Terrestrial gastropods are highly sensitive to environmental changes, which is further compounded by their limited ability to disperse or migrate quickly (Anderson and Coppolino 2007; Kralka 1986; Nunes and Santos 2012). Hence, the distribution and composition of these terrestrial gastropods are influenced by different environmental and geographic factors. These abiotic factors play a crucial role in determining their community structure within specific habitats (Coney et al. 1982; Wäreborn 1970). Several studies highlight the correlation between temporal and spatial changes in gastropod distribution and composition with temperature, rainfall patterns, elevation, and soil properties (Aubry et al. 2006; Bros et al. 2016; Liew et al. 2010; Tattersfield et al. 2001). These environmental factors impact the growth, reproduction and other physiological processes of gastropods.

Out of many gastropods, only a relatively small number are recognized as agricultural pests, i.e., those species that interfere with the production and utilization of crops and livestock, leading to economic losses (Godan 1983; Kozłowski 2012). Often these are exotic species that are inadvertently introduced into new environments. Altered environments such as agricultural lands provide opportunities for the passive dispersal of these exotic species (Wiktor 2001). Given such opportunities, these species can spread rapidly and even become invasive (Kozłowski and Kozłowski 2011; Wiktor 2001). Pest gastropods cause significant economic damage to crops in many countries (Baker 2002; Bishop 1977; Carlos Villalobos et al. 1995; Godan 1983; Kozłowski and Kozłowski 2011; Mc Donnell et al. 2009). In addition, these pests lower the quality of the crops by soiling them with slime and feces and can transmit pathogens or their eggs to plants, humans and livestock (Kozłowski 2012). Studies have shown that several species of terrestrial gastropods transmit various pathogenic nematodes, viruses, and fungal spores (Georgiev et al. 2003; Pokora 2001). Understanding the spatiotemporal population fluctuations and behaviors of these pests is crucial for effective control strategies.

However, similar to many other countries worldwide, there is a scarcity of research on the distribution and abundance of terrestrial pest gastropods, all of which are exotic species in Sri Lanka, in relation to environmental factors (Maheshini et al. 2019; Nunes and Santos 2012). Some of these pest gastropods cause considerable crop damage and economic losses to farmers in the island. Out of 21 exotic species, six slug species namely Laevicaulis alte, Deroceras reticulatum, Deroceras laeve, Mariella dussumieri, Arion intermedius, and Milax gagates and 12 snail species recorded by the Darwin Initiatives survey, are considered pests in Sri Lanka (MOE 2012; Mordan et al. 2003; Ranawana et al. 2003). Most of these exotic species are found in agricultural lands in the wet (mean annual rainfall of over 2,500 mm) and the intermediate (mean annual rainfall between 1,750 to 2,500 mm) zones of the country (Kumburegama and Ranawana 2001, 2002; Maheshini et al. 2019; Ranawana et al. 2003). Most of the European pest gastropods have primarily established themselves in the wetter parts of the wet zone, whereas the tropical invaders tend to thrive in relatively drier areas of the wet and intermediate zones of Sri Lanka (Bambaradeniya 2016; Marambe et al. 2011; Naggs and Raheem 2005). Previous baseline studies indicate that both European and tropical invaders thrive in the Nuwara Eliya district, located in the highlands of the Central Province within the wet zone (Kumburegama and Ranawana 2002; Ranawana et al. 2003).

The Nuwara Eliya district is the main upcountry vegetable, fruit and flower producer in Sri Lanka (Padmajani et al. 2014). Nearly 8.3% of the land in the district is dedicated to field crops, while nearly 7.6% of the total land area is used for home gardens (District Secretariat, Nuwara Eliya 2019). The topography and climatic conditions of this district seemingly provide an ideal environment for cultivating various crops. However, a handful of studies have highlighted the presence of exotic pest gastropods and the considerable damage they cause to crops in Nuwara Eliya (Kumburegama and Ranawana 2001, 2002; Maheshini et al. 2019). Despite some exploration of environmental factors such as air temperature, elevation and relative humidity, there has been limited analysis of the underlying reasons for the high abundance and wide distribution of exotic pest gastropods in agricultural lands within this district. Thus, the primary objectives of this study were to identify the composition, investigate spatiotemporal changes, and determine environmental preferences of exotic pest gastropods compared to the native species of terrestrial gastropods found in agricultural lands of Nuwara Eliya. Ultimately, this information could contribute to the identification of effective control measures and the reduction of economic losses caused by these pest gastropods.

Study site

The study was carried out for a period of two years from July 2017 to July 2019 in 80 agricultural lands in the Nuwara Eliya district located in the central highlands of Sri Lanka, between 80° 24’ 5” and 80° 57’ 8” east longitudes and 7° 16’ 5” and 6° 45’ 02” north latitudes (Fig. 1). The selection of agricultural fields for sampling of pest gastropods was based on 1:50,000 maps of the Survey Department of Sri Lanka. The chosen fields represented the entire district and included 65 agricultural fields, 5 vegetable poly-tunnels, 5 flower fields and 5 fruit fields. The selection criteria also ensured that molluscicides were not routinely used by the farmers. Irrespective of the extent, an agricultural field was defined as the land owned by a single farmer. Sampling was carried out during the night since terrestrial gastropods are highly active during this time. Each agricultural field was visited four times during the two-year study period, with two visits during the rainy and two visits during the non-rainy periods. This approach aimed to increase the sampling effort and investigate the temporal changes in pest gastropods.

