Can artificially induced habitat complexity alter macroinvertebrates diversity? A case study from a freshwater wetland ecosystem

Habitat complexity can enhance the resilience of wetlands against environmental stressors such as extreme weather events, pollution, and habitat loss. The introduction of artificial induced complexity (AIC) can play a significant role in reshaping the macroinvertebrate communities within wetland ecosystems by enhancing habitat quality in protected areas. Therefore, this study was designed to examine the variation of macroinvertebrates community structure in artificially induced complex sanctuary site (SS), partially protected (PP) and open sites (OS) from July 2019-April 2020. AIC in the sanctuary sites was established through the installation of cemented hexapods and ring pipes. Over the study period, a total of 665 macroinvertebrates were gathered, with 55.55% originating from SS, 31.14% from PP, and 18.21% from OS sites. The community consists mainly of Lymnaea acuminatatea and Tubifex tubifex, with the most abundant species being Limnodrillus hoffmeisteri and Branchiura sowerbyi. A notable positive impact of AIC was evident in the increased total abundance and diversity indices of macroinvertebrate communities. The Analysis of Similarity (ANOSIM) revealed significant distinctions in community structures among various intervention types, which was further corroborated by a non-metric multidimensional scaling (nMDS) plot. Similarity of Percentage Analysis (SIMPER) highlighted that Limnodrillus hoffmeisteri made the most significant contribution to the dissimilarity observed among the different intervention types. Canonical Correspondence Analysis (CCA) revealed a close association between the structure of the macroinvertebrate community and three key ecological factors: periphyton biomass, macrophyte cover, and sediment properties. These findings could offer a more effective approach for managers and policymakers engaged in the conservation of macroinvertebrates and the sustainable management of fisheries resources within wetland ecosystems.


