Introduction
The false chinch bug, Nysius cymoides (Spinola) (Hemiptera: Lygaeidae), is a significant agricultural pest that feeds on developing seeds and causes serious damage to canola (Brassica napus L.) and other oilseed crops across the Mediterranean Basin, temperate regions of Europe, and the Turanian zone of Central Asia (Scaccini and Furlan, Reference Scaccini and Furlan2021). The damage inflicted by this insect can lead to substantial economic losses for farmers and reduce the quality and quantity of oilseed crops, negatively impacting the production of vegetable oil and biodiesel (Mohaghegh et al., Reference Mohaghegh, Amir-Maafi, Shahrokhi and Pirhadi2016; Scaccini and Furlan, Reference Scaccini and Furlan2021). Among the host plants of the false chinch bug, canola is widely recognised, with the insect responsible for outbreaks and severe damage in Europe and Asia (Amini Khalaf Badam et al., Reference Amini Khalaf Badam, Mohaghegh and Ostovan2006; Mohaghegh et al., Reference Mohaghegh, Amir-Maafi, Shahrokhi and Pirhadi2016; Chapelin-Viscardi et al., Reference Chapelin-Viscardi, Tourton and Matocq2017; Scaccini and Furlan, Reference Scaccini and Furlan2021; González et al., Reference González, Aldanondo, Barredo and Alarcón-Elbal2022). In recent years, N. cymoides has emerged as a significant threat to canola production in Kermanshah province, Iran (Zamani et al., Reference Zamani, Pourian, Miri, Alizadeh, Moeini-Naghadeh, Jalilian, Dehnavi, Sharifi and Hemmati2022), with studies in the neighbouring Kurdistan province revealing varying levels of susceptibility among canola cultivars (Bidar et al., Reference Bidar, Sadeghi and Maroufpoor2024).
Post-harvest dispersal is a key determinant of pest population dynamics in the following growing season, yet it remains understudied for many cropping systems (Kennedy and Storer, Reference Kennedy and Storer2000). Following crop removal, phytophagous insects must either emigrate to alternative hosts or shelter within residual plant debris, with the outcome depending on species-specific habitat requirements (Mazzi and Dorn, Reference Mazzi and Dorn2012). In hemipteran pests, post-harvest movement can result in either mass emigration to surrounding crops or localised aggregation within the harvested field (Scaccini and Furlan, Reference Scaccini and Furlan2021). For N. cymoides, empirical data on such post-harvest behaviour are particularly scarce, limiting the development of targeted management strategies.
Among methodologies used for sampling arthropods, geostatistics – a branch of statistics that examines the spatial structure of data based on population coordinates – has gained importance for its ability to reduce the required number of samples while providing robust spatial insights (Liebhold et al., Reference Liebhold, Rossi and Kemp1993). Geographic Information Systems (GIS) maps – produced using manual data and remote sensing – enable the prediction and early warning of insect outbreaks and population trends, supporting informed management decisions by researchers and farmers. Recent technological advances in agriculture have further enhanced GIS applications, allowing for more accurate monitoring, detection, and timely pest management (Rano et al., Reference Rano, Afroz and Rahman2022), as in the site-specific integrated pest management (IPM), which relies on pest distribution maps and location-targeted control measures (Park et al., Reference Park, Krell and Carroll2007).
When integrated with GIS, geostatistical tools enable precise mapping of pest populations and support spatially informed management decisions (Liebhold et al., Reference Liebhold, Rossi and Kemp1993; Hassani Pak, Reference Hassani Pak1997). Integrated with GIS, it enables precise mapping, monitoring, and management of pest populations and plant diseases, enhancing decision-making through spatial analysis and predictive modelling (Liebhold et al., Reference Liebhold, Rossi and Kemp1993; Hassani Pak, Reference Hassani Pak1997; Alexa et al., Reference Alexa, Sumalan, Negrea, Bragea, Poiana, Radulov and Lazureanu2010).
