Multi-Sensor Geospatial Modelling to Address Complex Mangrove Dieback: Misattribution of Chemical Stressors Versus Physical Impact

Multi-Sensor Geospatial Modelling to Address Complex Mangrove Dieback: Misattribution of Chemical Stressors Versus Physical Impact | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Multi-Sensor Geospatial Modelling to Address Complex Mangrove Dieback: Misattribution of Chemical Stressors Versus Physical Impact Jade Farrugia This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8929984/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Accurately identifying the causes of mangrove dieback is essential for global coastal management, yet visually similar dieback events can result in incorrect attribution of causes. A significant mangrove dieback in Boambee Creek Estuary (2021) was initially linked to chemical contamination from nearby Coffs Harbour Airport. This study uses a forensic geospatial reconstruction to test the validity of this toxicity hypothesis against a physical-hydrological alternative. Going beyond the limitations of site-specific field sampling ( N = 2), we employed multi-sensor satellite telemetry (Sentinel-2), LiDAR-based geomorphic modelling, and ERA5 climatological reanalysis to assess the ecosystem at the landscape level ( N = 1,306). The investigation pinpoints a statistically extreme hailstorm on 20 October 2021 as the trigger event, representing a > 1-in-100-year anomaly ( Z = 4.10σ). Time-series diagnostics confirm an immediate structural collapse ( p < 0.001) coinciding with the storm, ruling out the signature of gradual chemical aging. Mortality followed a Death Curve, where the likelihood of death neared 100% at elevations below 1.5m AHD ( p < 10 − 8 ) within stagnant topographic basins (< 2° slope). Hydrological routing shows that the main runoff from the airport flows directly into the remaining forest, creating a Runoff Paradox that statistically discredits the chemical vector hypothesis. We conclude that the dieback resulted from a Hydrological Trap; sudden physical defoliation stopped canopy transpiration, causing rapid soil anoxia and root drowning in geomorphically unstable basins. Future management should focus on restoring hydrological connectivity rather than chemical remediation. This study highlights the vital need to incorporate landscape-scale multi-sensor remote sensing (Optical, Radar, and LiDAR) for validating localized field sampling and accurately diagnosing heterogeneous dieback events worldwide. Mangrove dieback Forensic ecology Sentinel-2 Hydrological trap Boambee Creek Restoration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction 1.1 Global Context: Mangrove Resilience and Vulnerability Mangrove ecosystems are among the most carbon-rich and biologically active environments on Earth, providing vital services from protecting coastlines to supporting fisheries (Donato et al., 2011 ). Despite their adaptations to the fluctuating intertidal zones, like aerial root systems (pneumatophores) for gas exchange and salt-exclusion mechanisms, mangroves operate near their physiological limits (Lovelock et al., 2015 ). They are increasingly threatened by a combination of human-related stressors and worsening climate conditions. Understanding what causes mangrove death is essential for conservation efforts. Ecological disruptions are usually classified as either pulse events (sudden, short-term shocks such as cyclones, hailstorms, or tsunamis) or press events (ongoing, long-term stressors like sea-level rise, chemical pollution, or changes in hydrology) (Duke et al., 2017 ). Although mangroves are evolutionarily adapted to recover from pulse events through epicormic resprouting and seedling recruitment, the combination of a physical pulse and persistent hydrological pressure can cause catastrophic ecosystem failure, known as peat collapse or drowning (Cahoon et al., 2003 ). Differentiating these drivers is often difficult because of the diverse nature of estuarine environments. A single mortality event may seem to result from a visible, immediate cause (such as pollution sources), while the underlying reason (such as geomorphic vulnerability) remains hidden. This study investigates one such complex mortality event in the Boambee Creek Estuary, New South Wales, where competing ideas, chemical toxicity versus physical-hydrological failure, have significant implications for management. 