Introduction
Complex habitats are helpful to enhance abundance and diversity of aquatic organisms compared with simple habitats for their role in reducing physical stress, increasing food resources or niche availability and surface area, decreasing competition, and shelter (Bartholomew et al 2000, Wang et al 2006, Matias et al 2010, Shimabukuro and Henry 2011, Kim et al 2019, Li and Asner 2023).Habitat complexity in wetland ecosystems provide a wide variety of niches and microhabitats, which boosts biodiversity and promotes species resilience (O'connor 1991(O'connor , Wang et al 2022)).This complexity helps organisms adapt to shifting environmental conditions by giving them a place to live and resources.The intricacy of habitats enables animals to locate ideal settings for feeding, nesting, and protection.This makes it more likely that one will survive throughout alterations or disruptions to the environment.Furthermore, more diverse food supplies and resources are frequently found in complex environments, which might act as a buffer against changes in the availability of certain resources (O'connor 1991(O'connor , Sueiro et al 2011)).Besides, the complex organization of wetland environments promotes ecological stability, guaranteeing species survival and persistence in the face of shifting obstacles (Mykrä et al 2011, St. Pierre & Kovalenko 2014).Species in subtropical region use the wide range of niches and microhabitats that are accessible to them to adapt to the complexity of their environments.Subtropical species often exhibit plasticity in their behaviors and life history traits, allowing them to thrive in dynamic habitats (Wang et al 2022).Furthermore, the complex structure of habitats in subtropic regions promotes interactions among species, facilitating resilience through processes such as competition, predation, and mutualism (Wang et al 2022).
The artificially induced complexity (AIC) can be enhanced by permanent or semi-permanent physical structures (McCoy andBell 1991, Downes et al 1998).Increasing the AIC could benefit the conservation of aquatic organisms in the wetland ecosystems characterized by periodic drought and flood events (Schwartz and Jenkins 2000).The creation of habitat heterogeneity through substratum complexity is known to reshape the algal community in an aquatic environment, which can also strongly influence the quality and quantity of macroinvertebrates community and fish assemblage (Merz et al 2005, Lusardi et al 2018, Atique et al 2020, Kim et al 2021).
Improvement in the macroinvertebrates community can increase the function of the ecosystem, which can help in the success of conservation programs (Batzer et al 2006, Rader 2001, Haque et al 2020).Macroinvertebrates are an essential component of the food web, providing sustenance to other invertebrates, fish, and birds.Furthermore, they assist in nutrient cycling and control of algae and plankton (Harvey andAssociates 2005, Atique andAn 2019).Their responses to physicochemical changes in the aquatic environment brand them reliable bioindicators (Brooks 2000, Anderson and Smith 2004, Furse et al 2006, Atique et al 2020).Therefore, AIC brought by incorporating complex-shaped structures could provide practical solutions to the sustainable conservation of aquatic resources in floodplain ecosystem.
Chalan Beel (CB) is the largest wetland in Bangladesh.It is home to a diverse variety of plants and animals (Sayeed et al 2015).It is rapidly losing its rich biodiversity due to overfishing, fishing by dewatering, habitat degradation, anthropogenic pollutants, siltation, and obstruction of migratory routes of fish species by the erection of water control structures (Hossain et al 2009, Galib et al 2018).During monsoon, CB habitat becomes more homogeneous and interconnected, enhancing the dispersal of aquatic organisms.However, during the dry season, dispersal of aquatic organisms becomes restricted due to the conversion of this waterbody into several isolated areas that renders the aquatic organisms more vulnerable to several stressors.In that situation, the establishment of sanctuaries (protected areas) becomes a suitable countermeasure to preserve and conserve the depleting biodiversity.To increase the production of fish in the CB, four sanctuaries were dug at a cost of taka two million under Singra upazila during the last decade.Nevertheless, despite the establishment of several sanctuaries in CB, there are no favorable outcomes due to the lack of proper planning and suitable intervention designs.Furthermore, there is no adequate monitoring system to evaluate the performance of established sanctuaries.
Most of the sanctuaries established in CB remained limited in achieving the anticipated conservation work due to low complexity and habitat quality and cannot fully conserve the threatened species.The ineffectiveness of the constructed sanctuary for the conservation of aquatic habitat can be attributed to habitat degradation caused by elevated siltation rates, unregulated use of herbicides, insecticides, and chemical fertilizers on croplands, as well as the full removal of water.Despite the existence of certain semi-permanent sanctuaries, their effectiveness in preserving the habitat quality of CB has been compromised by the indiscriminate removal of eggs, spawns, fry, undersized fish, and brood-stocks during the dry season (Hossain et al 2009).Therefore, the establishment of a sanctuary that can provide proper habitat heterogeneity is inevitable.
Research on macroinvertebrate diversity in Bangladesh has primarily concentrated on estuaries, freshwater ponds (Chakma et al 2015, Haider et al 2017, Sharmin et al 2018) and coastal wetlands (Mamun et al 2022, Khatun et al 2023).However, freshwater floodplain wetlands remain unexplored in terms of macroinvertebrate diversity.Additionally, there is a lack of studies addressing habitat complexity and its impact on macroinvertebrate communities in Chalan Beel (CB) of Bangladesh, which is the largest and most crucial wetland system in North Central Bangladesh, covering approximately 375 km 2 during the monsoon season, and sustaining around 5 million people primarily through fisheries and agricultural activities.CB primarily comprises numerous rivers and their tributaries, forming an extensive water network across the area.However, during the monsoon season (June-October), these rivers and channels lose their distinctiveness due to floodwater influx, merging into a vast expanse of water, becoming common property resources for all.In efforts to safeguard and preserve the ecosystem, government and non-government organizations implement various interventions e.g., sanctuaries (protected) and partial protection.Partially protected (PP) areas are delineated by encircling them with bamboo poles, while protected areas are reserved zones where hunting and fishing are prohibited or regulated, and open water or sites (OS) are accessible for fishing or activities by anyone.The objectives of this study were to describe spatial variations in macroinvertebrate assemblages in three distinct intervention types (protected, partially protected, and open for fishing sites), and to assess the performance of established sanctuaries in improving the ecological efficiency of CB.We hypothesized that environmental properties would differ in artificially intervened structures (hypothesis 1), leading to significant differences in macroinvertebrate community assemblages among intervention types (hypothesis 2).Furthermore, we hypothesized that specific macroinvertebrate species would predominate under certain environmental characteristics (hypothesis 3).Lastly, we hypothesized that artificial habitat structures would facilitate the establishment of macroinvertebrates, as measured by their abundance and diversity (hypothesis 4).