Figure 1. Study site of terrestrial pest gastropods in the Nuwara Eliya district, Sri Lanka. National highways in the island are graded A (red) or B (yellow) and the other smaller roads are indicated as Grade C (black).

The identification of rainy and non-rainy periods was done with the aid of a climatic diagram prepared using the packages “climate” and “iki.dataclim” in R (Guijarro 2018; Orlowsky and Seneviratne 2014). Nuwara Eliya experiences two non-rainy periods, from February to mid-March and mid-May to July, and two rainy periods, from mid-March to mid-May and the most prominent one from August to January (Fig. S1).

Survey design

Ten 1 × 1 m² plots were randomly established per crop in each agricultural land and sampled for pest gastropods for a maximum period of 15 minutes per plot. A total of 3,690 plots were sampled. In addition to the plot sampling, the entire field was scanned including the margins of the field, on plants and in soil to record any species that may have been missed during plot sampling. However, only the data from the plot sampling was used for the calculation of the density and relative abundance of the terrestrial pest gastropods. The abundance, microhabitats and different stages of the life cycle encountered during plot sampling were recorded for each species. The identification of all the species was done using available guides and published literature. The nomenclature followed the MolluscaBase (2023) website (https://www.molluscabase.org/) as well as Naggs and Raheem (2000) and Raheem et al. (2000).

Longitude, latitude, air temperature, elevation, soil pH, relative humidity and daily rainfall were recorded for each sampling site. The average values for air temperature, soil pH, relative humidity, and rainfall were used for data analysis. The geographical coordinates and elevation were estimated using a portable global positioning system receiver (Magellan eXplorist 310; MiTAC International Corp., Santa Clara, CA, USA). Air temperature and relative humidity were measured using a portable weather tracker (Kestrel 4000NV; Nielsen-Kellerman Co., Boothwyn, PA, USA). Rainfall data were collected from the Department of Meteorology, Colombo, Sri Lanka. Soil pH was measured using a soil pH meter (HANNA-HI-99121; Hanna Instruments Inc., Woonsocket, RI, USA). Furthermore, for each of the 80 fields visited, the crop type, crop stage (nursery, young and mature), agricultural land type (open agricultural land or polytunnel), and the extent of the agricultural land as an indicator of the degree of fragmentation were recorded. The extent of the field was categorized into three main categories, low (< 1 ha), medium (1–5 ha), and high (> 5 ha) degrees of fragmentation for further analysis.

Data analysis

Live individuals and shells of dead gastropods were both included in the analysis, following standard practice in terrestrial gastropod surveys. Sample-based rarefaction values and both Chao 2 and Jackkinfe 2 estimators were calculated for rainy and non-rainy periods using EstimateS version 9.1.0 (Chao et al. 2009; Colwell 2006). Rarefaction, abundance and density were plotted using SigmaPlot 10 Version 10.0.0.54 (Grafiti LLC., San Jose, CA, USA). All other statistical analyses were conducted using R version 1.2.1335 (Posit Software, PBC, Boston, MA, USA). The statistical significance for species densities and abundances was performed through one-way analysis of variance (ANOVA) at 95% confidence limits.

A simple regression analysis was carried out to establish the relationship between measured environmental variables and the abundance and richness of gastropod species. Graphs illustrating the preference of each species for various environmental factors were plotted, examining the impact of measured variables. These visualizations aimed to pinpoint the optimal range influencing the distribution of each encountered species. The relationships between species composition and environmental variables were determined using canonical correspondence analysis (CCA) using Canoco version 5.0 (Wageningen University & Research, Wageningen, Netherland) and generalized linear mixed model (GLMM) using the vegan package with a 95% significant level. In the GLMM, fixed factors included air temperature, elevation, soil pH, relative humidity, daily rainfall, agricultural land type, degree of fragmentation (extent of the field), crop type, and crop damage stage and abundance. Sampling locations were treated as random factors. The relative importance of the different environmental variables in explaining variation in species composition was assessed through forward selection (Raheem et al. 2008, 2009). Human disturbance was not considered as a factor since agricultural lands are inherently influenced by human activities and are surrounded by human settlements. The sampling locations were considered as different treatment levels and the GLMM was used to evaluate differences in species abundance in these sampling locations. The best fit model was determined using the Akaike Information Criterion.

where, y = Species abundance or richness, β_o = Intercept, β_n = Coefficient/slope for different random factors, x = Random factors (sampling location), γ = Coefficient/slope for different fixed factors, z = Fixed factors (air temperature, elevation, soil pH, relative humidity, daily rainfall, agricultural land type, and degree of fragmentation), and ε = Error factor.