1.2 The Boambee Creek Mortality Event In late 2021, a significant dieback event occurred in the mangrove forests of Boambee Creek, which includes Avicennia marina and Aegiceras corniculatum , near the Coffs Harbour Regional Airport. The dieback was patchy, marked by a clear Dead Zone with no canopy, and a nearby Survivor Zone that showed resilience despite being in the same climate zone. The event happened alongside a severe weather event on 20 October 2021, featuring heavy rain and hailstorms (Bureau of Meteorology, 2021 ). The ongoing dieback and failure to recover in some areas led to investigations into human-related causes. Initial studies indicated chemical contamination from the nearby airport was the main cause (Benkendorff et al., 2025 ). However, other interpretations pointed out that physical storm damage could be mistaken for toxicity (Farrugia, 2023 ). 1.3 The Chemical Toxicity Hypothesis: A Critical Review A recent study by Benkendorff et al. ( 2025 ) examined this event and concluded that chemical contamination was the main cause of death. Their study suggested that hydrocarbon deposits from the nearby airport runway and flight path, possibly from fuel dumping or unburned Jet A1 residue, built up in the sediment, causing phytotoxicity. This finding was largely based on the detection of Total Petroleum Hydrocarbons (TPH) in sediment samples taken after the deaths. Although Benkendorff et al. ( 2025 ) provided important biochemical data, the idea that chemical toxicity caused the mortality faces several methodological and theoretical issues that need re-examination. A key limitation of the previous analysis was its dependence on a binary comparison between two single locations ( N = 2). In complex estuarine environments, a sample size of two is inadequate to distinguish site-specific geomorphic features, such as elevation and drainage, from larger landscape stressors. Without a wider spatial assessment, it is impossible to determine whether the Impact site failed due to pollution or simply because it functioned as a topographic sink. Drawing landscape-scale conclusions from such a small sample carries a high risk of selection bias, particularly Pseudoreplication, which can easily cause confusion between correlation and causation (Hurlbert, 1984 ). Detecting hydrocarbons in dead mangrove stands does not necessarily prove they caused the death. Mangrove sediments are known to trap lipophilic contaminants (Tam & Wong, 2000 ), and Avicennia marina is well known for tolerating moderate hydrocarbon levels (Wardrop et al., 1987 ). The Association is not Causation fallacy is especially relevant here; the high volatility of low-molecular-weight hydrocarbons in subtropical estuaries usually prevents them from persisting long in the water column. Linking mortality to acute water-column toxicity needs evidence of a concentration gradient matching the pattern of death, rather than just detecting a point source after the event. The chemical hypothesis assumes that runoff from the airport is the cause of toxicity. This creates a hydrological paradox. If the runway runoff is the source of the toxin, hydrological flow models suggest that the Survivor site, which is also near the runway catchment, should have been equally exposed. If the Control site receives the same or more runoff than the Impact site but remains unaffected, the idea of acute water-column toxicity is logically contradicted. 1.4 The Physical-Hydrological Hypothesis An alternative hypothesis proposed here is that the mortality resulted from a biophysical failure arising from the interaction between Acute Physical Trauma and Hydrological Incompetence. Mangroves depend on atmospheric oxygen diffusing through lenticels on their aerial roots to oxygenate their subterranean root systems (Scholander et al., 1955 ). This process is driven by transpiration, which creates a negative pressure gradient known as biological pumping (Steudle, 2001 ). Severe hailstorms can cause complete defoliation, instantly stopping transpiration. If this defoliation occurs in a basin that cannot drain (a hydrological trap), the loss of biological pumping, combined with static floodwaters, leads to rapid soil anoxia (McKee, 1993 ). Under these conditions, root drowning can occur within days to weeks, causing irreversible mortality regardless of chemical presence. This mechanism indicates that Elevation and Slope, rather than proximity to a flight path, are the primary factors influencing survival. Trees in low-lying basins drown after defoliation, while trees on slopes that drain survive. 