Study area
The Chalan Beel is located between 24.35°to 24.70°N and 89.10°to 89.35°E.This study was conducted in a tributary of Atrai River (AR) and Gumani River (GR) that flows through the largest wetland in Bangladesh, i.e.Chalan Beel (CB) floodplain (figure 1) for 10 months from July 2019-April 2020.The water depth of this river is varied seasonally due to the heavy rainfall during the monsoon.Both the rivers are also influenced by floods causing inundations.Flooding of these rivers along with the others turns the floodplain into a massive water resource.During the high flood period, the inundated CB area reaches approximately 300-320 km 2 , which is shrunk to a water volume of 50-75 km 2 during the dry season (Hossain et al 2009, Samad et al 2009).Thus, during the dry season, small rivers flow through the CB are the only water holding areas, therefore, overexploitation becomes the main threat to the biodiversity of these rivers and its adjacent wetland areas with low water levels.The CB serves as a confluence for multiple smaller waterways and is subsequently drained by south-flowing channels, ultimately emptying into the Padma and Brahmaputra Rivers ( Samad et al 2006).

Intervention types
During the present study, we followed a manipulative approach to select the intervention types.Two locations were chosen along with the AR and GR, and the manipulative sites were designated as a sanctuary site (SS) with an area of 0.5 hectares each.Different structures were used for increasing the complexity of the SS sites (figure 2).Initially, the area chosen for AIC was demarked by encircled bamboo poles (200 pieces), and in the encircled area, several small size bamboos (150 pieces) with tree branches (150 pieces) were stacked at a given distance of 3 m.Finally, several cemented hexapods (50 pieces) and ring pipes (50 pieces) were placed in the encircled area.A generalized view of sanctuary sites with the arrangement of different artificial structures is shown in figure 3.
The manipulated site was subsequently compared with a partially protected (PP) temporary sanctuary site (SS) and an open site (OS).Three PP sites, characterized solely by encircled areas with bamboo poles, were chosen.Additionally, four OS sites, devoid of any manipulative complexities, were selected.The distances between the SS, PP, and OS sites ranged from approximately 5 to 5.5 km.Fishing activities were restricted at the SS site, while the PP and OP sites were open for fishing by local people.

Sampling and identification of macroinvertebrates
Samples of macroinvertebrates were collected monthly for 10 months during July 2019-April 2020.For each habitat type, three replicated samples were collected.Benthic macroinvertebrates (each with a sampling area of 1 m 2 ) were collected with a Peterson grab, meanwhile, epiphytic macroinvertebrates (each with a sampling area of 1 m 2 ) were sampled from the water column with a kick net (mesh of 420 μm).Samples were collected by allocating minor spatial differences in water depth (relatively shallow, moderate, and deep).The collected samples were preserved in 10% formaldehyde and immediately transported to the laboratory of the Department of Fisheries, University of Rajshahi, Bangladesh.In the laboratory, the macroinvertebrates were sorted, counted, and identified to the lowest possible taxonomic level.An attempt was made to identify macroinvertebrates to the species level or, if not feasible, to the genus level, utilizing taxonomic keys outlined in Gill (1998), Gooderham & Tsyrlin (2002), Merritt et al (2008), Oscoz et al (2011), Kotpal (2012), Sharma et al (2013), Fry (2021), and Muir and Hossain (2014).Species validation was conducted using the well-known taxonomic database, WoRMS.