Evaluating the spatial distributions patterns of gastropods and their ecological preferences involved examining species distribution maps for endemic, non-endemic native, and exotic pest snail and slug species. These maps, created using inverse distance weight interpolation, utilized abundance values. The classification of species abundance was achieved through the natural breaks (Jenks) criteria in ArcMap version 10.4 (ESRI Inc., Redlands, CA, USA).

Temporal fluctuations in gastropod populations were scrutinized employing an unpaired t-test, evaluating statistical significance in densities during rainy and non-rainy periods. Box plots visually depicted seasonal environmental variations. Biodiversity indices, such as Shannon Wiener, Simpson dominance, Shannon evenness, and Jaccard, were computed for both seasons using the BiodiversityR package. These metrics facilitated a comparison of species composition between rainy and non-rainy periods. The analyses aimed to unveil nuances in gastropod dynamics across varying climatic conditions, offering insights into the ecological responses of these species to seasonal shifts in the study area.

Species composition in agricultural lands

A total of 7,083 individuals belonging to 7 families, 12 genera and 14 species were recorded over the two-year period. Among the recorded species, five were native and nine were exotic species (Table 1). The exotic pest gastropods included five slug species (D. laeve, D. reticulatum, M. dussumieri, L. alte, and M. gagates) (Fig. 2A-F) as well as four snail species (Allopeas gracile, Bradybaena similaris, Lissachatina fulica, and Subulina octona) (Table 1, Fig. 2G-K). Most of the native species (Fig. 2L-P) were found along the margins of the agricultural lands. However, these native species were not identified as agricultural pests, as no evidence of their direct damage to crops was observed. The rarefaction curve represents the accumulative number of species recorded at each sampling site. In the present study it suggests that the total number of species may exceed the 14 species recorded (Fig. 3). The two non-parametric estimators, Chao 2 and Jackknife 2, estimated a species richness of 15 and 16 respectively, for the agricultural lands.

Figure 2. Terrestrial gastropods and some of their egg clutches from agricultural lands in the Nuwara Eliya district. (A-F) Exotic pest slugs: (A) Deroceras laeve, (B) Deroceras reticulatum, (C) Mariella dussumieri, (D) Milax gagates, (E) and (F) different color morphs of Laevicaulis alte. (G-K) Exotic pest snails: (G) Allopeas gracile, (H) Subulina octona, (I) Lissachatina fulica, (J) and (K) different color morphs of Bradybaena similaris. (L-P) Native snails: (L) Ariophanta bistrialis, (M) Cryptozona chenui, (N) Macrochlamys indica, (O) Ratnadvipia irradians, and (P) Euplecta emiliana. (Q-S) Egg clutches found from the marginal vegetation: Eggs of (Q) Lissachatina fulica, (R) Mariella dussumieri, and (S) Euplecta emiliana.

Figure 3. Species based rarefaction curves with lower and upper 95% confidence boundaries constructed for terrestrial gastropods encountered from the agricultural lands in the Nuwara Eliya district.

Table 1

Terrestrial gastropod species recorded from agricultural lands in Nuwara Eliya.

Family	Species	NCS	Invasive status	Occurrence in the agriculture field			Presence of gastropods				Density (ha-1)
				Field	Margin		Vegetable	Flower	Fruit		Rainy	Non-rainy	t value	p-value
Slugs
Agriolimacidae	Deroceras laevea	NE	Globally invasive	371	30		344	0	57		289	112	2.64	9.309e-14
Agriolimacidae	Deroceras reticulatuma	NE	Globally invasive	3,354	30		3,072	0	322		2,790	604	1.89	5.984e-06
Ariophantidae	Mariaella dussumieria	NE	Non-invasive	382	5		379	8	0		329	58	4.20	< 2.2e-16
Milacidae	Milax gagatesa	NE	Non-invasive	7	2		9	0	0		8	1	1.85	0.07
Veronicellidae	Laevicaulis altea	LC	Locally invasive	149	8		157	0	0		144	13	1.69	0.0001
Snails
Achatinidae	Lissachatina fulicaa	NE	Globally invasive	473	10		469	14	0		407	76	2.91	0.002
Ariophantidae	Ariophanta bistrialisc	LC	Non-invasive	0	11		11	0	0		11	0	-	-
Ariophantidae	Cryptozona chenuib	VU	Non-invasive	0	6		6	0	0		0	6	-	-
Ariophantidae	Euplecta emilianab	EN	Non-invasive	11	12		23	0	0		22	1	1.30	0.19
Ariophantidae	Macrochlamys indicac	DD	Non-invasive	0	362		333	29	0		299	63	1.95	0.05
Ariophantidae	Ratnadvipia irradiansb	VU	Non-invasive	0	4		2	2	0		4	0	-	-
Bradybaenidae	Bradybaena similarisa	NE	Globally invasive	1,609	65		1,497	19	158		1,283	391	3.80	< 2.2e-16
Subulinidae	Subulina octonaa	NE	Non-invasive	57	10		67	0	0		67	0	2.98	< 2.878e-07
Subulinidae	Allopeas gracilea	NE	Non-invasive	113	15		118	0	0		118	0	2.32	0.02
Species richness				10	14		13	5	3		13	10	6.98	0.001
Family richness				7	7		7	4	2		7	5	1.20	0.001
Species abundance				6,526	570		6,487	72	537		5,771	1,325	4.80	4.693e-07
Number of sampled plots				7,220	400		7,380	120	120		4,130	3,490	-	-

NCS: National Conservation Status; NE: not evaluated; LC: least concern; VU: vulnerable; EN: endangered; DD: data deficient.