1.5 Objectives of the Forensic Reconstruction To settle this debate, this study goes beyond the limitations of point sampling in the field to a landscape-scale forensic reconstruction ( N > 1,300). By combining multi-sensor satellite telemetry (Sentinel-1 SAR, Sentinel-2 MSI), climatological reanalysis (ERA5), and high-resolution LiDAR hydro-topographic modelling, we aim to uncover the physical and chemical factors behind the Boambee Creek dieback. This study pursues three main research goals. First, we assess the meteorological trigger to determine whether the event on 20 October 2021 was a black-swan shock capable of causing physical structural damage or simply a normal seasonal variation. Second, we use spatiotemporal diagnostics with Before-After-Control-Impact (BACI) analysis to establish if the dieback was instant, indicating physical trauma, or gradual, suggesting chronic chemical stress. Third, we examine the mechanism of mortality by modelling the statistical relationship between mortality, elevation, and drainage slope across the entire estuary. This includes explicitly mapping the airport runoff pathways to test the validity of the toxicity vector. This forensic approach offers a rigorous, statistically significant framework ( p < 0.001) for identifying the true cause of the ecosystem collapse. 2. Methodology 2.1 Study Area The study examines the Boambee Creek Estuary (30.35°S, 153.10°E), a subtropical estuarine system on the mid-north coast of New South Wales, Australia. The system is mainly composed of Grey Mangrove ( Avicennia marina ) and River Mangrove ( Aegiceras corniculatum ) (Duke, 2006 ). The site is bordered to the north by Coffs Harbour Regional Airport and experiences mixed semi-diurnal tides. To improve statistical accuracy, the Area of Interest (AOI) was defined not only by the two specific sites noted in previous research (Impact and Control) (Benkendorff et al., 2025 ) but also to encompass the entire connected mangrove area within the catchment, enabling stratified random sampling of over 1,300 pixels (Fig. 1 ). 2.2 Meteorological Forensics (ERA5) To objectively assess the severity of the alleged trigger event (20 October 2021), we used ECMWF ERA5-Land reanalysis data (Hersbach et al., 2020 ). We extracted hourly precipitation and wind gust vectors to characterise the storm's convective profile. A 20-year precipitation baseline (2000–2021) was established to determine the climatological mean ( \(\:\mu\:\) ) and standard deviation ( \(\:\sigma\:\) ) for the area. The storm's intensity was then standardised to a Z-Score ( Z = ( x - \(\:\mu\:\) ) / \(\:\sigma\:\) ). This measure helps identify how rare the event is statistically, differentiating between typical seasonal rainfall and extreme shock events that can cause physical structural damage. 2.3 Spatiotemporal Reconstruction (BACI) We used a Before-After-Control-Impact (BACI) design (Underwood, 1994 ) with satellite telemetry to reconstruct the collapse timeline. Sentinel-2 (Optical) imagery (Drusch et al., 2012 ) was accessed via Google Earth Engine (Gorelick et al., 2017 ) to create monthly median composites of the Normalised Difference Vegetation Index (NDVI) (Tucker, 1979 ) and the Normalised Difference Moisture Index (NDMI) (Gao, 1996 ) from 2020 to 2024. Cloud masking was applied using the QA60 band (cloud and cirrus bitmasks) to maintain radiometric integrity. A Welch’s t-test was conducted on the difference in vegetation indices (Δ NDVI ) before and after the storm to check for structural breaks in the time series. We examined the biophysical connection of the canopy. In healthy mangrove ecosystems, canopy greenness (NDVI) and leaf moisture (NDMI) are closely linked traits (Ceccato et al., 2001 ). We determined the Pearson correlation coefficient ( r ) between these two indices for both the pre-storm and post-storm periods. A notable decoupling of these indices was seen as a sign of severe defoliation and hydraulic failure. 