Traditional pest monitoring methods often provide limited information on pest progression and spatial dynamics, focusing on local pest presence and damage without assessing broader regional patterns, affected areas, damage severity, or forecasting future distribution trends. Instead, remote sensing and GIS have proven effective for pest monitoring across diverse systems, as applied to locusts (Orthoptera) in the Chinese Dagang wetland (Dong, Reference Dong, Xu, Liu, Du, Ye, Huang and Zhu2019), to monitor Agrotis ipsilon (Hufnagel) (Lepidoptera: Noctuidae) migration in response to climate change (Zeng, Reference Zeng, Liu, Zhang, Liu, Jiang, Wyckhuys and Wu2020), and predict the distribution of different pests (Bressan et al., Reference Bressan, Moral García, Sémétey and Boudon-Padieu2010; Liu and Shi, Reference Liu and Shi2020) including invasive species such as Spodoptera frugiperda Latreille (Lepidoptera: Noctuidae) and Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) (Pretty et al., Reference Pretty, Benton, Bharucha, Dicks, Flora, Godfray, Goulson, Hartley, Lampkin, Morris, Pierzynski, Prasad, Reganold, Rockström, Smith, Thorne and Wratten2018; Ramos et al., Reference Ramos, Kumar, Shabani and Picanço2018). In tomato fields, B. tabaci showed strong edge-originating spatial aggregation, demonstrating that mapping pest distribution with kriging can enable site-specific treatments, reducing pesticide use and costs (Ribeiro et al., Reference Ribeiro, Ramos, de Araújo, Soares, Paes, de Araújo, Bastos, Koch and Picanço2021).
Despite its growing relevance, knowledge of the post-harvest distribution of the false chinch bug remains limited, particularly in canola systems. It was hypothesised that following canola harvest, N. cymoides adults and nymphs would remain aggregated within the harvested field rather than dispersing to surrounding crops, due to shelter provided by crop residues. Therefore, this study aimed to analyse the spatiotemporal distribution and post-harvest dispersal of N. cymoides in a canola field using geostatistical kriging and GIS mapping.
Materials and methods
Study site and sampling methodology
This study was conducted in the experimental farm located on the Campus of Agricultural and Natural Resources, Razi University, Kermanshah, Iran (47°06′14-33″ E, 34°19′10-21″ N). The canola variety grown at the study site was Natalie, with planting taking place in October 2020.
The canola field selected as the study site had an area of 8 hectares and was irrigated using a pressurised sprinkler system. Row spacing was 30 cm, and seeds were sown every 5 cm along each row. The field was managed for commercial production, and IPM practices were implemented to prevent economic losses. During canola growth, two insecticide applications were performed with mixtures of chlorpyrifos and malathion, in accordance with recommended guidelines for the management of the false chinch bug. Harvesting was performed using a combine harvester, which left canola plant residues on the soil surface after harvest. The surrounding landscape featured areas of tree nursery, potato cultivation, pistachio, and apple orchards, all located over 100 m from the perimeter of the experimental canola field.
Following the canola harvest that occurred on 24 June 2021, an investigation was conducted to determine the distribution and movement of N. cymoides within the field, prompted by its presence during the crop growing period. The same experimental set-up was employed in 2022, when canola was harvested on 24 June, but the low number of false chinch bugs captured in the traps did not allow us to conduct a geostatistical analysis. Forty sampling points were determined using ArcGIS 10.3 software and a digital map of the study field (Esri, 2014). These points were arranged in eight cardinal directions radiating from the centre of the harvested field, with five points spaced approximately 60 m apart along each direction. This design resulted in a sampling radius of about 300 m per direction, covering a total study area of more than 28 hectares (fig. 1). A spacing of approximately 60 m between adjacent points was selected to ensure adequate resolution for detecting spatial dependence, based on the general principle that sampling intervals should be less than half the anticipated range of spatial autocorrelation (Webster and Oliver, Reference Webster and Oliver2007). The 300 m transect length was chosen to encompass the entire harvested field and a sufficient buffer of the surrounding landscape. The precise location of each pre-determined point was then recorded and marked in the field using a GPS device. During the 2021 post-harvest monitoring period (24 June to 3 September), the mean daily temperature was 28.6°C (range: 26.3–33.1°C), and the mean relative humidity was 13.2% (range: 7.0–28.0%). Precipitation was negligible, with only two recorded rainfall events of 0.1 mm each. These conditions are typical of the warm, dry summer climate of Kermanshah province. At each sampling point, both sticky traps and pitfall traps were initially deployed. Sticky traps consisted of a 21 × 29.7 cm2 panel covered with glue. Pitfall traps consisted of a beaker buried in the soil whose lip level was kept at the ground surface, and they were half-filled with a solution of water and detergent to reduce surface tension and trap insects.
