Hierarchical climatic forcing and population rhythms of bark and ambrosia beetles in tropical rainforests of Northern Borneo

Hierarchical climatic forcing and population rhythms of bark and ambrosia beetles in tropical rainforests of Northern Borneo | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 18 March 2026 V1 Latest version Share on Hierarchical climatic forcing and population rhythms of bark and ambrosia beetles in tropical rainforests of Northern Borneo Authors : Evahtira Gunggot , Roger Beaver , Maria Lardizabal , Jonathan Lucas , Sandra George , Anastasia Rasiah , Wilson Wong , and Naoto Kamata 0000-0002-8818-6991 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.177383746.63722976/v1 152 views 57 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract This study investigated the temporal dynamics and environmental drivers of bark and ambrosia beetles (Curculionidae: Scolytinae and Platypodinae) in a tropical rainforest in Northern Borneo. Utilizing an extensive long-term dataset (2017–2020), we employed Multi-Taper Method (MTM) spectral analysis and lagged Path Analysis to decode the underlying structure of population fluctuations. To isolate true cyclical relationships and ensure statistical rigor, both climatic variables and insect capture data were rigorously detrended prior to the path analysis.Our results revealed a multi-scale periodicity, most notably a dual-significant 35-day cycle in Total Trap Capture (TTC). We propose that these short-term oscillations are fundamentally rooted in intrinsic generation cycles reflecting the developmental duration of individual insect species. Within this framework, the 35-day MJO acts primarily as a ”gatekeeper” rather than the sole driver, periodically providing resources through windthrow while modulating flight activity through rainfall inhibition. These dynamics are further governed by a hierarchy of ”ecological memory,” where climatic forcing at specific lags determines realized abundance. We identified two distinct sub-annual biological legacy tiers: a ~3-month lag representing a host-stress response pathway, and an ~8-month lag driven by the accumulation of multivoltine generations and the obligate incubation period required for fungal symbionts to degrade wood substrates.Based on these findings, we propose the Resonance Hypothesis, suggesting that intrinsic rhythms are phase-locked by intraseasonal pulses and modulated by multi-month biological legacies. These cycles are ultimately nested within supra-annual climatic modes, such as ENSO and the IOD, which dictate long-term population baselines. Our results suggest that the apparent stochasticity of tropical insects is an interference pattern created by these overlapping temporal rhythms. Hierarchical climatic forcing and population rhythms of bark and ambrosia beetles in tropical rainforests of Northern Borneo Abstract: This study investigated the temporal dynamics and environmental drivers of bark and ambrosia beetles (Curculionidae: Scolytinae and Platypodinae) in a tropical rainforest in Northern Borneo. Utilizing an extensive long-term dataset (2017–2020), we employed Multi-Taper Method (MTM) spectral analysis and lagged Path Analysis to decode the underlying structure of population fluctuations. To isolate true cyclical relationships and ensure statistical rigor, both climatic variables and insect capture data were rigorously detrended prior to the path analysis.Our results revealed a multi-scale periodicity, most notably a dual-significant 35-day cycle in Total Trap Capture (TTC). We propose that these short-term oscillations are fundamentally rooted in intrinsic generation cycles reflecting the developmental duration of individual insect species. Within this framework, the 35-day MJO acts primarily as a ”gatekeeper” rather than the sole driver, periodically providing resources through windthrow while modulating flight activity through rainfall inhibition. These dynamics are further governed by a hierarchy of ”ecological memory,” where climatic forcing at specific lags determines realized abundance. We identified two distinct sub-annual biological legacy tiers: a and an ~8-month lag driven by the accumulation of multivoltine generations and the obligate incubation period required for fungal symbionts to degrade wood substrates.Based on these findings, we propose the Resonance Hypothesis, suggesting that intrinsic rhythms are phase-locked by intraseasonal pulses and modulated by multi-month biological legacies. These cycles are ultimately nested within supra-annual climatic modes, such as ENSO and the IOD, which dictate long-term population baselines. Our results suggest that the apparent stochasticity of tropical insects is an interference pattern created by these overlapping temporal rhythms. (264 words) Keywords: insect population cycle, hydroclimatic cycle, generation cycle, resonance hypothesis, host stress, legacy effect Introduction Tropical insect populations have historically been described as weakly structured or largely stochastic compared to their temperate counterparts. In temperate regions, strong seasonal shifts in temperature and photoperiod dictate synchronized life cycles, driving regular population dynamics such as winter diapause (Saunders 2012, Tauber, et al. 1986). In contrast, perpetually wet tropical forests such as those in Southeast Asia lack a distinctive dry season or cold winter. This relatively stable environment allows insects to reproduce continuously, resulting in overlapping generations and continuous multivoltine life cycles (Kishimoto-Yamada and Itioka 2015, Wolda 1978, Wolda 1988). Consequently, long-term studies in tropical rainforests often report irregular fluctuations in abundance, with population dynamics varying unpredictably among species and across years rather than following a strict calendar-bound seasonality (Ueno, et al. 2021). However, this apparent stochasticity may obscure deeper structural rhythms. The tropical climate in northern Borneo is not a flat, constant baseline, but is dynamically structured by interacting meteorological forcings (Qian 2008). While the region is often classified as aseasonal, its precipitation and temperature are heavily modulated by intraseasonal events like the Madden-Julian Oscillation (MJO) and interannual global modes such as the El Niño–Southern Oscillation (ENSO) (Aldrian 2003, Ropelewski and Halpert 1987, Zhang 2005), which are further buffered by regional topography and the Borneo Vortex (Chang, et al. 2005, Cheang 1977, Ng, et al. 2020). These forces create predictable but non-annual hydroclimatic cycles that dictates resource availability and physical environmental stress. Despite the profound impact of these climatic cycles, traditional population ecology often struggles to link tropical insect dynamics to weather because it searches for immediate, linear responses (Estay, et al. 2009, Ives 1995). In reality, biological systems operate with distinct time lags. To understand these dynamics, it is necessary to apply the framework of ecological memory, which is defined as the capacity of past environmental states to influence present-day ecological responses (Hughes, et al. 2019, Ogle, et al. 2015). For many forest insects, a severe climatic event (such as a storm of drought) does not merely trigger immediate dispersal; it generates a delayed resource bank of dead or stressed vegetation that requires months to biologically mature before it can be colonized (Anderegg, et al. 2012, Jactel, et al. 2012). Wood-boring beetles (Coleoptera: Curculionidae: Scolytinae and Platypodinae) represent an ideal model system to investigate these lagged dynamics. As colonizers of woody debris ranging from primary to secondary successional stages, these beetles are highly sensitive to the shifting physiological state of their host (Kirkendall, et al. 2015). Their population dynamics are not just dependent on ambient weather, but on the successional decay of wood (Peng, et al. 2022). While both groups frequently exhibit broad host-ranges, their colonization success is governed by distinct biological checkpoints. For early-successional species, establishment is often triggered by chemical legacies, specifically the tracking of ethanol cues released