2.4 Hydro-Topographic Auditing To test the Hydrological Trap hypothesis, a 1-metre LiDAR-derived Digital Elevation Model (DEM) (NSW Government Spatial Services, 2025 ) was used to extract geomorphic variables for each estuary pixel. We conducted an estuary-wide logistic regression to model the likelihood of mortality ( P death ) as a function of elevation. This involved sampling N = 1,306 pixels across the mangrove area, classifying them as Dead (dNDVI < -0.15) or Alive, and fitting a sigmoid probability curve (Zuur et al., 2009 ). This method advances beyond anecdotal site comparisons to establish a statistical rule at the landscape scale. Geomorphic drainage potential was evaluated by calculating the slope (degrees) and topographic curvature. Hydrological routing was performed in ArcGIS Pro (Esri, 2025 ) using the D8 flow direction algorithm (O’Callaghan & Mark, 1984 ) to map surface runoff pathways from the airport runway. This enabled verification that the Impact site was the primary recipient of runway runoff, thereby testing the validity of the chemical vector hypothesis. 2.5 Evaluation of Classification Accuracy To validate the spatial extent of the damage, a change detection map (dNDVI) was created by subtracting the post-storm median (Nov 2021–Jan 2022) from the pre-storm median (Jul–Sep 2021). The accuracy of this classification was confirmed using a pixel-level confusion matrix ( N = 114) (Congalton, 1991 ). Ground truth labels were obtained from high-resolution aerial imagery and the known status of the Impact and Control sites. Sensitivity, specificity, and overall accuracy were calculated to ensure that the identified Dead Zone represented a distinct population rather than random noise. 2.6 Statistical Analysis All statistical analyses were conducted within the Python environment (pandas, scipy.stats, sklearn) (Pedregosa et al., 2011 ). Differences in mean elevation and slope between the Impact and Control sites were evaluated using independent t-tests. The significance of the divergence in the time series was analysed with Welch’s t-test for unequal variances. The relationship between elevation and mortality probability was examined using logistic regression and Point-Biserial correlation. All tests were considered significant at the ⍺ = 0.05 level, with high-significance thresholds set at p < 0.001. 3. Results 3.1 The Meteorological Trigger: A Statistical Black Swan To validate the hypothesis of acute physical trauma, the meteorological conditions of the alleged trigger event (20 October 2021) were analysed against a 20-year climatological baseline. The ERA5 reanalysis data (Hersbach et al., 2020 ) confirmed that the Boambee Creek catchment experienced a statistically extreme hydrometeorological anomaly. The storm event registered a precipitation Z-Score of 4.10σ relative to the historical mean (Table 1 ). In standard normal distribution terms, a deviation of this magnitude corresponds to a probability of occurrence of less than 0.01%, classifying the storm as a shock event exceeding the 1-in-100-year recurrence interval. This event was characterised by low surface wind speeds (< 20 km/h) but extreme vertical precipitation intensity (Fig. 2 ). This sets the event apart from a wind-driven cyclonic disturbance. The combination of low horizontal wind shear and high-intensity precipitation aligns with a quasi-stationary convective cell capable of producing large-diameter hail. This explains the specific damage pattern observed: trees were not uprooted by wind, but rather defoliated and debarked on site by the vertical kinetic energy of falling hydrometeors. Table 1 Summary of the forensic meteorological and geomorphic review for the Boambee Creek mortality event. Meteorological data is sourced from ERA5 reanalysis (2000–2021 baseline). Geomorphic data is obtained from a 1-metre LiDAR DEM analysis. Parameter Value Classification/Interpretation Meteorological Trigger (ERA5) Event Date 20 October 2021 Hailstorm/Severe Storm Precipitation Z-Score 4.10σ Extreme Shock Event (> 1-in-100 Year Recurrence) Geomorphic Drainage (LiDAR) Impact Site (Site 1) Slope 1.99° Flat/Stagnant Basin (Hydrological Trap) Control Site (Site 2) Slope 5.09° Drainage Slope (Effective De-Watering) Drainage Ratio 2.6:1 Site 2 is 2.6x steeper than Site 1 3.2 Spatiotemporal Forensics: Instantaneous Ecosystem Collapse The reconstruction of the ecosystem's physiological trajectory using Sentinel-2 telemetry uncovers a clear temporal signature of collapse. The Before-After-Control-Impact (BACI) analysis shows that before the storm event, the Impact Site (Site 1) and the Control Site (Site 2) had synchronised phenological cycles, with no significant statistical difference in mean Normalised Difference Vegetation Index (NDVI). This synchrony suggests that both sites faced similar environmental conditions and had comparable baseline health. Coi ncident with the storm date, the time series shows an immediate structural break. The NDVI at Site 1 plummeted sharply, diverging from the Control