Environmental parameters analysis
Surface water samples (500 ml) were collected from each study site during each sampling event to assess water quality parameters using standard procedures.The water temperature (WT) was measured using a Celsius thermometer, pH was assessed using an electronic pH meter (Jenway 3020, UK), and dissolved oxygen (DO) was determined using a HACH Kit Box (Model DR-2010, USA).Water depth at each study site was recorded using a measuring tape, with five measurements taken from the nearest area of the bank to the deepest part of the channel, and the average depth was calculated.Water velocity was measured using a digital water flow meter (Model: 20307R6C, USA).Periphyton samples were collected by scraping about five-six square centimeter surface areas of hexapods, ring pipes, and macrophytes at SS site using a brush.However, at PP and OS site, stone/boulders, other available substrates, and macrophytes (if any) were scraped to collect periphyton samples.The collected samples were preserved in 4%-5% formalin in separate tubes.The samples were filtered with a Whatman filter paper and dried in a drier at 70 °C for 24 h.The dried specimens were ashed at 550 °C for 1 h using a muffle furnace for ash determination.The dry weight was then expressed as mg/cm 2 of periphyton biomass.Cover of aquatic macrophytes was estimated visually (Cai et al 2012) by scoring on an arbitrary scale of 0-4 (classes: 0,1-25, 26-50, 51-75, and 76%-100% cover of the sampling station, respectively).
At each sampling location, three replicated sediment samples were collected using a 0.1 m 2 van Veen grab.Approximately 500 g of sediment samples from each location was collected and kept in a Ziplock bag to transport to the laboratory.The collected sediment samples were air-dried and sieved with a 0.5 mm mesh sieve.Finally, oven-dried (105 °C for 24 h) sediment samples were analyzed for percentage of sand, silt, and clay by hydrometer method described by Huq and Alam (2005).The organic matter content of sediment was measured by Walkey and Black (1934) wet oxidation process.

Diversity indices
The datasets of macroinvertebrates abundance collected during the study period were subjected to the analysis of species richness, density (ind./m 2 ), Shannon index, evenness index, and Margalef index by the following formula: The Shannon diversity index: H ln

Statistical analyses
The data was submitted to the normality test to check the normal distribution prior to the detailed statistical analysis.To test the differences in environmental parameters, total abundance, species richness, density, and diversity indices among the intervention types, one-way ANOVA was performed using SPSS 20.0 following a post-hoc DMRT test.The normality and homoscedasticity of data were tested with the Kolmogorov-Smirnov test and the Shapiro-Wilk test before analysis.To deal with the zero values in species abundance, the whole species abundance data were transformed to log (x + 1) and environmental data were converted to Square-root to normalize the data set.Differences in macroinvertebrates assemblage based on the habitat types were determined by the Analysis of similarity (ANOSIM) using the function anosim and visualized by nonmetric multidimensional scaling (NMDS) ordination on the Bray-Curtis distance measure using the function metaMDS in the R package vegan 3.6.3(R Core Team, 2018).Similarity percentage (SIMPER) analysis was also calculated to identify contributory species that caused significant differences in community assemblages among the intervention types using the function simper in the R package vegan 3.6.3(R Core Team 2018).
We used the canonical correspondence analysis (CCA) to determine the relationship between environmental parameters and the macroinvertebrates abundance.Suitability of CCA as a unimodal model to discover this relationship was revealed through a detrended correspondence analysis (DCA) (length of the first axis > 2 SD).Multi-collinearity was conducted for all the environmental parameters and the parameters having a variance inflation factor (VIF =1 / (1 − R2) > 10 were considered for CCA analysis (Neter et al 1990).Furthermore, those taxa with a relative abundance of <1% were excluded for CCA analysis to minimize the effects of less abundant taxa.CCA was performed using the function cca, and the function ANOVA tested the significance of CCA model and axes with 9999 permutations in the R package vegan 3.6.3(R Core Team, Vienna, Austria).