Pitfall traps are set in and around the canola field in 2021 (the canola field boundary is identified with a red line).
Figure 1 Long description
The image shows an aerial view of a canola field with concentric circles overlaid on the landscape. The field boundary is marked with a red line. There are forty numbered sampling points distributed across the field and surrounding area. These points are arranged in eight cardinal directions radiating from the center, with numbers ranging from 0 to 39. The circles and lines create a grid-like pattern over the field, indicating a systematic sampling design. Buildings and roads are visible in the surrounding area, providing context to the field's location.
Traps were placed immediately after canola harvest, on 24 June 2021. Trap monitoring was conducted every 4–16 days, starting from 26 June 2021, and continued for more than 2 months post-harvest. Since sticky traps proved ineffective at capturing N. cymoides from the first sampling dates, their use was discontinued in favour of pitfall traps only. This method was considered appropriate given that N. cymoides is known to shelter in soil cracks and under plant residues post-harvest, making surface activity the primary mode of local dispersal during this period (Scaccini and Furlan, Reference Scaccini and Furlan2021). During each monitoring event, the content of the pitfall traps was collected and transported to the laboratory for morphological analysis under the stereomicroscope (Leica, WILD M3B). N. cymoides individuals captured in pitfall traps at each location were counted and separated into nymphs and adults. Insect count data, along with the corresponding GPS coordinates, were then entered into ArcGIS 10.3 software, and spatial distribution maps at different time points were generated using the kriging interpolation method. A paired t-test (α = 0.05) was conducted to compare the number of nymphs and adults captured in each pitfall trap across all sampling dates.
For the 2022 data, given the low number of collected insects, we tested whether traps placed within the canola field captured different numbers of individuals, by nymphs or adults, compared with those positioned outside the field. Traps were grouped into four replicate blocks corresponding to the cardinal directions to account for spatial replication. Differences in nymph and adult abundance between treatments were analysed using a generalised linear mixed model with a Poisson error distribution and a log link function. Treatment (in- vs. off-field) was included as a fixed effect, and models were fitted separately for each sampling date with at least one insect captured. Analyses were performed in SAS version 9.4 (SAS Institute, 2016) with the GLIMMIX procedure. Pairwise comparisons were adjusted using Tukey’s honestly significant difference test (α = 0.05). Model fit and residuals were evaluated using diagnostic plots generated by the procedure.
Semivariogram calculation for spatial dependence
A semivariogram is a fundamental geostatistical tool used to describe the spatial dependence of a regionalised variable – such as pest population density – across a study area. It quantifies the spatial correlation between measured data points as a function of both distance and direction (Madani, Reference Madani1994). The semivariogram, denoted as γ(h), is calculated using the following formula:
\begin{equation*}\gamma (h) = \frac{1}{{2N(h)}}\sum\nolimits_{i = 1}^{N(h)} {{{[z({x_i}) - z({x_i} + h)]}^2}} \end{equation*} In this formula, h is the horizontal distance between pairs of samples, z(xi) and z(xi + h) are the values of the regional variable at two points with coordinates xi and xi + h. Each of the calculated values of variogram or semivariogram, along with their corresponding h value, identifies a point on the (
$\gamma $–h) coordinate system (Madani, Reference Madani1994).
The equations for each of the models used are summarised in table 1.
List and details of model equations used in this study
Table 1 Long description
This table provides a comparison of six different model equations utilized in the study, highlighting their unique mathematical expressions. The spherical model includes conditions based on the relationship between two variables, while the exponential model describes a decay process. The Gaussian model involves a squared term, indicating a bell-shaped curve. The linear model represents a direct proportionality between variables. The J-Bessel model incorporates a Bessel function, often used in wave propagation studies, and the hole effect model includes a cosine function, suggesting periodic behavior. Each model is characterized by its specific approach to describing spatial or temporal relationships, offering diverse methods for analysis.
In these equations, γ(h) is the semivariogram for distance h, C 0 is the nugget effect, C is the structural variance, and R is the range, the distance where the semivariogram reaches a plateau, J₀ denotes the Bessel function of order zero, λ controls the wavelength, a is the damping factor, and b controls the frequency and decay of oscillations representing cyclic spatial patterns.