^aExotic pest species, ^bendemic species, ^cnon–endemic native species.

Species abundance reflects the composition and dominance of species within a particular ecosystem. The abundance of each encountered species in the agricultural fields was different (Fig. 4). The abundance of slugs was higher compared to the abundance of the sampled snails in the study site, although the difference was not statistically significant (t = 1.35, p = 0.41). Specifically, D. reticulatum and B. similaris accounted for 50% and 25% of the total abundance, respectively. The remaining 12 species were relatively less abundant (< 10%). Among the native species, except for Macrochlamys indica (8%), the others were relatively less abundant (< 1%) and primarily found along the margins of the agricultural fields (Table 1, Fig. 4).

Figure 4. Abundance and density of terrestrial gastropods in agricultural lands in the Nuwara Eliya district from July 2017 to July 2019. Simple alphabetical letters a, b, and c denote the significant level for each of the abundance values.

Effects of environmental variables on species composition

In addition to dispersal ability and biotic interactions, environmental factors play a crucial role in species distribution. Regression analysis unveiled the preferred environmental conditions for recorded gastropods. The results indicated that species abundances and richness exhibited an increasing trend with elevation, pH, and rainfall, while decreasing with atmospheric temperature (Figs. S2 and S3). Nonetheless, there were species-specific preferences for the environmental variables that were measured. The exotic pest gastropods exhibited broader preferences for elevation, daily rainfall, air temperature, soil pH and relative humidity compared to the native species. Specifically, the European invaders, D. laeve and D. reticulatum were found to prefer similar environmental ranges, thriving at elevations of 600–2,000 m a.s.l., temperature of 16°C–26°C, soil pH of 4.0–8.0, rainfall of 4–16 mm per day and relative humidity of 80%–90%. On the other hand, the tropical invaders such as L. fulica, B. similaris, L. alte, and M. dussumieri preferred elevations of 300–1,400 m a.s.l., temperatures of 20°C–26°C temperatures, soil pH of 4.0–8.0, rainfall of 4–16 mm per day and relative humidity of 70%–90% (Figs. S4 and S5). In contrast, most of the native species exhibited narrower ranges of tolerance to the measured environmental variables. For example, species such as M. indica and Euplecta emiliana were recorded at elevations between 1,300–1,350 m a.s.l., temperatures of 23°C–24°C, relative humidity of 90%–91%, rainfall of 11–12 mm per day and soil pH range of 7.5–8.0.

CCA was employed to assess the combined effects of environmental factors on the spatial arrangement of the sampled gastropods in agricultural fields in the Nuwara Eliya district (Fig. 5). The eigenvalues of the first two CCA axes were 0.54 and 0.23 respectively explaining a total of 73.23% of the variation. Elevation (F = 13.2, p = 0.002), relative humidity (F = 0.02, p = 0.02), soil pH (F = 0.03, p = 0.03), and daily rainfall (F = 2.0, p = 0.05) significantly contributed to the variations in pest gastropod composition and distribution in Nuwara Eliya. Species were primarily arranged along the first axis based on elevation. Along the second axis, species were primarily arranged by daily rainfall and secondarily by soil pH (Fig. 5, Table S1). While elevation and air temperature were significant factors that affected the distribution of most of the exotic slug species such as D. laeve, D. reticulatum, M. gagates, and M. dussumieri, the distribution of many of the native and exotic snail species strongly correlated with relative humidity and daily rainfall (Fig. 5). Furthermore, soil pH played a role in the spatial arrangement of Cryptozona chenui, M. indica, S. octona, and E. emiliana. Relative humidity was the governing factor in the distributions of A. gracile, L. alte, L. fulica, Ariophanta bistrialis, and Ratnadvipia irradians (Fig. 5).

Figure 5. CCA biplot of species with environmental parameters and field/crop type in agricultural habitats in Nuwara Eliya district. The light blue font represents exotic pest species, the green font represents non-endemic native species and the purple font represents endemic species. CCA: canonical correspondence analysis; AllpGrac: Allopeas gracile; BradSiml: Bradybaena similaris; ArioBist: Ariophanta bistrialis; CrypChen: Cryptozona chenui; DercLeva: Deroceras laeve; DercRetc: Deroceras reticulatum; EuplEmil: Euplecta emiliana; LaevAlta: Laevicaulis alte; LissFulc: Lissachatina fulica; MarcIndc: Macrochlamys indica; MariDuss: Mariella dussumieri; MilxGagt: Milax gagates; SublOctn: Subulina octona; RatnIrrd: Ratnadvipia irradians.