during the anaerobic fermentation of stressed tissues (Kelsey and Gladwin 2003). In contrast, certain ambrosia beetles associated with later successional stages are fundamentally tied to a nutritional-incubation lag; as obligate fungivores, their population peaks are limited by the moisture-dependent period required for their symbiotic fungi to reach sufficient biomass (Biedermann and Vega 2020, Hulcr and Dunn 2011, Li, et al. 2015). Thus, even among host-generalists, temporal dynamics are differentially regulated by the time required for host-tissue fermentation versus fungal proliferation. To disentangle this complexity, this study applies a macroecological approach (Brown and Maurer 1989, Lawton 1999) to a high-resolution, long-term trapping dataset from a Bornean tropical rainforest. By employing spectral and lagged path analyses, this study aims to bypass the noise of local community assembly to identify the dominant temporal frequencies governing beetle populations. Ultimately, this research investigates how immediate physical filters (like rainfall) and delayed biological legacies (like fungal incubation) interact with global climate modes, proposing a resonance hypothesis for how hyperdiverse tropical insect communities maintain dynamic stability in a changing environment. Material and Methods Study Site/Insect Sampling/Data source Insect data were obtained from a long-term monitoring program in the tropical forests of Long Miau, Sabah, Malaysia (4°23′–27′ N, 115°42′–47′ E). Sampling consisted of 80 continuous biweekly events from April 2017 to May 2020, conducted across three land-use types mature primary forest, disturbed secondary forest, and a monoculture rubber plantation ( Hevea brasiliensis ) using flight-intercept bottle traps baited with 95% ethanol. A total of 7,257 individuals across 154 species (134 Scolytinae and 20 Platypodinae) were recorded. Detailed descriptions of site topography and field protocols are provided in Gunggot et al. (Gunggot, et al. 2025). To ensure that fluctuations in beetle abundance were not artifacts of varying trap exposure times, raw capture data (Table S1) were standardized to two distinct temporal scales: uniform 14-day biweekly periods (Table S2) and calendar months (Table S3). For both scales, abundance was recalculated using a proportional weighting method based on the daily capture rates of the original sampling intervals that overlapped or spanned the boundaries of the target windows. Specifically, biweekly intervals were standardized to end on a fixed Saturday, while monthly data were aligned strictly with calendar months. Total Trap Capture (TTC) was defined as the aggregate abundance for each standardized period, and subsequent species-level analyses were restricted to the eleven most dominant taxa, each represented by more than 100 individuals. Climatic Data Daily meteorological variables, including temperature (°F), relative humidity (%), and rainfall (mm), were obtained from World Weather Online (https://www.worldweatheronline.com) for the Long Miau region, Sabah, Malaysia. The primary dataset spanned April 2017 to May 2020, comprising 1,128 daily observations. To align these data with the insect sampling scales, biweekly climatic variables were calculated by aggregating daily records into biweekly means for temperature and humidity, and cumulative totals for rainfall (Table S4). In contrast, monthly weather data were directly sourced and downloaded from the World Weather Online database to correspond with calendar-month sampling windows. Statistical Analysis All statistical analyses were performed using R-software ver.4.4.0 (R Core Team 2024). Multi-Taper Method (MTM) To identify periodicities in climatic and beetle population data, we conducted Multi-Taper Method (MTM) spectral analysis using the “mtmML96” function within the astrochron package (Meyers 2014). The analysis utilized a 1-day sampling interval for climatic variables (1,128 observations), establishing a Nyquist frequency of 1 cycle per day, while insect data were analyzed using a biweekly sampling interval (where 1 unit equals 14 days). By employing three orthogonal Slepian tapers (NW = 2), the method minimized spectral leakage and provided high-resolution estimates. To isolate significant narrow-band signals from an autocorrelated background, we implemented the robust red-noise estimation procedure developed (Mann and Lees 1996). True harmonic signals were distinguished from stochastic noise using a dual-significance criterion: peaks were required to simultaneously exceed the 95% confidence level (CL) against the AR(1) red-noise background estimated via a median-smoothed spectrum approach and the 90% CL for harmonic signals via Thompson’s F-test. Path Analysis To investigate the causal relationships between climatic variables and beetle abundance (TTC), we conducted a series of Path Analyses using the lavaan package (ver. 0.6-19) (Rosseel 2012) in R. To isolate the intrinsic coupling between climate and insect dynamics, all time-series datasets were first detrended using Ensemble Empirical Mode Decomposition (EEMD) via the “Rlibeemd” package (Helske and Luukko 2023). This pre-processing step involved decomposing each series into a finite set of Intrinsic Mode Functions (IMFs) and subtracting the final residual component representing non-stationary long-term trends and decadal noise to produce stationary datasets (Tables S5, S6, S7, and S8). This ensures that the resulting path coefficients (β) reflect true environmental forcing rather than coincidental alignment of seasonal cycles. We specified a structural model where detrended Temperature, Humidity, and Rainfall served as exogenous predictors of detrended TTC. Path coefficients were calculated using Maximum Likelihood (ML) estimation. Given that biological responses in insect populations are often delayed, we implemented a time-lagged analysis to identify the window of maximum environmental influence. For the biweekly scale, lags ranging from 1 to 22 units (14–308 days) were evaluated; for the monthly scale, lags from 1 to 16 months were tested. The optimal lag for each scale was determined by comparing the Coefficient of Determination ( R 2 ) and the significance of standardized path coefficients ( β ). Model fit was further evaluated using the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA). Finally, visual path diagrams were generated using the “semPlot” package (Epskamp 2022) to illustrate the magnitude and direction of these environmental drivers. Results Multi-frequency structure of the climate Spectral analysis of Temperature data revealed a highly stable sub-seasonal periodicity of 35 days, which attained statistical significance under both the Multi-Taper Method (MTM) (Table 1) and the Ensemble Empirical Mode Decomposition (EEMD) frameworks. Specifically, this 35-day cycle was physically isolated as the fifth Intrinsic Mode Function (IMF 5) in the EEMD results (Figure S1(a)). In contrast, while IMFs 1–4 captured high-frequency fluctuations representing short-term thermal variations, these components failed to reach statistical significance within the MTM framework. Beyond the sub-seasonal scale, MTM analysis identified robust periodicities at the annual (294–393 days) and supra-annual (695–999 days) scales, which were explicitly represented by IMF 8 and IMF 9, respectively. Rainfall patterns exhibited multi-scale complexity characterized by several high-frequency periodicities. MTM analysis identified distinct peaks at the 3, 17, 26–29, and 32–34-day scales (Table 1). Among these, only the 33-day periodicity achieved statistical significance against both the median and red-noise backgrounds. These high-frequency fluctuations align with the EEMD results; specifically, IMFs 3 and 4 correspond to the 3- and 7-day scales, while IMF 5 represents the 26–29 and 32–34-day periodicities (Figure S1(b)). Other high-frequency peaks were significant only when tested against the red-noise background. Furthermore, a discrete 120-day peak was detected against the AR(1) red-noise model. Beyond these high-frequency variations, rainfall displayed stable seasonal and long-term oscillations at semi-annual (181–195 days), annual (314–384 days), and supra-annual (788–930 