Environmental variables
The mean and standard deviations of the environmental variables among the studied locations are represented in table 1.Several environmental variables exhibited minimal changes among the different habitat types, with some displaying significant variations (P < 0.05).The mean values for water temperature (WT), dissolved oxygen (DO), pH, water depth, and velocity remained relatively stable across the intervention types (table 1).Notably, the highest mean periphyton biomass was significantly more abundant at SS, while the lowest biomass was observed at OS. Macrophyte cover showed minor variations between SS and PP but displayed substantial differences with OS.Sand particle content varied considerably across the habitat types, with the highest values recorded at OS. Similarly, silt, clay, and organic matter content exhibited significant variations among the habitat types, with the highest mean values of these particles found at SS and the lowest at OS.

Community composition and abundance
Throughout this study, a total of 665 macroinvertebrates was gathered, representing three distinct phyla, six classes, 16 orders, 40 families, and a diverse array of 55 species (table 2).Among these phyla, arthropods emerged as the most prominent, with a total of 29 species, while Mollusca accounted for 14 species, and Annelida featured 12 species.Remarkably, when considering species abundance, Annelida ranked the highest with 308 individuals, closely followed by Arthropods with 205 individuals, and Mollusca with 152 individuals.The most frequent species was Lymnaea acuminatea (83.30%) and Tubifex tubifex (83.30%), whereas the most abundant species were Limnodrillus hoffmeisteri (16.70%) and Branchiura sowerbyi (12.49%).Regarding the habitat types, 55.55% (336) of the individuals were recorded from SS belonging to 55 species.The most abundant species was Limnodrillus hoffmeisteri (19.74%) followed by Branchiura sowerbyi (13.90%) and Lymnaea acuminatea (9.80%).
In PP, 31.14% (207) of the individuals were collected which included 53 species with the most abundant was Limnodrillus hoffmeisteri (20.82%) followed by Branchiura sowerbyi (17.54%).However, in OS, the percentage of individuals collected was 18.21% (121) consisting of 44 species and the most abundant species was Chironomus sp.(32.51%) followed by Ephemera sp.(17.80%).The total abundance of macroinvertebrates was significantly (F = 11.104,P = 0.000) higher at SS compared to PP and OS.

Species distribution
Two types of distribution patterns were identified based on community structure (species composition, abundance and diversity) of macroinvertebrates among the habitat types.This notable difference within each habitat types is also shown in nMDS plot (stress = 0.14), calculated based on the same data of species abundance matrix.Species abundance has shown a slight overlap between SS and PP habitat types which showed their complete distinction from OS habitat type (figure 5).Analysis of similarity revealed significant differences in species abundance based on different habitat types (ANOSIM, Global R = 0.6156, P = 0.001) at the permutations of 9999.Pairwise tests of ANOSIM between SS and PP were insignificant (ANOSIM, Global R = 0.07844, P = 0.0863), while the differences were significant between SS, OS (ANOSIM, Global R = 0.922, P = 0.001) and PP, OS (ANOSIM, Global R = 0.8433, P = 0.002).SIMPER analysis revealed an overall average dissimilarly of 52.49%, 77.05% and 76.39% between SS-PP, SS-OS, and PP-OS sites, respectively, whereas the most contributory taxa were B. sowerbyi (5.00%), L. hoffmeisteri (5.37%), Chironomus sp.(6.73%), respectively  (table 3).However, the overall average dissimilarity among the intervention types was 68.64% and the most contributory taxon was L. hoffmeisteri (5.13%).

