diatom assemblage, including Cyclotella, Thalassiosira, Diploneis, Gyrosigma, Nitzschia, Achnanthidium, Cocconeis, Amphora, and Navicula, reflects high anthropogenic influence during this time. The prevalence of both freshwater and marine planktic diatoms points to high water columns, likely driven by an enhanced southwest monsoon during this warm and humid climate phase [91,92,126]. The moderate presence of freshwater and marine benthic diatoms suggests small-scale transport from both land and sea, a consequence of rising sea levels due to increased precipitation. The rise in silicoflagellates further supports the evidence of rising sea levels. Overall, this phase reflects a period of high monsoonal precipitation and limnological conditions influenced by both natural factors and human activity at Arookutty in the Vembanad wetland.

The phase from ca. 50 CE to 400 CE exhibits a decrease in sand content with a corresponding increase in silt and mud, although sand content rises significantly between ca. 300 CE and 400 CE. This period can be divided into two distinct depositional phases: ca. 50 CE to 300 CE and ca. 300 CE to 400 CE. During the first phase, the wetland experienced low water levels, leading to the proliferation of Botryococcus algae in brackish marine conditions [134]. Terrestrial phytoclasts and Gonyaulacoid dinocysts also decreased significantly, likely due to high turbidity and the dominance of Botryococcus. Although freshwater and marine planktic diatoms remained relatively abundant, their numbers decreased compared to earlier periods. Freshwater benthic diatoms increased, while marine benthic assemblages declined. Silicoflagellates diminished significantly and eventually disappeared, indicating a period of low rainfall and reduced water levels, likely due to a weakening southwest monsoon along the coastal region. While the exact causes remain unclear due to the lack of other biogeochemical proxies, it is speculated that variations in solar activity may be responsible for the changing climate [135].

In the later phase, from ca. 300 CE to 400 CE, there is a marked increase in medium and coarse sand content and a decrease in mud, indicating higher runoff. This period is characterized by a resurgence of terrestrial phytoclasts, including opaque and degraded brown organic matter, reflecting increased terrestrial influx. Botryococcus algae, which thrive in calm, low-energy waters, are absent in this phase, while dinocysts reappear in greater numbers, suggesting higher primary productivity. Overall, diatom abundance decreases, likely due to the high-energy conditions indicated by the grain size data, which would have disrupted the accumulation of delicate diatom valves.

This phase indicates that even during the Roman Warm Period (RWP), the southwest coast of India experienced extended periods of reduced sedimentation due to weaker southwest monsoons. However, conditions began to recover during the final phase from ca. 300 CE to 400 CE, aligning with global records of the RWP. Finally, by integrating multiple proxies, we can reconstruct past monsoonal patterns and better understand their effects on coastal environments along the Indian coast. Late Holocene sea level changes along the Kerala coast were shaped by a combination of global sea level fluctuations, regional tectonics, and monsoonal variability. During this period, stable sea levels facilitated the accumulation of coastal sediments, leading to the formation of beach ridges, organic-rich layers, lagoons, estuaries, and barrier islands. By around 2000 years ago, sea levels had largely stabilized, with only minor fluctuations, allowing for the development of these coastal features.

The strength and variability of the southwest monsoon played a key role in shaping Kerala’s coastal landscape during the late Holocene. Strong monsoons led to coastal erosion and sediment redistribution, while weaker monsoons allowed for the deposition of finer sediments in protected areas. Paleoenvironmental studies, including palynofacies and diatom analyses, have helped reconstruct past sea levels and coastal conditions, indicating that higher sea levels coincided with increased marine influence, while lower sea levels were associated with more terrestrial input.

In the holistic view of the Arookutty core from Vembanad wetland we found that there has been a significant change in the environmental setup from ca. 500 BCE to 400 CE. However, using the diatoms and palynofacies the human impact has also been observed during the period based on stress indicator taxa around the Vembanad wetland. As it is known the period 500 BCE to 200 CE has been globally recorded as Roman Warm Period (RWP) during which ancient trade connections have been developed with the Roman Empire, Arabia, and Southeast Asia [136,137,138,139]. The study area and the surroundings of the Vembanad wetland is also recognized for its natural and historical connections. The Vembanad Lake in Kerala, a vital part of the state’s extensive waterway network, has historically served as a key hub for both regional and international trade. Its proximity to ancient port towns like Muziris (near modern-day Kodungallur) positioned it strategically within trade networks that linked Kerala to distant regions, including the Roman Empire, Arabia, and Southeast Asia. Archeological excavations around Vembanad have uncovered pottery, semi-precious stones, and other artifacts, attesting to these early trade connections and the cultural exchanges they fostered [137,138].

Significant finds, such as Roman coins and amphora fragments, indicate active trade relationships with Roman merchants who frequented the Kerala coast for valuable spices, particularly black pepper. This trade likely left a lasting impact on the area’s cultural practices, urban development, and economic systems. Additionally, archeological evidence of Iron Age settlements around the Vembanad includes items like iron tools, pottery, and beads, which provide a glimpse into the technological advancements and evolving material culture of these early communities [139,140]. Such discoveries help archeologists trace the development of settlement patterns and the social organization in the region.

These sea level changes also had significant impacts on human settlements along the Kerala coast, influencing resource availability, land use patterns, and potentially causing displacement due to rising sea levels and coastal erosion.

6. Conclusions

The Arookutty core analysis from the Vembanad wetland reveals key shifts in environmental and climatic conditions along Kerala’s coast throughout the late Holocene, particularly emphasizing the impact of monsoon variability, sea-level changes, and early human activity.

• This period (ca. 500 BCE to 350 BCE) reflects a diverse aquatic ecosystem influenced by alternating freshwater and marine conditions, pointing to variable monsoon intensity. Increased sand content and environmentally sensitive diatoms indicate significant terrestrial runoff and suggest early human impacts on the wetland.
• The time span ca. 350 BCE to 50 CE shows continuously high sand levels, highlighting strong monsoon-driven runoff and elevated marine influence, likely due to rising sea levels. The presence of anthropogenically linked diatom species suggests growing human interaction with the wetland.
• In the period from ca. 50 CE to 400 CE, the region is initially marked by decreased sand and increased mud, this period shows a shift to lower-energy conditions, possibly due to a weakened SWM because of the variations in solar insolation during the Holocene. Later, sand content rises again, signaling intensified monsoons and higher terrestrial input, alongside dinocysts that reflect increased coastal productivity fueled by nutrient-rich runoff.
• During the aforementioned period close to historic port towns like Muziris, Vembanad was strategically placed within trade routes connecting Kerala to distant regions, including the Roman Empire, Arabia, and Southeast Asia. Archeological discoveries, including Roman coins, amphora fragments, pottery, and semi-precious stones, highlight these extensive trade connections and cultural exchanges.
• Additionally, Iron Age artifacts such as tools, pottery, and beads reveal early technological advancements and settlement patterns around Vembanad, offering insights into the area’s evolving material culture and social organization, reflecting its historical significance in trade and community during the deposition period.

Based on the above findings the Vembanad wetland can further be investigated by analyzing how past communities adapted to monsoon-driven environmental changes. The human–environment interactions using diatom and sediment profiles could provide a clearer picture of anthropogenic impacts on coastal ecosystems over time. Expanding archeological investigations around Vembanad and its ancient trade connections with Rome, Arabia, and Southeast Asia could enhance our understanding of cultural exchanges and the socio-economic structure of early settlements. These perspectives can contribute valuable data for regional climate adaptation, ecosystem management, and heritage conservation along Kerala’s coast.