days) scales, which were explicitly captured by IMF 7, IMF 8, and IMF 9, respectively. Humidity patterns largely mirrored those of temperature, although they exhibited distinct variance in the lower frequency domains (Table 1). While high-frequency fluctuations were identified at the 5-day and 156–163-day scales, these components did not meet the dual-significance criteria. These periodicities were also captured in the EEMD results as IMF 2 and IMF 7, respectively (Figure S1(c)). The most robust harmonic cycles for humidity were identified at the annual (311–395 days) and supra-annual (836–1,131 days) scales, corresponding to IMF 8 and IMF 9. Notably, the oscillations decomposed into IMFs 3 through 6 by the EEMD were not identified as statistically significant cycles within the MTM framework. Table 1. Summary of environmental periodicities identified via Multi-Taper Method (MTM) spectral analysis. Temperature 34─36 35 35 215─2191 294─393 294─393 695─999 695─999 Rainfall 3 17 26─29 32─34 33 33 120 144─930 181─195 181─195 313─384 314─384 788─930 788─930 Humidity 5 156─163 219─2191 311─395 311─395 836─1131 836─1131 Note: ’Dual Significant’ indicates peaks meeting both Red Noise (>95%) and Harmonic (>90%) confidence levels. Power represents the normalized MTM spectral amplitude. Values exceeding the 1,128–day observation period are derived from a 6-year spectral projection. Insect Population Cycles Spectral analysis of the Total Trap Capture (TTC) and individual species revealed a structured hierarchy of periodicities across multiple temporal scales (Table 2). At the high-frequency level, the aggregate beetle assemblage exhibited distinct harmonic cycles at the 2.05–2.06 units (28.7–28.8 days) and 2.19–2.21 units (30.7–30.9 days) scales. Notably, a prominent sub-seasonal cycle was identified at 2.47–2.5 units (34.6–35.0 days); this signal achieved dual statistical significance and closely aligned with the 35-day sub-seasonal temperature periodicity, suggesting a strong thermal driver for these population fluctuations. The IMF 1 decomposed by the EEMD corresponds to these short-term cycles (Figure S2(a)). Beyond these immediate cycles, additional TTC periodicities were detected at sub-annual scales ranging from 2.70–5.33 units (37.8 to 74.6 days), which were further validated by oscillations captured in the IMF 2 of the EEMD analysis. Moving into the broader temporal domain, long-term harmonics occurred at 8.33 units (~116.6 days) and 22.22 units (~311.1 days), which were isolated within IMF 3 and IMF 4, respectively. Finally, a primary supra-annual background signal was centered at 133.33 units (~1,866.6 days), indicating underlying population trends that extend well beyond annual cycles. Individual species demonstrated varying degrees of rhythmic complexity across these temporal categories, reflecting species-specific biologies (i.e. life cycles) and responses to environmental drivers. Eidophelus ( Scolytogenes ) sp. SA1 (Sp. 1) exhibited a broad spectrum of significant periodicities, ranging from high-frequency pulses to supra-annual oscillations. According to MTM analysis, high-frequency activity was concentrated in short-term pulses of 2.52–2.64 units (~35.3–37.0 days) and reached up to 4.30–4.35 units (~60.2–60.9 days). These results were well-supported by IMF 1 (Figure S2(b)). Intermediate rhythms were detected at 5.06–5.19 units (~70.84–72.66 days) and 7.0–7.17 units (~98.0–100.38 days), representing distinct sub-annual activity captured in IMF 2 and IMF 3, respectively. Furthermore, MTM peaks identified longer cycles at 12.12 units (~169.7 days) and 18.18 units (~254.5 days), which corresponded to the structure of IMF 4. Finally, the longest periodicity, a supra-annual oscillation of 44.44 units (~622.2 days), was clearly identified and confirmed by the trends in IMF 5. Hypothenemus eruditus (Sp. 2) exhibited a broad spectrum of periodicities, with activity primarily concentrated at short-term and sub-annual scales. According to MTM analysis, high-frequency rhythms were dominant, featuring significant spectral peaks at 2.05 units (~28.7 days), 2.5–2.53 units (~35–35.42 days), 3.60–3.64 units (~50.4–50.96 days), and 4.34–4.49 units (~60.8–62.9 days). These rapid fluctuations were well-supported by the structure of IMF 1 (Figure S2(c)). Within the intermediate range, activity was detected at 5.13–5.26 units (~71.82–73.64 days) and 6.25–6.61 units (~87.5–92.54 days) in IMF 2, while a distinct sub-annual rhythm at 8.51–8.89 units (~119.1–124.5 days) was identified and confirmed by IMF 3. Furthermore, a prominent long-term cycle was identified at 40.00–44.44 units (~560.0–622.2 days), with this supra-annual influence clearly reflected in the trends of IMF 5. Dryocoetiops moestus (Sp. 3) also exhibited an extensive suite of periodicities, spanning from high-frequency short-term scales to supra-annual intervals. According to MTM analysis, numerous high-frequency rhythms were identified at 2.01 units (~28.1 days) and across several clusters, including 2.18–2.19 units (~30.52–30.66 days), 2.31–2.48 units (~32.34–34.72 days), 2.72–2.76 units (~38.08–38.64 days), and 3.28–3.33 units (~45.92–46.62 days). These spectral peaks were well-supported by IMF 1 (Figure S2(d)). Intermediate cycles were detected at 4.49–4.65 units (~62.86–65.10 days) and 5.19–5.41 units (~72.66–75.74 days), corresponding to IMF 2, while sub-annual oscillations at 8.00–8.51 units (~112–119.14 days) and 10.81–11.43 units (~151.3–160.0 days) were captured by IMF 3. Notably, the species displayed robust activity at broader temporal scales, with significant power between 25.0 and 200.0 units (~350.0–2,800.0 days) reflected in both IMF 4 and IMF 5. These findings highlight a prominent supra-annual influence, specifically a 40.0–57.14 unit oscillation (~560.0–800.0 days) that was further corroborated by EEMD results. Temporal profiles among other taxa exhibited significant heterogeneity, reflecting diverse rhythmic hierarchies. Ambrosiodmus asperatus (Sp. 4) presented a broad-spectrum profile, where MTM analysis identified high-frequency activity characterized by initial cycles at 2.07–2.10 units (~28.98–29.4 days), 2.25–2.27 units (~31.5–31.78 days), and 2.59–2.84 units (~36.26–39.76 days). Within this same high-frequency range, a 3.70–3.77 unit (~51.8–52.78 days) sub-annual rhythm was isolated, with these rapid fluctuations well-captured by IMF 1 (Figure S2(e)). This complexity extended through multiple sub-annual pulses, including intermediate oscillations at 6.35–6.55 units (~88.9–91.7 days) and 7.69–8.10 units (~107.66–113.4 days), which corresponded to IMF 2 and IMF 3, respectively. The hierarchy culminated in significant supra-annual activity identified at 36.36 units (~509.04 days) and a major oscillation at 80.0 units (~1,120.0 days), both of which were confirmed by the low-frequency trends in IMF 5. In contrast, Eccoptopterus spinosus (Sp. 5) exhibited a narrower rhythmic range, limited primarily to short-term and sub-annual oscillations. High-frequency cycles were identified in IMF 1(Figure S2(f)), specifically at intervals of 2.47–2.48 units (~34.58–34.72 days), 2.88 units (~40.32 days), and 3.67–3.70 units (~51.38–51.8 days). MTM analysis revealed sub-annual signals characterized by cycles of 5.56–6.78 units (~77.8–94.92 days) and 7.14–7.54 units (~99.96–105.56 days), which were further supported by oscillations captured in IMFs 2 and 3 of the EEMD analysis. Temporal fluctuations in the trap capture numbers of this species appear to be driven predominantly by seasonal factors rather than supra-annual drivers, as evidenced by the lack of significant MTM-detected cycles within IMFs 4 and 5 extracted by EEMD. Hypothenemus areccae (Sp. 6) revealed a complex hierarchy of ten distinct cycles, ranging from short-term rhythms to supra-annual oscillations. Using MTM analysis, several high-frequency peaks were identified at 2.07–2.26 units (~28.98–31.64 days), 2.65 units, 2.86–2.89 units (~40.46 days), 3.22–3.31 units (~45.08–46.34 days), and 3.67 units (~51.38 days). These results were supported by IMF 1(Figure S2(g)), which captured these rapid fluctuations. Intermediate cycles were detected at 4.0–4.12 units (~56–57.68 days), 5.88 units (~82.32 days), and 7.14 units (~99.96 days), corresponding with the structure of IMF 2. Additionally, IMF 3 highlighted sub-annual rhythms at 10.0–10.53 units (~140–147.42 days). The long-term temporal architecture was further confirmed by IMF 4 at 13.79–14.81 units (~193.06–207.34 days) and a significant supra-annual rhythm in IMF 5 at 57.14–66.66 units (~800.0–933.24 days). Scolytoplatypus nanus (Sp. 7) exhibited the most intricate temporal profile, featuring fourteen identified cycle ranges across the entire hierarchy. According to MTM analysis, high-frequency activity was exceptionally dense, with several significant peaks at 2.05 (~28.7 days), 2.12–2.29 (~29.68–32.06 days), 2.21–2.22 (~30.94–31.08 days), 2.44–2.45 (~34.16–34.3 days), 2.61–2.63 (~36.54–36.82 days), 2.70–2.75 (~37.8–38.5 days), and 3.20–3.23 units (~44.8–45.22 days). These short-term scales were well-captured and supported by IMF 1(Figure S2(h)). An intermediate rhythm was also detected at 6.25–6.35 units (~87.5–88.9 days), corresponding to IMF 2. While mid-range activity was absent, the species displayed a complex suite of long-term oscillations. These included a distinct 40.0-unit (~560 days) rhythm and a broad, supra-annual dual-oscillation spanning 57.14–100.0 (~800–1,400 days) and 100.0–400.0 units (~1,400–5,600 days), both of which were confirmed by IMF 5. The presence of these diverse cycles suggests that S. nanus synchronizes its activity with both immediate monthly changes and multi-year environmental shifts. Xyleborinus andrewesi (Sp. 8) exhibited a diverse temporal profile characterized by activity across multiple scales. Using MTM analysis, high-frequency rhythms were identified starting at 2.05 units (~28.7 days), with several distinct peaks at 2.12–2.29 units (~29.68–32.06 days), 2.21–2.22 units (~30.94–31.08 days), 2.44–2.45 units (~34.16–34.3 days), 2.61–2.63 units (~36.54–36.82 days), 2.70–2.75 units (~37.8–38.5 days), and 3.20–3.23 units (~44.8–45.22 days). These spectral peaks were strongly supported by IMF 1(Figure S2(i)), which captured the fine-scale fluctuations. An intermediate sub-annual rhythm was detected at 6.25–6.35 units (~87.5–88.9 days), corresponding to the structure of IMF 2. Similar to other complex profiles, Sp. 8 displayed significant supra-annual activity, with prominent oscillations identified at 40.0 units (~560.0 days) and 57.14–100.0 units (~799.96–1,400.0 days). A lower-frequency cycle spanning 100.0–400.0 units (~1,400.0–5,600.0 days) was also observed, with both long-term patterns confirmed by the trends in IMF 5. Eidophelus sp . SA1 (Sp. 9) followed a predominantly sub-annual pattern, with rhythms primarily concentrated at short-term scales. According to MTM analysis, high-frequency activity was isolated at 2.41 units (~33.74 days), 2.61–2.63 units (~36.54–36.82 days), 2.7 units (~37.8 days), 2.72–3.08 units (~38.08–43.12 days), and 2.85–2.88 units (~39.9–40.32 days). These rapid oscillations were well-supported by the structure of IMF 1(Figure S2(j)). The species’ temporal hierarchy was further defined by intermediate pulses identified at 5.0–5.13 units (~70–71.82 days), with the longest periodicity for this taxon detected at 5.71–5.79 units (~79.94–81.06 days). These intermediate cycles were captured and confirmed by IMF 2. Conversely, Hypothenemus sp. SB06 (Sp. 10) exhibited the most constrained temporal profile among the surveyed taxa, with activity markedly limited to short-term cycles. MTM analysis identified fluctuations exclusively within high-frequency ranges, characterized by two primary spectral clusters: 2.19–2.63 units (~30.66–36.82 days) and 2.92–3.39 units (~40.88–47.46 days). These results were supported by IMF 1(Figure S2(k)), which captured the entirety of the species’ temporal dynamics. The absence of significant power in lower frequencies suggests that the population dynamics of Sp. 10 are driven almost entirely by immediate, high-frequency environmental triggers rather than seasonal or supra-annual cycles. Finally, Xylosandrus morigerus (Sp. 11) demonstrated a structured progression from high-frequency short-term cycles to expansive supra-annual background signals. MTM analysis identified multiple short-term periodicities at 2.05–2.06 units (~28.7–28.84 days), 2.29–2.5 (~32.06–35 days), 2.39–2.41 (~33.46–33.74 days), 2.76–2.79 (~38.64–39.06 days), and 4.12–4.21 units (~57.68–58.94 days). These initial fluctuations were well-supported by the structure of IMF 1 (Figure S2(l)). Intermediate rhythms were detected at 5.0–5.06 units (~70.0–70.84 days), corresponding to IMF 2, while a sub-annual oscillation was identified at 12.5–13.79 units (~175–193.06 days) and confirmed by IMF 4. The temporal profile culminated in significant long-term population cycles ranging from 40.0–44.44 units (~560.0–622.2 days) to expansive supra-annual signals reaching 80.0–400.0 units (~1,120.0–5,600.0 days), both of which were reflected in the low-frequency trends of IMF 5. Table 2. Summary of significant population cycles for Total Trap Capture (TTC) and eleven bark and ambrosia beetles species. Cycles are expressed in biweekly units (1 unit = 14 days). ‘Dual Significant’ denotes peaks meeting both Red Noise (AR(1)>95%) and Harmonic (CL>90%) Confidence thresholds. Total Trap Capture (TTC) 2.05-2.06 2.19-2.21 2.39-2.54 2.47-2.5 2.47-2.5 2.70-2.73 2.89-2.94 3.22-3.31 4.49 5.19-5.33 8.33 22.22 133.33 [Sp. 1] Eidophelus ( Scolytogenes ) sp. SA1 2.52-2.64 3.08-3.13 3.54-3.60 4.30-4.35 5.06-5.19 7.0-7.17 12.12 18.18 44.44 133.33 [Sp. 2] Hypothenemus eruditus cx Westwood 2.05 2.5-2.53 3.60-3.64 4.34-4.49 5.13-5.26 6.61-6.25 8.51-8.89 40.0-44.44 [Sp. 3] Dryocoetiops moestus (Blandford) 2.01 2.18-2.19 2.31 2.35-2.44 2.45-2.48 2.72-2.76 3.28-3.33 4.49-4.65 5.19-5.41 8-8.51 10.81-11.43 25-200 40-57.14 40-57.14 [Sp. 4] Ambrosiodmus asperatus (Blandford) 2.07-2.10 2.07-2.09 2.07 2.25-2.27 2.59-2.84 2.70-2.74 3.70-3.77 6.35-6.55 7.69-8 10 36.36 80 400 [Sp. 5] Eccoptopterus spinosus (Olivier) 2.88 2.47-2.48 3.67-3.70 6.56-6.78 7.14-7.54 [Sp. 6] Hypothenemus areccae cx (Hornung) 2.07-2.26 2.65 2.86-2.89 3.22-3.31 3.67 4.0-4.12 5.88 7.14 10-10.53 13.79-14.81 57.14-66.66 [Sp. 7] Scolytoplatypus nanus Schedl 2.16-2.19 2.31-2.34 2.89-2.96 3.47-3.54 3.92 4.44 5.63-5.88 8.69-8.89 10.53-11.76 12.5-13.79 12.90-14.28 15.38-16.67 20-21.05 50-80 [Sp. 8] Xyleborinus andrewesi (Blandford) 2.05 2.12-2.29 2.21-2.22 2.44-2.45 2.61-2.63 2.70-2.75 3.2-3.23 6.25-6.35 40 57.14-100 100 100-400 [Sp. 9] Eidophelus sp. SA1 2.41 2.61-2.63 2.7 2.72-3.08 2.85-2.88 5-5.13 5.71-5.79 [Sp. 10] Hypothenemus sp. SB06 2.19-2.63 2.92-3.39 [Sp. 11] Xylosandrus morigerus (Eichhoff) 2.05-2.06 2.29-2.5 2.39-2.41 2.76-2.79 4.12-4.21 5-5.06 12.5-13.79 40-44.44 80-400 Causal Mechanism via Path Analysis The path analysis revealed distinct temporal patterns regarding the influence of environmental variables on TTC across both biweekly and monthly scales (Tables S9–S10, Figure 1). The three model fit indices (CFI = 1.000, TLI = 1.000, RMSEA = 0.000) indicated a saturated model structure where the degrees of freedom equal zero (df = 0), resulting in a perfect mathematical fit to the observed covariance matrix. At the biweekly scale, statistically significant models were primarily identified at extended temporal intervals, specifically between Lag 16 (~224 days) and Lag 22 (~308 days). The most robust biweekly model occurred at Lag 17 (~238 days; R 2 = 0.180, p = 0.006), where temperature ( β = 0.579, p = 0.002), humidity ( β = 0.907, p < 0.001), and rainfall ( β = ─0.430, p = 0.012) were all identified as significant predictors of beetle abundance. Throughout this interval, temperature and humidity consistently exhibited positive associations with TTC, while rainfall demonstrated a sustained negative influence. However, at Lag 21 (~294 days) and Lag 22 (~308 days), the predictive power of temperature diminished and lost statistical significance (p > 0.05). At these extended lags, humidity and rainfall remained the primary environmental drivers governing population dynamics. In comparison, the monthly scale models demonstrated higher predictive power, with R 2 -square values ranging from 0.290 to 0.530. The best-fitting model was found at Lag 20 (~600 days; R 2 = 0.530, p = 0.005), where rainfall ( β = 0.632, p = 0.009) emerged