Macroinvertebrates assemblage and environmental parameters
Twenty-six species had a relative abundance >1% and were therefore selected for CCA analysis (table 1).All the environmental parameters showed no apparent multi-collinearity (VIF <10) among them, therefore, all these parameters were included in the CCA analysis of macroinvertebrate species assemblage and environmental factors (figure 6).Among the eigenvalues of all CCA axis, only the first (Proportion of variance explained 0.458) axis was significant (P < 382 0.01, ANOVA).It displayed a strong correlation with the periphyton biomass, macrophytes, sand, clay and organic matter (correlations of − 0.52, − 0.72, 0.48, − 0.45 and − 0.46, respectively).This axis mainly reflected the dominance of intervention types on the distribution of microinvertebrates, which supported the second hypothesis.Since the proportion of was high in the OS sites (see table 1), while proportions of macrophytes and silt increased in the PP, and even more so in the SS sites (table 1), this axis reflects a gradient from OS sites with high sand towards the positive side of the axis, to the artificial habitats PP and SS with high proportions of macrophytes towards the negative side of the axis.For  instance, among the selected 26 species, T. tubifex, Chironomus sp., Ephemera sp., Ephemerella sp., Heptagenia sp. and Corixa sp. were positively correlated with axis I, indicating the influence of sand particles in their distribution.In contrast, the remaining 20 species (Hirudo sp., Pheretima sp., Branchiodrilus sp., Dero sp., Nais sp., B. sowerbyi, L. hoffmeisteri, Coenagrion sp., Lestes sp., Isoperla sp., Brachycentrus sp., B. bengalensis, L. acuminatea, Planorbis sp., Bithynia sp., Goniobasis sp., T. tuberculata, Brotia sp., L. marginalis and P. corrugata) seemed to favour the AIC that provided higher periphyton biomass and macrophyte cover.These species are mainly detritivores and grazing macroinvertebrates associated with areas of dense algae and macrophytes cover.

Discussion
4.1.Impacts of AIC (SS) on habitat ecological variables Our findings clearly indicated a substantial positive impact of complex structures, Artificial Induced Complexity (AIC), on various aspects of the study sites.This influence extends periphyton biomass, macrophyte cover, as well as the composition of sand, silt, clay, and organic matter within these locations.However, being a continuous water area in Chalan Beel (CB), WT, DO, pH, water depth, and velocity values were considerably similar among the study sites with different interventions (SS, OS and PP).The key ecological variables specially WT, DO, pH were within the suitable range indicated suitable ecosystem which supporting rich biodiversity.
However, these values of water quality parameters for wetlands can vary depending on the catchment land use, specific wetland type, local ecological conditions, and the intended use of the wetland (e.g., conservation, recreation, agriculture).In the SS (sanctuary site), the presence of hexapods, ring pipes, and the subsequent growth of macrophytes created more favorable conditions for periphyton growth, leading to an increase in mean periphyton biomass.Conversely, the scattered gravel and rocks in the mud of the OS (open site) provided limited anchoring space for periphyton attachment, resulting in lower biomass.Several studies, including Osório et al (2019) and Tokeshi and Arakaki (2011), have highlighted the positive influence of habitat complexity on the richness of periphytic communities.Increased complexity, facilitated by physical elements, offers diverse niche opportunities for periphyton settlement, likely contributing to the higher biomass observed in the SS.Furthermore, the SS exhibited significantly greater macrophyte cover compared to the OS, where macrophytes were absent.This greater macrophyte presence in the SS further supported the abundant development of periphytic communities.Similar findings were reported by Taniwaki et al (2013), emphasizing that intricate aquatic macrophyte habitats promote increased diversity and richness of periphyton.
In terms of sediment composition, the SS displayed significantly higher levels of silt, clay, and organic matter, except for sand particles, which were more abundant in the OS.The presence of organic matter and sediment characteristics has been noted by previous studies such as Thorp and Covich (2001) and Jones et al (2012) as influential factors in the occurrence of benthic fauna.The sediment grain size is typically influenced by various factors, including sedimentation processes, re-suspension, inflow, currents, biological deposition, and physical features within river channels, as documented by Scheffer (1998).Additionally, the size of sediment grains plays a pivotal role in regulating the organic enrichment of the waterbody.A higher proportion of finegrain sizes, such as silt and clay, effectively captures and retains organic matter carried by inflowing water within the bottom sediments.This phenomenon has been supported by studies conducted by Silva et al (2006) and Spruzen et al (2008).Conversely, the reduced macrophyte cover in the OS led to decreased sediment diversity and a noticeable increase in the retention of sand particles.These findings align with the research conducted by Bücker et al (2010), Sychra et al (2010), and Luiza-Andrade et al (2017).