Kriging interpolation for spatial distribution maps
Kriging interpolation was used to map the spatial distribution of N. cymoides density. This geostatistical method applies a weighted moving average, estimating unsampled values based on neighbouring observations and weights (λᵢ) derived from a fitted semivariogram, minimising error variance and ensuring unbiased predictions. Specifically, the kriging equation is (Webster and Oliver, Reference Webster and Oliver2007):
\begin{equation*}Z_{{x_0}}^* = \mathop \sum \limits_{i = 1}^n {\lambda _i}{Z_{\left( {{x_i}} \right)}}\end{equation*} In this equation, 
$Z_{{x_0}}^*$ is the estimated value at point x 0, 
${Z_{\left( {{x_i}} \right)}}$ is the real value at point xi, n is the number of neighbouring locations, and λi is the statistical weight assigned to the sample 
${Z_{({x_i})}}$ located in the neighbouring of point x 0.
The root-mean-square error statistics were used to evaluate the performance of different interpolation models.
\begin{equation*}{\text{RMSE}} = \sqrt {\frac{1}{n}\mathop \sum \limits_{i = 1}^n {{({Z_{{x_i})}} - Z_{({x_i})}^*)}^2}} \end{equation*} In this equation, 
$Z_{{x_i}}^*$ is the estimated value at point xi, 
${Z_{\left( {{x_i}} \right)}}$ is the real value at point xi, and n is the number of points. Mathematically, when this measure equals zero, it means that the estimated value is exactly equal to the true quantity.
Results
Following canola harvest on 24 June 2021, pitfall traps were monitored regularly for over 2 months. No N. cymoides individuals were captured during the first 4 weeks of sampling (26 June to 13 July). The first captures occurred on 20 July, when a total of 18 nymphs and 11 adults were recorded; however, numbers remained too low for geostatistical analysis. From 27 July onwards, trap catches increased markedly, particularly at locations within the harvested field, and remained elevated through early September. Sufficient captures for kriging interpolation were obtained on five sampling dates: 27 July, 5 August, 21 August, 25 August, and 3 September. Throughout the monitoring period, trap densities were consistently higher inside the harvested canola field than in the surrounding landscape.
In the second year of sampling (2022), pitfall traps captured very few N. cymoides individuals, with mean catches per trap remaining close to zero across most sampling dates for both nymphs and adults (Figure S1), not allowing geostatistical analyses. Despite being modest, a peak in abundance – particularly of nymphs – was recorded from late June to early July, with captures first detected on 29 June (Figure S1). Only two sampling dates yielded both nymphs and adults in pitfall traps; therefore, the analysis was limited to these two occasions. Comparisons between in-field and off-field traps revealed a significantly higher abundance of nymphs on 29 June in traps placed within the harvested canola field compared with those located outside it (F 1,6 = 15.33; p = 0.0078), while no significant differences were detected on the subsequent sampling date (F 1,6 = 0.01; p = 0.9793) for adult captures on either date (29 June: F 1,6 = 1.92; p = 0.2150; 6 July: F 1,6 = 0.01; p = 0.9811; Figure S1).
The results of the statistical analysis of the N. cymoides spatial distribution in 2021 are summarised in table 2, which compares the performances of the geostatistical models tested. The exponential model was most frequently selected as the best fit based on the lowest error values or other model selection criteria. Furthermore, analysis of the semivariograms indicated a high degree of spatial correlation in all cases, signifying that the distribution patterns were significantly structured and not due to random or spatially unstructured effects. Analysis of the best-fit models revealed that the percentage of spatial dependence for both nymph and adult stages ranged from 63.9% to 100% when using exponential, spherical, J-Bessel, and Hole effect models.
Geostatistical information on the best variogram model for determining the distribution map of Nysius cymoides using kriging method on different sampling dates in 2021
Table 2 Long description
The table presents geostatistical data on variogram models used to map the distribution of Nysius cymoides across different sampling dates in 2021. It compares nymph and adult instars, showing that exponential models are frequently used for nymphs, while adults utilize a variety of models including circular, J-Bessel, and hole effect. The nugget effect is generally zero for nymphs, indicating no spatial variability at small scales, while adults show some variability. The range and partial sill values vary, with the range extending up to 375.39 for adults. The percentage of spatial dependency is consistently high, often reaching 100 percent, except for some adult samples. RMSE values indicate the accuracy of the models, with lower values suggesting better model fit, ranging from 3.67 for adults on August 5 to 105.73 for nymphs on August 25.