as the primary environmental driver of beetle abundance. The direct effect of temperature was limited to shorter time intervals, remaining statistically significant only at Lag 3 (~90 days; p = 0.001) and Lag 4 (~120 days; p = 0.019). At longer lags, temperature became statistically non-significant. In contrast, humidity remained a consistent positive predictor at mid-range intervals, such as Lag 8 ( β = 1.118, p < 0.001) and Lag 13 ( β = 0.817, p = 0.008). Meanwhile, rainfall showed a significant negative relationship at Lag 3 (p = 0.008) and Lag 18 (p = 0.003), before transitioning into a strong positive predictor at Lag 20 (p = 0.009). A comparison of the two temporal resolutions indicates that the monthly scale models provided substantially higher predictive power, with R 2 values ranging from 0.290 to 0.530, compared to 0.128 to 0.180 for the biweekly models. Despite this difference in explanatory power, both scales identified a consistent biological window at extended intervals, specifically between the 8 and 10-month lags (Biweekly Lags 16–22 and Monthly Lags 8–10). Throughout this period, the environmental predictors followed a distinct process: humidity and rainfall were the most stable drivers, with humidity maintaining a significant positive association and rainfall showing a sustained negative influence on beetle abundance. In contrast, the effect of temperature was less stable; while it acted as a significant positive predictor in the early biweekly lags (Lags 16–18), its influence became statistically non-significant at the longest biweekly intervals (Lags 21–22) and was largely absent from the extended monthly models. These results suggest that while temperature has a short-term or scale-dependent impact, the long-term environmental signal is most effectively captured at the monthly scale through the combined influence of high humidity and shifting rainfall dynamics. Figure 1: Path analysis diagrams illustrating the causal relationships between environmental variables and Total Trap Capture (TTC) at different temporal scales. Panels (a) and (b) show Monthly Lags 3 and 8, while (c) and (d) represent Biweekly Lags 3 and 16. Green solid lines indicate positive standardized path coefficients ( β ), while red dashed lines indicate negative relationships. Line thickness is proportional to the magnitude of the effect. Variables include Temperature (Tmp), Humidity (Hmd), and Rainfall (Rnf). Values adjacent to variable boxes represent explained variance ( R 2 ), and values on the paths represent standardized coefficients Discussion Hierarchical Climatic Forcing in the Tropics Our spectral analysis reveals that the tropical climate in southern Sabah is not a stochastic background of white noise, but a structured hierarchy of coupled hydroclimatic oscillations (Chong, et al. 2021). Contrary to the traditional classification of this region as aseasonal or ever-wet, prior work in Bornean rainforests has highlighted the presence of distinct dry periods and hydroclimatic variability (Malhi, et al. 2011, Walsh 1996, Walsh and Newbery 1999). We detected distinct cycles ranging from intraseasonal (35 days) to interannual (>1,100 days) scales. These lower-frequency interannual signals likely reflect the influence of the El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD), which are known to drive multi-year rainfall anomalies and prolonged dry spells across the Indo-Pacific (Saji, et al. 1999). The dominant 35-day cycles observed in temperature and rainfall (Table 1) strongly align with the Madden-Julian Oscillation (MJO), the primary mode of intraseasonal variability in the tropical atmosphere. The MJO is characterized by a large-scale coupling of atmospheric circulation and deep convection moving eastward (Madden and Julian 1994, Zhang 2005, Zhang 2013). Our detection of synchronous peaks in temperature and humidity confirms this coupling, a relationship statistically validated by cross-correlation function (CCF) analysis at both monthly (Figure S3) and biweekly (Figure S4) resolutions. These CCF results reveal a near-zero lag between peak precipitation and temperature minima, confirming that the deep convective phase of the MJO brings heavy precipitation and cloud cover, which simultaneously reduces solar insolation and lowers surface temperatures (Hendon and Glick 1997). This creates a hydroclimatic cycle where wetting and cooling events occur in unison, rather than acting as isolated variables, establishing the fundamental high-frequency climatic tempo of the forest (Kiladis, et al. 2005, Maloney and Sobel 2004). Intraseasonal Generation Cycles Spectral analysis of the Total Trap Capture (TTC) and dominant species revealed several high-frequency periodicities. We emphasize that these 4–5 week rhythms represent ”generation cycles (Bjørnstad, et al. 2004, Nakamura, et al. 2004)”, the intrinsic developmental timing of the insects rather than a direct result of climatic forcing. These rhythms indicate that the population tempo is governed primarily by the metabolic developmental duration of the insects rather than external seasonal cues. This aligns with observations (Sanguansub, et al. 2020), which identified a spectral peak matching the 30–40 day developmental timeframe of specific ambrosia beetles. In the absence of a thermal reset like winter diapause, the Bornean environment allows for free-running, continuous multivoltine cycles characterized by the superposition of multiple overlapping cohorts (D.L. 1986, Nakagawa, et al. 2000). Our findings underscore that even in such aseasonal environments, generation cycles originating from the length of a single generation can still manifest as significant spectral signals. These rhythms indicate that the population tempo is governed primarily by the metabolic developmental duration of the insects. The developmental durations can be vary based on whether species is phyloephagous (e.g., Eidophelus spp.) or xylomycetophagous (e.g., Ambrosiodmus spp.), the tropical thermal regime in Sabah appears to converge these diverse life histories into a coherent 4–6 week (30–40 day) spectral peak. Role of Madden-Julian Oscillation Our MTM spectral analysis revealed a pervasive high-frequency rhythm across the beetle community. A prominent ~35-day cycle (2.47–2.5 biweekly unit) dual-significant cycle identified in the TTC matched the MJO established 30–60 day cycles in temperature anomalies (Table 2), suggesting that the MJO serves as the primary environmental pacemaker (Huang, et al. 2024). We propose that episodic westerly wind events associated with the MJO create a resource cycle consisting of broken branches and windthrow, which effectively resets the colonization clocks of multiple species simultaneously. However, the influence of the MJO is characterized by a distinct duality, while it facilitates resource availability, its active convective phase also imposes a significant, albeit temporary, constraint on beetle activity. This suppression is driven by two concurrent mechanisms. First, the mechanical inhibition hypothesis (Wolda 1988) posits that intense tropical precipitation acts as a physical barrier, restricting flight windows and reducing trap captures regardless of actual population density. Second, we propose a host-stress alleviation effect operating on rapid physiological timescales. Wood-boring beetles utilize stress-induced plant volatiles, specifically ethanol, to locate weakened host trees (Kelsey and Gladwin 2003, Wood 1982). Even in perennially wet tropical forests, brief dry spells can induce transient physiological water deficits. However, we argue that significant hydroclimatic pulses can rapidly reverse this vulnerability. The recovery of host defences occurs across two distinct temporal stages. The first stage of this recovery occurs within hours to a single day, as root water uptake restores cell turgor pressure (Warren, et al. 2011). This reactivates constitutive defences, such as resin and gum exudation, which rely on hydrostatic pressure to mechanically repel boring beetles (Lewinsohn, et al. 1993). In the second stage, occurring over several days, the restoration of aerobic metabolism halts the production of ethanol and other anaerobic byproducts (Kimmerer and Kozlowski 