The distribution of pest gastropods in this district was also influenced by crop type, agricultural land type, field extent, and crop stage (Fig. 5). Most pests showed a strong association with crops such as leeks, cabbage, carrot, broccoli, radish, and knol khol (German turnip) particularly during the nursery and young stages (Fig. 5). The density of most pest gastropods was higher in medium and high-extent agricultural lands compared to smaller fields and poly-tunnels, which had smaller extents (Fig. 5).

To examine the combined effect of environmental variables, crop type, field type, crop stage, degree of fragmentation, and the sampling locations in the Nuwara Eliya district on species richness and abundance, multi-linear regression models were applied. The analysis revealed a significant collective effect of the sampling sites on species richness and abundance, along with some measured environmental variables, indicating the impact of treatment units (sampling sites). The species abundance was influenced by sampling locations in conjunction with elevation, crop type, field type, and crop stage. Therefore, species richness varied across locations due to the combined effects of elevation, crop type, field type, and crop stage (Tables 2, 3).

Table 2

Comparison of the generalized linear mixed models affecting the composition and distribution of terrestrial pest gastropods.

Model	AIC value	Conditional R²	p-value
Richness–elevation + rainfall + relative humidity + temperature + soil pH + crop + field type + stage + [site]	678.30	0.23	< 0.05
Richness–elevation + crop + field type + stage + [site]a	659.30	0.04
Richness–elevation + crop + field type + stage	675.82	0.20
Abundance–elevation + rainfall + relative humidity + temperature + soil pH + crop + field type + stage + [site]	5,893.21	0.98	< 0.05
Abundance–elevation + crop + field type + stage + [site]a	650.12	0.97
Abundance–elevation + crop + field type + stage	6,409.40	0.96

AIC: Akaike Information Criterion.

^aBest fitted model was selected using maximum likelihood method (AIC values).

Table 3

Best fit generalized linear mixed models models for species richness and abundance in the montane agricultural lands.

Factor	Richness				Abundance
	Estimated coefficient	Standard error	p-value		Estimated coefficient	Standard error	p-value
Intercept	0.0130	0.3354	0.969		–1.5872	0.3997	7.15e-05
Elevation	0.0007	0.0002	0.004		0.0009	0.0003	1.92e-04
Crop-cabbage	1.9525	0.1817	0.001		1.0665	0.2526	2.42e-05
Crop-leeks	1.9497	0.1362	0.001		1.2943	0.1609	8.96e-16
Crop-lettuce	1.5605	0.1383	0.001		0.6860	0.1662	3.64e-05
Crop-radish	0.6613	0.1473	7.16e-06		–0.8325	0.1826	5.12e-06
Crop-strawberry	2.1340	0.1625	0.001		1.4050	0.2314	1.27e-09
Crop-knolkhol	1.2008	0.1981	1.34e-09		0.6268	0.2736	0.02
Crop-Chinese cabbage	0.9509	0.1805	1.38e-07		0.7479	0.2190	6.38e-04
Stage-young	0.3545	0.0994	4e-04		0.6916	0.1755	8.12e-05
Field-polytunnel	–0.7307	0.0709	0.001		0.2286	0.1181	0.05

Spatial distribution of the gastropods

The distribution patterns of the exotic, native and endemic species exhibited clear spatial variations across the agricultural lands in the Nuwara Eliya district. The majority of the exotic species showed wide distributions, particularly concentrated in the southeast part of the district (Diagama, Agarapathana, Labukele, Kudaoya, Bogawanthalawa and Ragala) (Fig. 6A, B). The native non-endemic species were mainly observed in the southern part (Bogawanthalawa, Hatton, Nallathanniya and Norwood) (Fig. 6C), while the endemics showed narrower distributions primarily in the northeast and northwest parts (Keenagolla, Walapane, Udagama and Palagolla) of the district (Fig. 6D).

Figure 6. Distribution of terrestrial gastropods in agricultural lands in the Nuwara Eliya district. Distribution of (A) exotic slugs, (B) exotic snails, (C) non-endemic native snails, and (D) endemic snail species.

Furthermore, the spatial arrangement of individual species in agricultural fields of the district also exhibited distinct variations with respect to environmental factors. Dry and wet areas of the district could be recognized according to rainfall distribution (Fig. S6). Most of the tropical pest snails such as A. gracile, S. octona, L. fulica, and L. alte, were dominant in the drier parts of the district, whereas B. similaris (Fig. S7), and the European slug species such as D. laeve and D. reticulatum were widely distributed throughout the wetter/cooler part of the district (Fig. S8). On the other hand, native species like A. bistrialis, C. chenui, E. emiliana, M. indica, and R. irradians were predominantly found in the drier areas of the district (Fig. S9).