RMSE, root-mean-square error.
Contour maps generated via kriging interpolation showed that the highest concentrations of both N. cymoides nymphs and adults remained within the harvested canola field for up to 2 months post-harvest (figs 2 and 3). However, insect catches in pitfall traps 1 month after harvest were minimal. Notably, trap catches first increased on 27 July. Nymphs showed in general higher numbers than those of the adults on all sampling dates, as confirmed by a paired t-test (t = 7.52, p < 0.0001). No directional bias in trap captures was detected for either nymphs or adults across the eight radial transects.
Kriging model maps with contour lines to display the distribution of Nysius cymoides nymphs in and around the canola field at five post-harvest dates: (A) 27 July, (B) 5 August, (C) 21 August, (D) 25 August, and (E) 3 September 2021 (numbers refer to estimated N. cymoides individuals trap−1; red points indicate the locations of pitfall traps; the boundary of the study field is marked with a white line).
Figure 2 Long description
The image consists of five contour maps labeled A to E, each displaying the distribution of nymphs in a canola field at different post-harvest dates. Map A shows the distribution on 27 July, with contour lines indicating varying concentrations of nymphs. Map B represents the distribution on 5 August, showing changes in nymph concentrations. Map C illustrates the distribution on 21 August, with contour lines marking different levels of nymph presence. Map D depicts the distribution on 25 August, highlighting areas of higher concentration. Map E shows the distribution on 3 September, with contour lines indicating the spread of nymphs. Each map includes red points marking pitfall trap locations and a white line outlining the study field boundary. The maps use kriging interpolation to visualize the data, with numbers referring to estimated individuals trapped.
Kriging model maps with contour lines to display the distribution of Nysius cymoides adults in and around the canola field at five post-harvest dates: (A) 27 July, (B) 5 August, (C) 21 August, (D) 25 August, and (E) 3 September 2021 (numbers refer to estimated N. cymoides individuals trap−1; red points indicate the locations of pitfall traps; the boundary of the study field is marked with a white line).
Figure 3 Long description
The image consists of five kriging model maps labeled A to E, each displaying contour lines over a satellite view of a canola field. Each map shows the distribution of individuals at different post-harvest dates. Red points indicate pitfall trap locations. The maps are as follows: (A) 27 July, (B) 5 August, (C) 21 August, (D) 25 August and (E) 3 September 2021. Contour lines represent estimated individual counts, with numbers indicating trap estimates. The boundary of the study field is marked with a white line. The maps illustrate changes in distribution over time, with varying concentrations of individuals within the field. Each map includes a scale bar indicating 100 and 200 meters and coordinates are marked along the edges.
Discussion
Geostatistical analyses in different crop-pest systems have revealed pest-specific spatiotemporal patterns, with some species concentrating in field centres (Bressan et al., Reference Bressan, Moral García, Sémétey and Boudon-Padieu2010), while others exhibit higher densities along field edges (Ribeiro et al., Reference Ribeiro, Ramos, de Araújo, Soares, Paes, de Araújo, Bastos, Koch and Picanço2021). Studies also highlight the role of surrounding non-crop vegetation as sources of early infestation (Carvalho et al., Reference Carvalho, Pereira, Sarmento, Farias, Lima, Santos and Picanço2020), and the influence of soil type on pest distribution, being also related to insect oviposition behaviour (Rogers et al., Reference Rogers, Guimarães, Evans and Rogers2015).
In geostatistical research, the effective range is a key metric for optimising pest monitoring and sampling efficiency. Sampling distances can be increased by up to 75% of the effective range without compromising accuracy (Hassani Pak, Reference Hassani Pak1997). In our study, the minimum effective range for N. cymoides in canola was approximately 163 m (as obtained from the best-fit semivariogram model), indicating that future sampling intervals could extend to about 123 m. Spatial analysis using kriging revealed an aggregated distribution pattern of N. cymoides, consistent with findings by Mohaghegh et al. (Reference Mohaghegh, Amir-Maafi, Shahrokhi and Pirhadi2016), who reported similar aggregation using Taylor’s power law and Iwao’s regression in canola fields in Iran. However, our geostatistical approach provides more precise spatially explicit maps, offering deeper insight into pest density and aggregation intensity within crop fields.