1982). This physiological shift effectively ”masks” the host landscape, as the cessation of distress kairomones prevents beetles from orienting toward previously weakened resources. By integrating these processes, it becomes evident that while the 35-day generation cycle establishes the fundamental population rhythm, MJO-driven hydro-climatic pulses act as a decisive ”gatekeeper.” Furthermore, we propose that episodic westerly wind events associated with the MJO generate a structured resource cycle. While the genus Eidophelus (spp. 1 and 9) is typically oligophagous (Marchioro, Vallotto, et al. 2024), the majority of the species identified in this study are characterized by broad polyphagy. In particular, H. eruditus (sp. 2) and X. andrewesi (sp. 8) are highly polyphagous (Marchioro, Vallotto, Ruzzier, Besana, Rossini, Ortis, Faccoli and Martinez-Sañudo 2024, Ruzzier, et al. 2023). This ecological plasticity is further augmented by the capacity of certain species to exploit small-diameter substrates, which are highly susceptible to wind-induced fracture. For instance, D. moestus (sp. 3) colonizes twigs and small branches (Beaver, et al. 2019), while Hypothenemus spp. (spp. 2, 6, and 10), along with the highly polyphagous E. spinosus (sp. 5), and X. morigerus (sp. 11), utilize a much broader niche,exploiting not only phloem but also the pith of twigs, seeds, and fruits (Vega, et al. 2015). Because these smaller distal components are the most readily fractured and shed during high-wind events, species capable of utilizing them can capitalize on a more immediate and abundant resource pulse compared to those restricted to large-diameter trunks. These broad dietary and structural niches enable the beetle community to exploit MJO-driven disturbances across a vast range of host families simultaneously. By transcending the reproductive phenology or ephemeral availability of any single host species, the community can respond to MJO-induced disturbances as a synchronized functional unit, effectively translating episodic climatic energy into concentrated population surges. The ~3-Month Lag Our path analysis revealed that climatic influence on beetle populations operates through a structured hierarchy of temporal scales rather than a single immediate trigger (Boggs and Inouye 2012, Kéfi, et al. 2014). We interpret these significant time lags as ecological memory, defined as the capacity of past environmental states to influence present-day ecological responses (Dakos and Bascompte 2014, Ogle, Barber, Barron-Gafford, Bentley, Young, Huxman, Loik and Tissue 2015). While mechanical inhibition acts as an immediate filter at the biweekly scale, the strong explanatory power of our models at Monthly Lag 3 ( R 2 =0.421) and Monthly Lag 8 ( R 2 = 0.412) indicates that MJO-driven hydroclimatic pulses initiate specific biological processes that require fixed durations to mature. The first tier of this hierarchy is the stress-response pathway (~3-Month Lag), a response best exemplified by the super-generalist Hypothenemus eruditus (Marchioro, Vallotto, Ruzzier, Besana, Rossini, Ortis, Faccoli and Martinez-Sañudo 2024). Unlike specialist species constrained by the phenology of specific hosts, H. eruditus colonizes an extreme diversity of tissues across hundreds of plant families, provided they share a specific physiological state. This species is highly attracted to ethanol, a volatile byproduct of anaerobic respiration in stressed or dying tissues (Biedermann and Vega 2020, Kelsey 1994, Wood 1982). The ~8-Month Lag and Legacy Effect The strongest explanatory power in our model emerged at a Monthly Lag 8 ( R ² = 0.412, p =0.002). Our path analysis also identified a significant negative coefficient for rainfall ( β =0.494, p = 0.005) at biweekly Lag 16 (Figure 1). Although Lag 16 represents an approximately 8-month window (224 days), indicating that rainfall patterns exert a consistent modulating pressure on realized population abundance over sub-annual scales. This approximately 8-month window represents a legacy effect, where favourable moisture conditions specifically high humidity ( β= 1.118 , p<0 .001 ) experienced nearly a year prior facilitate host tree health or fungal proliferation, ultimately driving beetle abundance in future generations. We propose three non-mutually exclusive pathways to explain the observed ~8-month (approx. 250-day) sub-annual periodicity: The Direct Demographic Pathway ( Hypothenemus areccae and Scolytoplatypus nanus ): For species such as H. areccae and S. nanus , the observed 193–233 day (6.4–7.8 months) signal likely reflects multivoltine accumulation. While the specific biology of S. nanus (Sp. 7) remains largely uncharacterized, and its current host record is limited to the genus Vernonia (Asteraceae) (Kalshoven 1959), the prevailing polyphagy within this genus (Marchioro, Besana, et al. 2024) suggests that S. nanus is also a generalist. Similarly, H. areccae (Sp. 6) is known to exploit a broad range of host families and diverse plant organs, including phloem (Beaver and Gebhardt 2006, Marchioro, Besana, Rossini, Vallotto, Ruzzier, Ortis, Martinez-Sañudo and Faccoli 2024). These expansive dietary and structural niches likely facilitate a sequence of rapid, overlapping generations. In the absence of diapause-driven resets typical of seasonal environments (Danks 2007), these populations can build up continuously over approximately ~7 months, eventually culminating in the detectable population peaks identified in our spectral analysis. The Developmental Legacy Pathway ( Ambrosiodmus asperatus ): Spectral analysis revealed significant population oscillations at 12.9–14.2 biweekly units (~250 days, or ~8 months) for Ambrosiodmus asperatus (Table 2). Among bark and ambrosia beetles, A. asperatus maintains a symbiotic relationship with the white-rot fungus Flavodon spp. (Fukasawa 2021, Li, Simmons, Bateman, Short, Kasson, Rabaglia and Hulcr 2015, Peris, et al. 2021), which degrades lignocellulose through a slow successional process (Castaño, et al. 2022, Riley, et al. 2014). The 8-month lag is dictated by specialized fungal agriculture. Unlike phloeophagous species (e.g. Eidophelus spp.), which may utilize phloem directly, xylomycetophagous species depend on the delignification process. Because white-rot enzymatic activity is strictly moisture-dependent (Schmidt 2006), the ~8-month lag represents the obligate incubation period required for the fungus to transform fresh timber into the delignified substrate necessary for brood development. Sub-annual Hydroclimatic Rhythms: This sub-annual periodicity also reflects a dominant ~8-month hydroclimatic rhythm in the tropical environment, which modulates habitat availability and resource quality without imposing a strict annual reset. Long-term Resonance Beyond annual scales, we detected significant supra-annual periodicities in the biological data, most notably at the ~17-month scale for A. asperatus (Table 2). These results indicate that tropical insect populations resonate with large-scale climatic modes. In the absence of a rigid one-year solar pacemaker typical of seasonal forests, the alignment between these supra-annual climatic periodicities and observed population cycles suggests a coupling with oscillations such as the ENSO and the IOD. ENSO, in particular, is recognized for inducing severe and extraordinarily prolonged drought stress within non-seasonal tropical ecosystems (Walsh and Newbery 1999). IOD also induce severe prolonged drought stress in southeast Asia, especially in non-seasonal tropics (Aldrian 2003, Saji, Goswami, Vinayachandran and Yamagata 1999). Such physiological stress enhances the susceptibility of host trees, creating a resource landscape that is highly preferable for secondary bark and ambrosia beetles (Anderegg, et al. 2015, Raffa, et al. 2008). Ultimately, we propose a Resonance Hypothesis: intrinsic generation cycles (~35 days) establish the fundamental biological rhythm, which is phase-locked by intraseasonal climatic pulses (MJO). These coupled dynamics are further modulated by sub-annual biological processes, such as the physiological recovery of host plants, successional insect demography, and fungal incubation. This entire hierarchical structure