Impacts of AIC (SS) on macroinvertebrate community attributes
The current research revealed that there was a significantly higher level of diversity, richness, and abundance of macroinvertebrates in SS habitats in comparison to PP and OS habitat types.This phenomenon can be attributed to the influence of the AIC.The AIC in SS provided an increased complexity resulted in a variety of microhabitats within the wetland.The AIC plays a pivotal role in shaping these differences due to its ability to expand niche space, enhance food quality, and provide a protective environment against predators (Li and Asner 2023).Taniguchi & Tokeshi (2004) also reported increased diversity of benthos with habitat complexity in differential environmental settings.Previous studies showed that habitat complexity in wetland ecosystems creates a mosaic of niches, shelters, and resources that support a rich diversity of species (Taniguchi & Tokeshi 2004, Tews et al 2004, Li and Asner 2023).Different species have specific habitat preferences, and greater complexity means a broader range of niches.For example, submerged vegetation, fallen trees, and aquatic macrophytes create distinct microenvironments that support a diverse array of species.Macroinvertebrates, and amphibians use submerged vegetation, submerged woody debris, and underwater structures as hiding places from predators.This increased safety encourages the presence of a wider range of species.Different species have varying feeding strategies and preferences, and diverse habitats provide a variety of food sources, from algae and detritus to smaller prey organisms which leads to increased diversity.In addition, increased complexity means more suitable locations for reproductive activities.
These findings align with the concept of spatial heterogeneity as proposed by Huston (1979) and Tews et al (2004).In essence, they suggest that the complexity of a habitat directly influences the complexity of the faunal assemblage it supports.This notion is further substantiated by studies conducted by Kovalenko et al (2012) and Smith et al (2014), which emphasize the significant impact of habitat complexity on the variations observed in macroinvertebrate assemblages.
Additionally, research conducted by Hunter and Sayer (2009), Hackradt et al (2011), and Gatts et al (2014) on macroinvertebrate faunal assemblages in both artificial and natural reefs underscore the effect of reef structural complexity on the community of colonizing macroinvertebrates.This intricate habitat structure provides a diverse array of substrates for various life functions, including living, feeding, and reproduction, while also offering protection from predatory invertebrates and fishes, as noted by Wang et al (2006).Consequently, these factors collectively contribute to heightened diversity and species richness among macroinvertebrates.
The present findings are further corroborated by the studies of Zbikowski and Kobak (2007), Cai et al (2012), and Gleason et al (2018), which highlight a positive correlation between taxa abundance and richness and the presence of macrophyte cover originating from the AIC.Moreover, Hanson et al observed that heavily vegetated areas in wetlands exhibit greater species abundance among macroinvertebrates compared to less or nonvegetated sites, playing a pivotal role in shaping macroinvertebrate assemblages (Bilotta and Brazier 2008).Nevertheless, it is worth noting that the higher presence of silt and clay at SS sites could also contribute significantly to the increased diversity and species richness of macroinvertebrates in this habitat.
Similar conclusions have been drawn by Bilotta and Brazier (2008) and Anderson (2008), who emphasized the crucial role of sediment particle size in the distribution of macroinvertebrates within stream ecosystems.Therefore, the discovery of the highest macroinvertebrate abundance in SS habitats in the current study is consistent with these earlier findings.It is important to note that while sediment characteristics are influential, the response of species to habitat complexity holds greater significance.There exists an upper threshold beyond which further structural alterations can lead to a decline in species diversity and abundance, as highlighted by Kelaher (2003).
In designing the structural components utilized to create the Aquatic Invertebrate Community (AIC) in SS habitats, careful consideration was given to intervention types and complexity.An intermediate level of complexity was aimed to be maintained in SS habitats, with recognition that exceeding this level by introducing permanent structures could potentially impede water flow and increase sediment deposition.This, in turn, could lead to reduced water depth and decreased species diversity, as indicated by research conducted by Atilla et al (2005) and Kovalenko et al (2012).Thus, striking a balance between habitat complexity and the maintenance of essential ecological processes is deemed vital for preserving macroinvertebrate diversity and abundance.Our findings determined that T. tubifex, Gerris sp., L. acuminatea, and B. bengalensis were the most common taxa (frequency of occurrence > 80%, whereas the most dominant taxa were B. sowerbyi, L. hoffmeisteri, Chironomus sp. and L. acuminatea (relative abundance > 5%).The presence of abundant taxa also varied with habitat types.L. hoffmeisteri was the most abundant species in both the SS and the PP habitat types, while the abundance of Chironomus sp. was the highest in OS habitat type.The dominance of the Oligochaeta group was also reported from several ecosystems in Bangladesh (Mustafa et al 2013, Nupur et al 2013).Similarly, Jung et al (2008) stated that collector-gatherers (mainly Oligochaetes and Gastropods) were more abundant in temperate and subtropical rivers in East Asia.
These results offer a crucial basis for putting into practice focused tactics meant to maximize wetland ecosystems.Managers may create and carry out habitat enhancement programs that successfully support biodiversity and ecosystem resilience by knowing how artificial habitat structures affect habitat quality and macroinvertebrate diversity.By using this information, conservation programs may identify places that need to be prioritized for habitat development and restoration, helping to protect important wetland ecosystems.These results also highlight how crucial it is to include artificial habitat complexity in wetland management strategies in order to prevent habitat degradation and maintain the long-term viability of freshwater ecosystems.Overall, the knowledge gathered from this study is a useful instrument for directing conservation and management efforts meant to protect and restore the integrity of fresh water wetlands like Chalan Beel in Bangladesh.