Notably, our results seem to be in contrast with the initial hypothesis that N. cymoides would readily disperse from canola fields following harvest, as previously observed by González et al. (Reference González, Aldanondo, Barredo and Alarcón-Elbal2022) and D. Scaccini (pers. obs.). Contrary to expectations, a considerable proportion of false chinch bug nymphs and adults appeared to remain within the field, likely sheltering in the soil or under residual plant debris. This behaviour aligns with known preferences of the false chinch bug for plant residues and higher infestations in fields that were not ploughed (Chapelin-Viscardi et al., Reference Chapelin-Viscardi, Tourton and Matocq2017, Reference Chapelin-Viscardi, Tourton, Ruck, Pozzebon and Scaccini2021), such as in no-till systems (Scaccini and Furlan, Reference Scaccini and Furlan2019). Moreover, the insects captured in our field – particularly nymphs – may also have originated from eggs laid and hatching in the soil, as observed in other Nysius spp. that deposit eggs in soil cracks and plant residues (Tiwari and Wratten, Reference Tiwari and Wratten2019).
The high degree of spatial dependence detected in the semivariogram analysis (63.9–100%) indicates that the post-harvest distribution of N. cymoides was not random but strongly structured by factors operating within the field itself, such as the heterogeneous distribution of crop residues or soil moisture. Furthermore, the observation that nymphs exhibited a broader spatial dispersion than adults suggests differing post-harvest habitat requirements between life stages. Nymphs, being less mobile and more susceptible to desiccation, may have dispersed more widely in search of suitable microhabitats, whereas adults likely remained within the harvested area where residual plant material provided both shelter and a continuing food source in the form of fallen seeds. This pattern is consistent with the known biology of Nysius spp., where nymphal development is closely tied to ground-level microclimatic conditions (Scaccini and Furlan, Reference Scaccini and Furlan2021).
Although the 2022 dataset was limited in terms of captures, its inclusion provides complementary evidence to the overall analysis. While the small sample size only partially supports this interpretation, the results are consistent with the limited dispersal capacity of nymphs following canola harvest as observed in 2021. Moreover, the very low numbers of both nymphs and adults captured underscore the outbreak-prone dynamics of N. cymoides populations, which are known to fluctuate markedly between years, as documented for this and other Nysius spp. (McDonald and Farrow, Reference McDonald and Farrow1988; Scaccini and Furlan, Reference Scaccini and Furlan2021). Such variability may be influenced by climatic conditions, host plant availability, or local agronomic practices, as documented for other Nysius spp. (McDonald and Farrow, Reference McDonald and Farrow1988; Scaccini and Furlan, Reference Scaccini and Furlan2021). These findings reflect the exploratory nature of our study and highlight the importance of multi-year monitoring to better capture the full range of population dynamics of N. cymoides. From a pest management perspective, these results suggest that IPM strategies should remain flexible and account for both outbreak years and periods with negligible pest pressure.
In Nysius spp., migration and infestation dynamics are influenced by multiple factors, including insect species, sex, developmental stage, seasonal timing, time of day, and environmental conditions such as temperature and other abiotic factors (Kehat and Wyndham, Reference Kehat and Wyndham1974; McDonald and Farrow, Reference McDonald and Farrow1988; Demirel and Cranshaw, Reference Demirel and Cranshaw2006; Wei, Reference Wei2014; Moradi-Vajargah and Parry, Reference Moradi-Vajargah and Parry2017). Weed cover also plays a key role in shaping pest movement and colonisation patterns. For example, Nysius spp. have been linked to the removal or degradation of nearby host weeds, which prompts the insects to disperse in search of alternative food sources (Wene, Reference Wene1958; Tiwari and Wratten, Reference Tiwari and Wratten2019; Scaccini and Furlan, Reference Scaccini and Furlan2021). Consequently, the removal of weeds near cultivated fields should be carefully timed and managed to avoid promoting pest dispersal.