site’s trajectory straight after the event (Fig. 3 ). A Welch’s t-test on the pre- and post-storm mean differences ( NDVI ) confirms this divergence is highly statistically significant ( t = 22.88, p < 0.001). The rapid occurrence of this break, within a single satellite capture interval, rules out the hypothesis of chemical senescence. Phytotoxicity caused by soil hydrocarbons usually appears as a slow physiological decline (chlorosis) over months or years. The sudden, steep drop in vegetation indices is a clear spectral sign of acute physical defoliation. This structural collapse was further supported by an analysis of the biophysical state space. In the pre-storm phase, the ecosystem maintained a tightly linked relationship between canopy greenness (NDVI) and leaf moisture content (NDMI), with a Pearson correlation coefficient of r = 0.89. This strong link indicates a functional hydraulic system where chlorophyll content increases linearly with turgor pressure. In the post-storm phase, this relationship decoupled considerably ( r = 0.38) (Fig. 4 ), suggesting a chaotic breakdown of physiological regulation consistent with the severing of the hydraulic continuum due to extensive leaf loss and subsequent xylem cavitation. 3.3 Spatial Extent and Classification Accuracy The spatial analysis confirms that the mortality was not diffuse or random, as might be expected from a volatile chemical plume, but was instead spatially coherent and geomorphically confined. The change detection map (dNDVI) delineated a specific Dead Zone characterised by a mean dNDVI of -0.51, indicating total loss of photosynthetic biomass. The surrounding Survivor Zones maintained a mean dNDVI of -0.07 (Fig. 5 ), representing only minor canopy thinning consistent with recoverable storm damage. The pixel-level accuracy assessment ( N = 114) confirmed this zonation with an Overall Classification Accuracy of 93.9%. The confusion matrix showed a sensitivity of 100% for the Impact zone, meaning that every sampled pixel within the geomorphic basin of Site 1 was correctly identified as Severely Damaged. The high specificity (87.5%) of the Control zone indicates that, while the Survivor site experienced some peripheral stress, it remained a statistically distinct population from the Dead Zone. This precise spatial delineation argues against a gradient-based toxicity model and instead suggests a threshold-based driver of mortality. 3.4 Topographic Determinism: The Death Curve The most decisive outcome of this forensic audit is the identification of elevation as the key variable governing survival. The estuary-wide logistic regression ( N = 1,306) uncovered a strong, non-linear relationship between elevation and mortality. The resulting probability curve (The Death Curve) shows a sigmoid pattern, with the likelihood of mortality approaching 100% as elevation falls below 1.5 metres relative to the Australian Height Datum (AHD)(Fig. 6 ). The statistical significance of this link is undeniable ( p < 1.6 ✕ 10 − 8 ), establishing elevation as a universal predictor of survival throughout the entire estuary. A comparative analysis of the specific study sites explains the difference in their outcomes. Site 1 (The Dead Zone) has a mean elevation of 0.86 m, placing it firmly within the high-risk bathtub zone identified by the logistic model. Site 2 (The Survivor Zone) has a mean elevation of 2.86 m and a vertical difference of 2.00 m. Geomorphic slope analysis shows a key difference in drainage potential. Site 1 has a very gentle slope of 1.99°, forming a flat, stagnant basin. Site 2 has a mean slope of 5.09°, which is 2.6 times steeper than the impact site. This topographic setup confirms that Site 1 acted as a hydrological sink, holding floodwaters and runoff, while Site 2 served as a drainage slope, allowing quick dewatering after the storm surge. This topographic control is further supported by the linear regression of post-storm recovery rates (Fig. 7 ), which demonstrates a strong positive correlation between elevation and vegetation regrowth. 