Conclusions
The aim of this study was to investigate whether the artificially constructed habitat complexity could improve both habitat quality and the diversity of macroinvertebrates within Chalan Beels, the largest freshwater wetland in Bangladesh.The results demonstrated that the macroinvertebrate fauna made more efficient use of the artificially designed sanctuary site, which featured cemented hexapods and ring pipes as tools to elevate the quality of the aquatic habitat.The results revealed substantial improvements in the habitat, including increased periphyton biomass, greater macrophyte cover, and enhanced sediment quality parameters.These enhancements had a notably positive impact on the overall abundance and diversity of macroinvertebrate communities.Our findings strongly advocate for the augmentation of artificial habitat complexity to an intermediate level as a valuable strategy for effectively conserving macroinvertebrates in the Chalan Beel.By deliberately constructing more complex structures within the sanctuary site, we created an environment that better supports macroinvertebrates, thereby promoting their conservation and bolstering biodiversity in this largest wetland.

Figure 1 .
Figure 1.Study area map showing the study site at Singra Upazila of Natore district of Bangladesh.SS = Sanctuary site, PP = Partially protected site, OS = Open site.

D
(Shannon and Wiener (1949).H = the diversity index, ni = the relative abundance (S/N), S = the number of individuals for each species, and N = total number of individuals.Evenness index (J): e H L n S = [L n = The natural logarithm] followed fromPielou (1966).H = is the Shannon-wiener's diversity index and S = is the number of different species in the sample.Margalef's (1968) species richness (S): = Margarlef's richness index, S = Number of different species in the sample,N = Total number of individual species.

Figure 3 .
Figure 3. Generalized view of sanctuary site (A, upper view, B, lower view) with the arrangement of different artificial structures.

Figure 4 .
Figure 4. Species richness, density and diversity indexes of different habitat types.

Figure 5 .
Figure 5. Non-metric multidimensional (NMDS) plot showing differences in species assemblage among the studied intervention types.(SS = Sanctuary site, PP = Partially protected site, OS = Open site).

Table 1 .
Physicochemical water quality and other environmental variables recorded from the study sites.Values in the same raw having different superscript letters a indicates significantly higher b indicates intermediate situation and c indicates significantly lower values) indicated significant difference (p > 0.05) among the intervention types.

Table 2 .
Abundances of macroinvertebrate taxa according to their taxonomic affiliations along with their frequency of occurrence (FoO %) and relative abundance (RA %) in the intervention types in Chalan Beel (CB) during 2019-2020.
'+' indicates the taxa used in CCA analysis.

Table 3 .
Results of SIMPER analysis among the species assemblage of the studied intervention types.