rides atop a supra-annual wave driven by large-scale oscillations, specifically ENSO and the IOD, which dictate long-term environmental carrying capacity. Such a multi-layered architecture explains why tropical insect populations may appear stochastic in short-term observations; their apparent randomness is, in fact, an interference pattern created by these nested, multi-scale temporal drivers. Conclusions Our study contests the historical paradigm that tropical insect dynamics are purely stochastic or seasonally stable. By integrating MTM spectral analysis with lagged path analysis, we demonstrate that beetle populations in Sabah are structured by a multi-frequency hierarchy of climatic and biological drivers. We propose the Resonance Hypothesis, which posits that population dynamics emerge from the interference of three distinct temporal scales: First, intrinsic generation cycles (~35 days) establish the internal tempo of the community. Second, intraseasonal climatic pulses (MJO) act as a gatekeeper, periodically providing resources through windthrow while simultaneously modulating flight activity through rainfall inhibition. Third, biological legacy effects modulate population peaks through two distinct sub-annual tiers. The first tier (~3-month lag) represents a stress-response pathway where hydroclimatic pulses initiate physiological shifts in host trees. The second tier (~8-month lag) constitutes an ”ecological memory” driven by two mechanisms: the continuous accumulation of overlapping multivoltine generations, and the obligate incubation period required for fungal symbionts to degrade wood into suitable larval substrates. Ultimately, these cycles are nested within supra-annual climatic modes, specifically ENSO and the IOD, which influence long-term population baselines. By inducing prolonged drought stress and altering tree mortality rates, these large-scale oscillations dictate the availability of breeding substrates, effectively driving the long-term resonance of the community. Our findings suggest that the apparent stochasticity of tropical insects is an interference pattern created by these overlapping rhythms. Recognizing these lagged, non-linear relationships is critical for predicting how tropical forest ecosystems will respond to shifting climatic modes in a changing global environment. Supporting information The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: Intrinsic Mode Functions (IMFs) of daily meteorological time series; Figure S2: Intrinsic Mode Functions (IMFs) of biweekly insect time series data; Figure S3: Cross-correlation functions (CCF) of monthly weather variables derived from the detrended monthly time-series dataset; Figure S4: Cross-correlation functions (CCF) of biweekly weather variables derived from the detrended biweekly time-series dataset; Table S1: Primary data for the total trap capture (TTC); Table S2: Adjusted biweekly total trap capture (TTC) and eleven major species; Table S3: Adjusted monthly total trap capture (TTC); Table S4: Biweekly weather data calculated from daily records; Table S5: Detrended biweekly total trap capture (TTC) time series; Table S6: Detrended monthly total trap capture (TTC) time series; Table S7: Detrended biweekly weather data; Table S8: Detrended monthly weather data; Table S9: Model fit indices for the path analysis of lagged relationships between hydroclimatic drivers and total trap capture (TTC) at a biweekly resolution.; Table S10: Model fit indices for the path analysis of lagged relationships between hydroclimatic drivers and total trap capture (TTC) at a monthly resolution. References Aldrian, E. S., R.D. 2003. Identification of three dominant rainfall regions within Indonesia and their relationship to sea surface temperature. - International Journal of Climatology 23 : 1435-1452.Anderegg, W. R., et al. 2015. Tree mortality from drought, insects, and their interactions in a changing climate. - New Phytologist 208 : 674-683.Anderegg, W. R. L., et al. 2012. The roles of hydraulic and carbon stress in a widespread climate-induced forest die-off. - Proceedings of the National Academy of Sciences 109 : 233-237.Beaver, R. A. and Gebhardt, H. 2006. A review of the oriental species of Scolytoplatypus Schaufuss (Coleoptera, Curculionidae, Scolytinae). - Deutsche Entomologische Zeitschrift 53: 155-178.Beaver, R. A., et al. 2019. A review of the genus Dryocoetiops Schedl, with new species, new synonymy and a key to species (Coleoptera: Curculionidae: Scolytinae). - Zootaxa 4712: 236–250.Biedermann, P. H. W. and Vega, F. E. 2020. Ecology and evolution of insect-fungus mutualisms. - Annual Review of Entomology 65 : 431-455.Bjørnstad, O. N., et al. 2004. Trends and cohort resonant effects in age-structured populations. - Journal of Animal Ecology 73 : 1157-1167.Boggs, C. L. and Inouye, D. W. 2012. A single climate driver has direct and indirect effects on insect population dynamics. - Ecology Letters 15 : 502-508.Brown, J. H. and Maurer, B. A. 1989. Macroecology: The division of food and space among species on continents. - Science 243 : 1145-1150.Castaño, J. D., et al. 2022. Metabolomics highlights different life history strategies of white and brown rot wood-degrading fungi. - mSphere 7 : e0054522.Chang, C.-P., et al. 2005. Synoptic disturbances over the equatorial South China Sea and western Maritime Continent during boreal winter. - Monthly Weather Review 133 : 489-503.Cheang, B. K. 1977. Synoptic features and structures of some equatorial vortices over the South China Sea in the Malaysian region during the winter monsoon, December 1973. - Pure and Applied Geophysics 115 : 1303-1333.Chong, K. L., et al. 2021. Spatiotemporal variability analysis of standardized precipitation indexed droughts using wavelet transform. - Journal of Hydrology 605: 127299.D.L., D. 1986. Dormancy in Tropical Insects. - Annual Review of Entomology. 31:239-264 31: 239-64.Dakos, V. and Bascompte, J. 2014. Critical slowing down as early warning for the onset of collapse in mutualistic communities. - Proceedings of the National Academy of Sciences 111 : 17546-51.Danks, H. V. 2007. The elements of seasonal adaptations in insects. - The Canadian Entomologist 139 : 1-44.Epskamp, S. 2022. semPlot: Path Diagrams and Visual Analysis of Various SEM Packages’ Output.Estay, S. A., et al. 2009. Predicting insect pest status under climate change scenarios: Combining experimental data and population dynamics modelling. - Journal of Applied Entomology 133 : 491-499.Fukasawa, Y. 2021. Ecological impacts of fungal wood decay types: A review of current knowledge and future research directions. - Ecological Research 36 :Gunggot, E., et al. 2025. List of Scolytinae and Platypodinae (Coleoptera: Curculionidae) captured by ethanol-baited traps in a tropical rainforest in southern Sabah, Malaysia. - Miscellaneous Information of The University of Tokyo Forests 71 : 83-94.Helske, J. and Luukko, P. J. J. 2023. Rlibeemd: Ensemble empirical mode decomposition (EEMD) and its complete variant (CEEMDAN). - In.Hendon, H. H. and Glick, J. 1997. Intraseasonal air–sea interaction in the tropical Indian and Pacific Oceans. - Journal of Climate 10 : 647-661.Huang, W. R., et al. 2024. The impact of the Madden-Julian oscillation on spring and autumn afternoon diurnal convection in Sri Lanka. - npj Climate and Atmospheric Science 7: 42.Hughes, T. P., et al. 2019. Ecological memory modifies the cumulative impact of recurrent climate extremes. - Nature Climate Change 9 : 40-43.Hulcr, J. and Dunn, R. R. 2011. The sudden emergence of pathogenicity in insect–fungus symbioses threatens naive forest ecosystems. - Proceedings of the Royal Society B: Biological Sciences 278 : 2866-2873.Ives, A. R. 1995. Predicting the response of populations to environmental change. - Ecology 76 : 926-941.Jactel, H., et al. 2012. Drought effects on damage by forest insects and pathogens: A meta-analysis. - Global Change Biology 18 : 267-276.Kalshoven, L. G. E. 1959. Studies on the biology of Indonesian Scolytoidea 4. Data on the habits of Scolytidae. Second part. - Tijdschrift voor Entomologie 102:Kéfi, S., et al. 2014. Early warning signals of ecological transitions: methods for spatial patterns. - PLoS One 9: e92097.Kelsey, R. G. 1994. Ethanol and ambrosia