3.5 The Runoff Paradox The hydrological routing analysis carried out in ArcGIS Pro challenges the chemical toxicity hypothesis. The flow accumulation modelling shows that the main surface runoff pathways from the Coffs Harbour Airport runway and tarmac surfaces flow directly into the catchment of Site 2 (The Survivor) (Fig. 8 ). Site 1 remains hydrologically separated from direct runway runoff, receiving mainly runoff from the nearby non-industrial forested catchment. This finding presents a Runoff Paradox for the chemical hypothesis. If the runoff contained lethal levels of hydrocarbons or unburnt fuel, Site 2, which directly receives the hydrological flow, should have experienced equal or higher mortality than Site 1. Site 2 showed significant resilience and fully recovered. The site most exposed to the supposed pollution source's runoff empirically indicates that the runoff water quality was sublethal. The mortality at Site 1 cannot be credited to the water source (airport runoff), but must be due to the water's residence time, which is entirely determined by the basin's geomorphology. 4. Discussion 4.1 The Mechanism of Mortality: The Hydrological Trap The results of this forensic reconstruction offer compelling evidence that the mangrove dieback at Boambee Creek was caused by a biophysical failure best described as a Hydrological Trap. While the initial trigger was certainly the physical trauma from the 20 October 2021 hailstorm, confirmed by the 4.10σ precipitation Z-Score (Table 1 ) and the immediate structural break in the Sentinel-2 time series (Fig. 3 ), the specific morphology of the damage is crucial. The meteorological data confirm a low-wind, high-intensity vertical event (Fig. 2 ), which explains why the trees were defoliated in situ rather than uprooted. The ongoing dieback at Site 1 was entirely determined by basin geomorphology. The logistic regression analysis establishes a statistically significant connection ( p < 10 − 8 ) between elevation and survival, identifying a mortality threshold of approximately 1.5 metres AHD (Fig. 6 ). This finding aligns with the theoretical framework of peat collapse and drowning described in mangrove ecophysiology (Cahoon et al., 2003 ). Mangroves depend on a delicate balance between hydroperiod and aeration; the aerial root systems must be exposed to the atmosphere for a critical part of the tidal cycle to enable oxygen diffusion to the rhizosphere (Scholander et al., 1955 ). The 2.00m elevation deficit and negligible slope (< 2°) at Site 1 characterise it as a topographic depression with limited drainage capacity (Table 1 ). Following the catastrophic defoliation by hail, the cessation of canopy transpiration, the primary mechanism for pumping water out of the soil profile (Steudle, 2001 ), combined with the stagnant topography, led to a state of permanent inundation. Similar mechanisms of delayed mortality due to ponding have been observed in Florida following Hurricane Irma (Radabaugh et al., 2020 ). Lacking the gravitational head to drain (unlike the steeper Site 2), the soil at Site 1 likely transitioned rapidly to severe anoxia (McKee, 1993 ). Under such conditions, root respiration ceases, toxic sulphides accumulate, and mortality becomes irreversible within weeks, regardless of any chemical contaminants. This Hydrological Trap mechanism explains why the dieback was spatially variable despite the uniform delivery of the meteorological shock. The storm did not target Site 1 because it was near a flight path; it targeted Site 1 because it was the only area where the physical removal of the canopy resulted in a fatal hydrological failure. 4.2 Global Parallels: Pulse-Press Interactions in Mangrove Diebacks The interaction between an acute pulse disturbance (hail) and a chronic press stressor (poor drainage) observed in Boambee Creek reflects mechanisms found in major mangrove diebacks worldwide. The large 2015 dieback of Avicennia marina in the Gulf of Carpentaria, Australia, was linked to a similar combination of climatic stress (El Niño-driven drought) and local hydrological failure, where lower sea levels disconnected the mangroves from their water source (Duke et al., 2017 ). After Hurricane Irma (2017) in Florida, patchy mortality was associated with ponding in microtopographic depressions where storm surge water couldn't recede, leading to drowning in a manner very similar to the Boambee Creek event (Radabaugh et al., 2020 ). These global case studies reaffirm the validity of the physical-hydrological hypothesis. They show that when the biophysical limits of mangrove resilience are exceeded by a pulse event, the recovery or collapse path is almost always shaped by the local hydrological regime (Fig. 7 ). The Boambee Creek event adds to this global knowledge by illustrating that hailstorms, often overlooked in favour of cyclones, can generate enough kinetic energy to trigger this peat collapse cascade in vulnerable, low-lying basins (Houston, 1999 ). 