Temporal population dynamics of the gastropods

The two main seasons in the Nuwara Eliya district were characterized based on rainfall and temperature patterns. The average daily rainfall during the rainy and non-rainy periods was 7.0 ± 3.1 mm and 4.0 ± 1.2 mm, respectively. Additionally, air temperature, soil pH and relative humidity exhibited temporal variations corresponding to the rainfall pattern in the district. The average night temperatures during the non-rainy and rainy periods were 25.0°C ± 2.1°C and 22.0°C ± 1.7°C, respectively. Soil pH displayed a wider variation during the rainy period (5.3–7.0) compared to the non-rainy period (5.4–6.4). The mean relative humidity between the non-rainy and rainy periods was 82% and 90%, respectively (Fig. S10).

A marked difference between the population structure of pest gastropods was observed during the rainy and non-rainy periods. A total of 5,771 individuals during the rainy period and 1,325 individuals during the non-rainy period were encountered in the agricultural lands (t = 5.04, p = 4.693e^-07). Altogether 13 species belonging to 7 families were identified during the rainy period, while 10 species belonging to 5 families were identified during the non-rainy period (t = 6.98, p = 0.001). However, the two non-parametric estimators of Chao 2 and Jackknife 2 predict 14 species during the rainy period and 11 species during the non-rainy period. Almost all the species except M. gagates, S. octona, A. gracile, A. bistrialis, and R. irradians were found in both seasons while C. chenui was only recorded in the non-rainy period (Table 1). Surprisingly, we discovered that some terrestrial gastropods rest and lay eggs in marginal vegetation and dumping or composting sites adjacent to the agricultural lands during the non-rainy period (Fig. 2Q-S).

Density is another parameter that can be used to predict population trends and understand how species arrange themselves in space according to limited resources. The density of B. similaris (t = 2.69, p < 0.05), D. reticulatum (t = 2.46, p < 0.05), D. laeve (t = 2.32, p < 0.05), L. fulica (t = 2.30, p < 0.05), L. alte (t = 1.94, p < 0.05), M. dussumieri (t = 2.64, p < 0.05), M. gagates (t = 2.04, p < 0.05), and M. indica (t = 1.93, p < 0.05) were higher during the rainy period (Table 1). In addition, the diversity of the terrestrial pest gastropods was greater in the rainy period (H’ = 1.60 and D_s = 0.70) compared to the non-rainy period (H’ = 1.45 and D_s = 0.69). Most of these gastropods exhibited an even distribution in the non-rainy period (S_e = 0.63) (Table 4). The similarity of the sampled gastropods between the two seasons was further examined using the Jaccard index, revealing a 57% level of similarity between the rainy and non-rainy periods.

Table 4

Terrestrial pest gastropods diversity between the rainy and non-rainy periods.

Diversity index	Rainy	Non-rainy
Shannon wiener index (H’)	1.60	1.45
Simpson dominance index (Ds)	0.70	0.69
Shannon evenness index (Se)	0.62	0.63
Jaccard index (Cj)	0.57

Information on the composition, distribution and impact of environmental factors on pest gastropods in in Sri Lanka is meagre. The present study on exotic, terrestrial pest gastropods represent the first comprehensive survey undertaken in the central highlands of Sri Lanka to document species richness, distribution, abundance and their correlation with environmental variables in agricultural lands. The accidental introduction of species worldwide has resulted in significant alterations to native ecological communities and has contributed to the decline or extinction of numerous native species. Exotic pest gastropods, in particular, have been unintentionally introduced to numerous countries, posing a significant threat to agricultural lands by causing extensive damage to crops including cereals, fruits, vegetables, and ornamental plants (Baker 2002; Borkakati et al. 2009; Cowie et al. 2009; Godan 1983; Kozłowski and Kozłowski 2011; Maheshini et al. 2019). Hence, for the effective control these pests and prevent their further spread, it is crucial to gather information on their distribution, composition, and environmental preferences, underscoring the importance of the present study.

Previous investigations in the Nuwara Eliya district reported only three gastropod families namely Agriolimacidae, Bradybaenidae, and Subulinidae (Kumburegama 2000). However, the present study, in alignment with Maheshini et al. (2019), identified a total of seven gastropod families, namely Achatinidae, Agriolimacidae, Ariophantidae, Bradybaenidae, Milacidae, Subulinidae and Veronicellidae, in Nuwara Eliya, indicating that these terrestrial gastropods have established themselves in agricultural lands relatively quickly. Future sampling efforts should explore the diversity, distribution, and composition of gastropods in cultivated fields, enhancing our understanding of their ecological dynamics.

Effects of environmental variables on species composition

Only a few studies have investigated the factors affecting terrestrial gastropod composition and distribution in modified ecosystems such as agricultural lands, compared to natural or forest landscapes (Maheshini et al. 2019). According to the present study, almost all the species recorded show unique preference ranges for the measured environmental factors. Interestingly, however, most of the exotic pest gastropods have wider range of preferences for the measured environmental variables compared to the natives. This enables the exotic pest gastropods to survive in relatively larger numbers even during the non-rainy season, posing a huge threat to farmers. Temperature and elevation appear to be key factors that affect the distribution of most of the slug species recorded, while relative humidity, rainfall, and soil pH affect the distribution of many of the native and exotic snails. Accordingly, slugs are more common at higher elevations, while a large proportion of tropical exotic pest snails occur in agricultural lands at low elevations with high relative humidity and relatively high temperatures. Undoubtedly, most terrestrial gastropods cannot withstand high temperatures and low humidity since they depend on water or high humidity for an active life and are relatively susceptible to desiccation, which tends to restrict them (Baker 1958; Getz 1974). Furthermore, moist conditions are necessary for their respiration, reproduction and mucus production (Cameron 1970; Coney et al. 1982). Therefore, temperature, rainfall, and humidity significantly affect terrestrial gastropod populations (Sionek and Kozłowski 1999; Wiktor 2001).