Agricultural practices such as residue removal, ploughing, and weed management can disrupt pest microhabitats and reduce N. cymoides populations in subsequent crops (Whitworth et al., Reference Whitworth, McCornack and Davis2012; Mohaghegh et al., Reference Mohaghegh, Pirhadi and Amini-Khalaf2015; Scaccini and Furlan, Reference Scaccini and Furlan2021; Hangartner and McDonald, Reference Hangartner and McDonald2025). These strategies align with broader recommendations for hemipteran pest management, which emphasise controlling ground cover and weeds to suppress pest build-up (Rijal et al., Reference Rijal, Joyce and Gyawaly2021). On the other hand, while removing plant residues may help manage Nysius pests, it can negatively affect ecosystem services by reducing soil organic matter, fertility, and crop yield (Blanco-Canqui and Lal, Reference Blanco-Canqui and Lal2009), highlighting the need for case-by-case evaluation and trade-offs. Additional management practices against N. cymoides could include the use of trap crops planted near canola fields that may attract and concentrate this pest, enabling their targeted removal through attract-and-kill strategies (McDonald and Smith, Reference McDonald and Smith1988; Alizadeh et al., Reference Alizadeh, Avand-Faghih, Mohaghegh and Porshekoh2009; Pirhadi et al., Reference Pirhadi, Mohaghegh Nishaboori, Shahverdi, Nasrolahi, Mohiseni, Nabati and Khorasanizadeh2012; Tiwari et al., Reference Tiwari, Dickinson, Saville and Wratten2018). Crop rotation and selection may also play a critical role in pest management; for instance, the previous canola cultivation was shown to increase N. cymoides populations that later damaged soybean in Italy, whereas soybean following winter wheat did not show such high infestation levels (Mohaghegh et al., Reference Mohaghegh, Pirhadi and Amini-Khalaf2015). Insecticide use should complement these measures but must be applied judiciously to minimise non-target and environmental impacts (Barzman et al., Reference Barzman, Bàrberi, Birch, Boonekamp, Dachbrodt-Saaydeh, Graf, Hommel, Jensen, Kiss, Kudsk, Lamichhane, Messéan, Moonen, Ratnadass, Ricci, Sarah and Sattin2015; Kogan and Higley, Reference Kogan and Higley2019), in accordance with established guidelines for pest control.
However, limitations of this study should be acknowledged. The geostatistical analysis was based on data from a single canola field during one outbreak season (2021), and the observed aggregation patterns may therefore not fully represent the post-harvest behaviour of N. cymoides under different environmental conditions or landscape contexts. In addition, the low population densities recorded in 2022 precluded robust multi-year comparisons, highlighting the outbreak-prone dynamics of this pest. Consequently, the findings presented here should be interpreted as preliminary, and further studies across multiple sites and seasons are required to validate the generality of the observed dispersal patterns.
Conclusions
Kriging maps derived from variogram analysis effectively visualised the spatial distribution patterns and temporal dynamics of N. cymoides populations during the canola post-harvest period. Our understanding of the false chinch bug movement and dispersal after crop harvest remains limited. As the geostatistical analysis was based on a single field during an outbreak season, the findings should be regarded as site- and time-specific pending further validation. The post-harvest behaviour of hemipteran pests is known to vary in response to annual weather conditions and landscape structure. Therefore, further research across multiple fields and seasons is needed to clarify its behaviour and to determine effective pest management strategies. Despite these limitations, this study provides useful insights into the dispersal behaviour of N. cymoides, which should be carefully considered when planning effective IPM strategies.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0007485326101096.
Acknowledgements
We are grateful to Razi University and the Management and Planning Organization of Kermanshah Province for supporting this project.
Author contributions
A.A.Z. conceived and designed research. A.A.Z., H.R.P., B.M., Z.D., Z.S., and M.H. performed field sampling. M.A., N.M.N., F.J., and D.S. provided scientific consultation and expert advice in conducting the research. A.A.Z. and D.S. analysed data. A.A.Z. and B.M. wrote the original draft of the manuscript. All authors read and approved the manuscript.
Financial support
This project was funded by Razi University and the Management and Planning Organization of Kermanshah Province through the grant No. 15631-222538.
Competing interests
The authors have declared that no conflict of interest exists.