beetles in Douglas fir logs with and without branches. - Journal of Chemical Ecology 20 : 3307-19.Kelsey, R. G. and Gladwin, J. 2003. Ethanol in ponderosa pine as an indicator of physiological injury from fire and its relationship to secondary beetles. - Canadian Journal of Forest Research 33 : 870-884.Kiladis, G. N., et al. 2005. Zonal and vertical structure of the Madden–Julian Oscillation. - Journal of the Atmospheric Sciences 62 : 2790-2809.Kimmerer, T. W. and Kozlowski, T. T. 1982. Ethylene, ethane, acetaldehyde, and ethanol production by plants under stress. - Plant Physiology 69 : 840-847.Kirkendall, L. R., et al. 2015. Chapter 3 - Evolution and diversity of bark and ambrosia beetles. - In: F. E. Vega and R. W. Hofstetter (eds), Bark beetles: Biology and ecology of native and invasive species. Academic Press, pp. 85-156.Kishimoto-Yamada, K. and Itioka, T. 2015. How much have we learned about seasonality in tropical insect abundance since Wolda (1988)? - Entomological Science 18 : 407-419.Lawton, J. H. 1999. Are there general laws in ecology? - Oikos 84 : 177-192.Lewinsohn, E., et al. 1993. Oleoresinosis in grand fir ( Abies grandis ) saplings and mature trees: Modulation of this wound response by light and water stresses. - Plant Physiology 101 : 1021-1028.Li, Y., et al. 2015. New fungus-insect symbiosis: Culturing, molecular, and histological methods determine saprophytic Polyporales mutualists of Ambrosiodmus ambrosia beetles. - PLOS ONE 10 : e0137689.Madden, R. A. and Julian, P. R. 1994. Observations of the 40–50-Day tropical oscillation—A review. - Monthly Weather Review 122 : 814-837.Malhi, Y., et al. 2011. The allocation of ecosystem net primary productivity in tropical forests. - Philosophical Transactions of the Royal Society B: Biological Sciences 366 : 3225-3245.Maloney, E. D. and Sobel, A. H. 2004. Surface fluxes and ocean coupling in the tropical intraseasonal oscillation. - Journal of Climate 17 : 4368-4386.Mann, M. E. and Lees, J. M. 1996. Robust estimation of background noise and signal detection in climatic time series. - Climatic Change 33 : 409-445.Marchioro, M., et al. 2024. The first host plant dataset of Curculionidae Scolytinae of the world: miscellaneous tribes (Part 2). - Scientific Data 11: 1217.Marchioro, M., et al. 2024. The first host plant dataset of Curculionidae Scolytinae of the world: miscellaneous Tribes. - Scientific Data 11: 120.Meyers, S. R. 2014. astrochron: An R Package for Astrochronology. - In.Nakagawa, M., et al. 2000. Impact of severe drought associated with the 1997–1998 El Niño in a tropical forest in Sarawak. - Journal of Tropical Ecology 16 : 355 - 367.Nakamura, K., et al. 2004. Generation cycles in Indonesian lady beetle populations may occur as a result of cannibalism. - Proceedings of the Royal Society of London. Series B: Biological Sciences 271 : S501 - S504.Ng, C. K.-C., et al. 2020. Precipitation trend and heterogeneity of Sabah, North Borneo. 28 : 19-43.Ogle, K., et al. 2015. Quantifying ecological memory in plant and ecosystem processes. - Ecology Letters 18 : 221-235.Peng, Y., et al. 2022. Succession of ambrosia beetles colonizing the logs of fallen alder and birch trees. - Insects 13 : 223.Peris, D., et al. 2021. Origin and evolution of fungus farming in wood-boring Coleoptera—A palaeontological perspective. - Biological Reviews 96 :Qian, J.-H. 2008. Why precipitation is mostly concentrated over islands in the Maritime Continent. - Journal of the Atmospheric Sciences 65 : 1428-1441.R Core Team 2024. R: A Language and Environment for Statistical Computing. - In: R Foundation for Statistical Computing.Raffa, K. F., et al. 2008. Cross-scale drivers of natural disturbances prone to anthropogenic amplification: The dynamics of bark beetle eruptions. - BioScience 58 : 501–517.Riley, R., et al. 2014. Extensive sampling of basidiomycete genomes demonstrates inadequacy of the white-rot/brown-rot paradigm for wood decay fungi. - Proceedings of the National Academy of Sciences 111 : 9923-8.Ropelewski, C. F. and Halpert, M. S. 1987. Global and regional scale precipitation patterns associated with the El Niño/Southern Oscillation. - Monthly Weather Review 115 : 1606-1626.Rosseel, Y. 2012. lavaan: An R Package for Structural Equation Modeling. - Journal of Statistical Software 48 : 1 - 36.Ruzzier, E., et al. 2023. The first full host plant dataset of Curculionidae Scolytinae of the world: tribe Xyleborini LeConte, 1876. - Scientific Data 10: 166.Saji, N. H., et al. 1999. A dipole mode in the tropical Indian Ocean. - Nature 401: 360-363.Sanguansub, S., et al. 2020. Influence of seasonality and climate on captures of wood-boring Coleoptera (Bostrichidae and Curculionidae (Scolytinae and Platypodinae)) using ethanol-baited traps in a seasonal tropical forest of northern Thailand. - Journal of Forest Research 25 : 223-231.Saunders, D. S. 2012. Insect photoperiodism: seeing the light. - Physiological Entomology 37:Schmidt, O. 2006. Wood and tree fungi: Biology, damage, protection, and use . - Springer-Verlag.Tauber, M. J., et al. 1986. Seasonal adaptations of insects . - Oxford University Press.Ueno, H., et al. 2021. Six-year population dynamics of seven passalid species in a humid tropical rainforest in Borneo. - Entomological Science 24 : 399-409.Vega, F. E., et al. 2015. Chapter 11 - The Genus Hypothenemus , with Emphasis on H. hampei , the Coffee Berry Borer. - In: F. E. Vega and R. W. Hofstetter (eds), Bark Beetles. Academic Press, pp. 427-494.Walsh, R. P. D. 1996. Drought frequency changes in Sabah and adjacent parts of northern Borneo since the late nineteenth century and possible implications for tropical rain forest dynamics. - Journal of Tropical Ecology 12 : 385-407.Walsh, R. P. D. and Newbery, D. M. 1999. The ecoclimatology of Danum, Sabah, in the context of the world’s rainforest regions, with particular reference to dry periods and their impact. - Philosophical Transactions of the Royal Society B: Biological Sciences 354 : 1869-1883.Warren, J. M., et al. 2011. Ecohydrologic impact of reduced stomatal conductance in forests exposed to elevated CO 2 . - Ecohydrology 4 : 196-210.Wolda, H. 1978. Seasonal fluctuations in rainfall, food and abundance of tropical insects. - Journal of Animal Ecology 47 : 369-381.Wolda, H. 1988. Insect Seasonality: Why? - Annual Review of Ecology, Evolution, and Systematics 19 : 1-18.Wood, S. L. 1982. The bark and ambrosia beetles of North and Central America (Coleoptera: Scolytidae), a taxonomic monograph. - Great Basin Naturalist Memoirs 6 : 1-1359.Zhang, C. 2005. Madden-Julian Oscillation. - Reviews of Geophysics 43 :Zhang, C. 2013. Madden–Julian Oscillation: Bridging weather and climate. - Bulletin of the American Meteorological Society 94 : 1849-1870. Information & Authors Information Version history V1 Version 1 18 March 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords generation cycle host stress hydroclimatic cycle insect population cycle legacy effect resonance hypothesis Authors Affiliations Evahtira Gunggot Universiti Malaysia Sabah View all articles by this author Roger Beaver Freelance View all articles by this author Maria Lardizabal Universiti Malaysia Sabah View all articles by this author Jonathan Lucas Universiti Malaysia Sabah View all articles by this author Sandra George Universiti Malaysia Sabah View all articles by this author Anastasia Rasiah Universiti Malaysia Sabah View all articles by this author Wilson Wong Universiti Malaysia Sabah View all articles by this author Naoto Kamata 0000-0002-8818-6991 [email protected] The University of Tokyo Graduate School of Agricultural and Life Sciences Faculty of Agriculture View all articles by this author Metrics & Citations Metrics Article Usage 152 views 57 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Evahtira Gunggot, Roger Beaver, Maria Lardizabal, et al. Hierarchical climatic forcing and population rhythms of bark and ambrosia beetles in tropical rainforests of Northern Borneo. Authorea . 18 March 2026. DOI: https://doi.org/10.22541/au.177383746.63722976/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.177383746.63722976/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fde6a8a0a48dfa9',t:'MTc3OTE0NDU3Nw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();