In addition to the above environmental factors, luminosity, canopy cover, litter humidity, soil moisture, soil pH, vegetation characteristics, type, and availability of food significantly affect the terrestrial gastropod composition and distribution (Astor et al. 2017; Nunes and Santos 2012). According to some studies, soil characteristics are the principal determinants of gastropod distributions (Evans 1968; Mörzer Bruijns et al. 1959), whereas others suggest that litter characteristics, soil characteristics, and vegetation type are the main determinants (Bishop 1977; Ondina et al. 2004). In Sri Lankan lowland rainforests, the variation in terrestrial gastropod assemblage, primarily composed of non-endemic and endemic native species, is mainly determined by elevation, canopy cover, and soil pH (Raheem et al. 2008). The absence or the limited canopy cover and leaf litter in human-maintained agricultural lands may therefore significantly impact the terrestrial gastropod composition.

Food availability also plays a crucial role in the distribution of pest gastropods. Depending on the crop, the pest gastropods may damage the seedlings, leaves, roots, shoots, flowers, or fruits. It results in a loss or decrease in harvest due to direct damage, loss of market value due to vegetables soiled by these pests, or due to secondary infestation of bacteria and fungi (Borkakati et al. 2009). Nursery and young stages of the crops are especially vulnerable to pest gastropod damage. While the recorded pest species were voracious feeders on agricultural crops, the few native species did not feed on these crops. Investigating the food preferences of native species in comparison to pest species would be valuable in understanding why the native species do not pose a significant threat to agricultural crops.

Spatial distribution of pest gastropods

Species distribution is influenced by various environmental factors, and global trade and commerce have facilitated the introduction of exotic species worldwide. Once introduced, terrestrial pest gastropods can spread and thrive, especially in the absence of natural predators. Many of the pest gastropods recorded in this study are widely reported in other regions and are classified as invasive. Notable, the European gastropods species D. reticulatum and D. laeve have become invasive in various areas including Northern Africa, North America, Australia, New Zealand, South Asia, and Southeast Asia (Godan 1983; Kozłowski 2012; Kozłowski and Kozłowski 2011; Tulli et al. 2009). Of particular concern is L. fulica, which is a widely distributed invasive alien pest species that is mainly distributed in association with human dwellings and cultivated areas (Fontanilla et al. 2014; Foon et al. 2017; Ratnapala 1984).

While all the identified pest gastropods species are non-native to Sri Lanka, the country itself boasts a rich diversity of terrestrial gastropods, with over 80% of the recorded 253 species being endemic (Raheem et al. 2006). Many of the Sri Lankan gastropods require forests and most of the native gastropod species are found concentrated in either lowland or submontane/montane rainforests, especially in the wet zone of the country (Raheem et al. 2008; Ranawana 2016). However, the conversion of forested areas into agricultural lands has led to a significant shift in the composition of terrestrial gastropods, with exotic species now dominating the modified environments. Furthermore, the proximity of some agricultural lands to natural forests in this region may have adverse effects on the native species inhabiting these habitats which needs to be investigated.

Temporal population dynamics of gastropods

Population fluctuations are highly influenced by environmental and climatic factors, which determine the rates of reproduction, growth and survival. However, the effects of seasonal fluctuations on terrestrial gastropod pest communities have been poorly studied (Kozłowski 2012; Sionek and Kozłowski 1999). Studies conducted in Egypt, England, Australia and Scotland report that the pest population sizes of certain species including D. reticulatum and D. laeve are highest during the spring (Baker 2008; Eshra 2013; Griffiths et al. 1998; Joe and Daehler 2008; Sionek and Kozłowski 1999; Shahawy 2019; Willis et al. 2006), and they reproduced during the autumn. Furthermore, several of these studies indicate that pest species are least abundant during winter (Baker 2008; Willis et al. 2006).