4.3 The Runoff Paradox: Challenging the Chemical Excuse The spatial routing of airport runoff offers the strongest rebuttal to the chemical toxicity hypothesis. The fact that the main runway discharge flows into the remaining forest (Site 2) presents a Runoff Paradox that cannot be explained by a toxicity model (Fig. 8 ). In ecotoxicology, the dose-response relationship states that the ecosystem receiving the highest level of a contaminant should show the most severe symptoms. The survival and quick recovery of Site 2 clearly show that the airport runoff did not contain phytotoxic levels of hydrocarbons or herbicides sufficient to cause death. This conclusion is further supported by the survival of the seagrass ( Zostera muelleri ) bio-indicators near the impact zone. Seagrasses are widely regarded as indicators of estuarine water quality because they are highly sensitive to water-column turbidity and dissolved hydrocarbons (Wilson & Ralph, 2012 ). Their continued survival indicates that the water remained chemically benign throughout the event. The selective death of the mangroves at Site 1, therefore, points to mechanisms that affect trees but not submerged grasses: aerial physical trauma and root-zone anoxia. 4.4 Methodological Strengths: From Anecdote to Audit The main strength of this study is its shift from the anecdotal scale of field observation ( N = 2) to the statistical robustness of landscape-scale remote sensing ( N > 1,300). Previous evaluations were limited by selection bias, where the choice of sampling sites influenced the outcomes, a common issue of pseudoreplication (Hurlbert, 1984 ). By performing a stratified random audit across the entire estuary, this study removed such bias and uncovered landscape-scale laws (e.g., the Death Curve) (Fig. 6 ) that were hidden in point-based sampling. The application of a BACI design using high-frequency satellite telemetry enabled precise temporal isolation of the trigger event. While field sampling conducted months after the event can only infer the cause of death from residual evidence (often leading to the Association is not Causation error regarding hydrocarbons), the Sentinel-2 time series provided a real-time physiological trace of the collapse (Fig. 3 ). This ability to identify the exact week of the change in state ( p < 0.001) was crucial in confirming the hailstorm as the key factor. 4.5 Limitations Despite the strength of the geospatial approach, this study recognises certain limitations. The spatial resolution of Sentinel-2 (10m) and the SRTM/LiDAR DEMs (1-30m) inherently generalise micro-topographic features. While adequate for landscape-scale modelling, sub-metre variations in topography (crab-hole micro-relief) that affect seedling recruitment cannot be resolved. Although the Bathtub Effect hypothesis is strongly supported by the convergence of topographic, spectral, and global comparative evidence (Fig. 7 ), no real-time soil redox potential sensors were active in the basin during the storm. The diagnosis of anoxia is inferred from geomorphic boundary conditions rather than direct measurement. While the chemical hypothesis is statistically refuted by the spatial patterns of survival, this study did not perform new chemical assays; it relied on the falsification of the vector hypothesis (the Runoff Paradox) to dismiss toxicity (Fig. 8 ). 4.6 Future Directions and Management Implications The findings of this study indicate a shift in management strategy for Boambee Creek and similar urbanised estuaries. Remediation efforts centred on chemical cleanup or altering flight paths are unlikely to prevent future diebacks, as the cause is geomorphic, not toxicological (Fig. 6 ). Future research should focus on installing in situ hydrological monitoring stations (piezometers) to measure the hydroperiod and drainage rates in low-lying basins. This forensic protocol, which combines BACI satellite analysis with hydrotopographic auditing, should be adopted as a standard first-response tool for investigating coastal vegetation diebacks. By quickly ruling out or confirming physical causes, managers can avoid costly and ineffective interventions based on speculative chemical links. For Boambee Creek, the priority must be restoring hydrological connectivity to the Site 1 basin, possibly through engineering drainage channels, to prevent water stagnation during future extreme weather events (Fig. 7 ), a strategy aligned with the principles of Ecological Mangrove Restoration (Lewis, 2005 ). 