Unlike temperate countries that experience four seasons annually, Sri Lanka, being a tropical country, only experiences rainy and non-rainy periods based on the rainfall pattern. Most agricultural practices and gastropod populations in Sri Lanka rely on rainfall pattern. As expected, the abundance of most endemic and exotic gastropod species is significantly higher during the rainy period. This highlighted the fact that rain is one major factor that causes seasonal fluctuations of the gastropods, as adequate rainfall provides suitable micro habitats (Carlos Villalobos et al. 1995), helps prevents desiccation and dehydration of snails and slugs, and evenly disperses and increases the availability of limiting resources. Interestingly, however, some non-endemic native species were more abundant during the non-rainy period. This implies that these non-endemic native species are more tolerant of drier conditions compared to exotics and endemics. Moreover, egg clutches and estivating gastropods were found in soil and under decaying organic matter along field margins during the non-rainy period, suggesting that these species breed during this period while the young hatch during the rainy period, leading to a sudden increase in the population. In contrast to the exotic pest gastropods recorded in this study, species like Succinea costaricana, which damage ornamental plants are highly abundant during periods of low rainfall in Costa Rica and least abundant during periods of high rainfall (Carlos Villalobos et al. 1995).

Most of the limiting resources, such as food, microclimate, and microhabitats, could become scarce and clumped in some areas during the non-rainy period (Ruthven 1920). Consequently, the terrestrial gastropod pests become more evenly dispersed during the rainy period, while in the non-rainy period, they cluster around limiting resources, exhibiting random and clumped species distributions. Furthermore, inter- and intra-specific competition can be less during the rainy period due to the availability of resources (Hairston 1987).

To effectively implement integrative pest management (IPM) tools in the management of terrestrial pest gastropod populations, comprehensive knowledge of their biology and ecology, including reproduction, and how the abundance and densities change is essential. Mechanical, chemical and biological control methods can then be utilized more effectively in managing pest gastropod populations. According to the present study, since there is a significant difference in the pest population during the non-rainy period, IPM strategies combining manual and chemical methods may be more effective during this time. However, continuous application of IPM methods in both periods is necessary to keep the pest gastropod population at low levels. Furthermore, the removal of egg clutches during the non-rainy period through tillage, clearing of field margins, and composting sites could greatly influence the reduction of the abundance of terrestrial pest gastropods during the rainy period.

The present study provides a comprehensive understanding of the species richness, composition, abundance, and distribution patterns of gastropods in agricultural lands of the Nuwara Eliya district. The gastropod composition included both native and exotic species, with exotic pest gastropods exhibiting a higher abundance. Notably, while the exotic species, especially D. reticulatum and B. similaris, accounted for a significant proportion of the total abundance, the native species were predominantly found along the margins of agricultural lands and did not exhibit direct damage to crops. Environmental factors played a crucial role in shaping species composition and distribution. Overall, exotic pest gastropods exhibited broader environmental preferences compared to native species, highlighting the adaptability of invaders to a range of conditions. The results also illustrated the combined effects of environmental factors on the spatial arrangement of gastropods in agricultural fields, with elevation, relative humidity, soil pH, and daily rainfall emerging as significant contributors. Spatial distribution patterns indicated clear variations among exotic, native, and endemic species across different regions of the district. Crop type, field type, field extent, and crop stage also influenced the distribution of pest gastropods. Pest gastropod abundance and diversity is significantly higher during the rainy season. However, some species exhibited seasonal preferences, and terrestrial gastropods were found to rest and lay eggs in specific sites during the non-rainy period. The present study contributes valuable insights into the complex interactions between gastropods, environmental factors, and agricultural practices in the Nuwara Eliya district. Understanding these dynamics is crucial for implementing effective pest management strategies.

Supplementary Information accompanies this paper at https://doi.org/10.5141/jee.23.042.

Table S1. Forward selection results: interactive-forward-selection for canonical correspondence analysis. Fig. S1. Climatic diagram of the Nuwara Eliya district. Fig. S2. Regression analysis highlights the relationships between specific environmental variables and gastropod abundance. Fig. S3. Regression analysis between selected environmental variables and species richness. Fig. S4. Environmental factor preferences of the pest snail species in Nuwara Eliya district. Fig. S5. Environmental factor preferences of the pest slug species in Nuwara Eliya district. Fig. S6. Environmental factor distribution in Nuwara Eliya district. Fig. S7. Distribution of pest snails from agricultural lands in Nuwara Eliya district. Fig. S8. Distribution of pest slugs from agricultural lands in Nuwara Eliya district. Fig. S9. Distribution of the native gastropods from agricultural lands in Nuwara Eliya district. Fig. S10. Selected environmental parameter ranges between two seasons in agricultural fields in the Nuwara Eliya District for a period of two years from July 2017 to July 2019.

CCA: Canonical correspondence analysis

ANOVA: Analysis of variance

GLMM: Generalized linear mixed model

IPM: Integrative pest management

KGDDT conducted fieldwork, data analysis and wrote manuscript. GNH conducted fieldwork. KBR conducted research design and planning. SK conducted research design and planning, research supervision, data analysis and worte manuscript.

This study was funded by the National Science Foundation (NSF), Sri Lanka (Grant No: RG/2017/EB/05).

All data generated or analyzed during this study are included in this published article (and its supplementary information files). Voucher specimens are available at the Department of Zoology, Faculty of Science, University of Peradeniya, Sri Lanka.

Ethical clearance is not required for studies on pest gastropods. Informed consent from the farmers or owners of each agricultural field was obtained prior to the start of data collection.

Not applicable.

The authors declare that they have no competing interests.