5. Conclusion The investigation into the 2021 mangrove dieback at Boambee Creek Estuary is a key case for applying forensic geospatial science in coastal ecology. By shifting the analytical approach from a small field comparison ( N = 2) to a landscape-scale audit ( N = 1,306), this study has fundamentally redefined the event from a localised chemical spill to a system-wide geomorphic failure. The evidence indicates that the Chemical Toxicity hypothesis is unlikely and supports a Physical-Hydrological model of mortality. The meteorological reconstruction identifies the trigger mechanism as an acute atmospheric shock rather than a chronic pollutant accumulation. The storm of 20 October 2021 was a statistically extreme anomaly (Z = 4.10σ), representing a > 1-in-100-year recurrence event. The kinetic energy associated with this event was sufficient to cause catastrophic canopy defoliation, an impact confirmed by the instantaneous structural break observed in the Sentinel-2 time series ( p < 0.001). This study explains the deadly physiological process that occurred after this physical trauma. Mangrove species like Avicennia marina are obligate halophytes that depend on active leaf excretion to control internal salt levels. The near-total defoliation at the Impact Site severely compromised this osmoregulatory ability. Without leaves, the trees lost their primary means of removing salt. The loss of canopy transpiration disrupted the biological pump that maintains root aeration. The persistence of this failure was due to the basin's Hydrological Trap. The logistic regression model ( p < 10 − 8 ) and topographic analysis show that the Impact Site is at a mean elevation of 0.86m with a minimal slope of 1.99°, effectively serving as a stagnant basin. In this topographic depression, the halt in transpiration caused rapid ponding and soil anoxia. The mangroves faced a double impact: severe salt toxicity from their inability to excrete saline loads, and root asphyxiation from their inability to drain floodwaters. The Survivor Site (Site 2), located 2.00m higher and on a slope 2.6 times steeper, maintained enough drainage capacity to avoid ponding, enabling the trees to sustain osmoregulation through residual foliage and quick epicormic recovery. The Runoff Paradox offers a strong rebuttal to the airport contamination theory. The hydrological modelling shows that the main runoff pathways from the runway flow directly into the Survivor Site. The resilience of the forest that receives the direct hydrological load clearly demonstrates that the runoff was not phytotoxic. The survival of the nearby seagrass bioindicators further indicates that there was no water-column toxicity. The Dead Zone at Boambee Creek is not a chemical hotspot but a geomorphic vulnerability revealed by a stochastic weather event. This finding serves as a warning for estuarine management: blaming mortality on convenient human-made sources (such as airports) without thorough geospatial validation can mask the true physical causes. Future conservation efforts should focus on restoring hydrological connections to low-lying basins to reduce the risk of drowning during the increasingly unpredictable storm events forecast under climate change. Declarations Funding Source This research did not receive any specific funding from public, commercial, or not-for-profit agencies. Declaration of Competing Interest The author states that they have no known conflicts of interest or personal relationships that might have influenced the work reported in this paper. CRediT Author Statement Jade Farrugia: Conceptualisation, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing, Visualisation, Project administration. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process During the preparation of this work, the author used Gemini (Google) to enhance language, readability, and structure. After utilising this tool, the author reviewed and edited the content as necessary and accepts full responsibility for the published article. References Benkendorff, K., Briggs, R., Caraco, S., Shilling, J., Islami, M., Davey, A., Boland-Hoskins, E., & Dowell, A. (2025). Investigation of Bara-Baruga (mangrove) ecosystem recovery after a hail storm in Boambee Creek, Gumbaynggirr, NSW. Marine and Freshwater Research , 76 (6), NULL-NULL. Bureau of Meteorology. (2021). Monthly Weather Review—Australia: October 2021 . Australian Government. https://www.bom.gov.au/climate/mwr/aus/mwr-aus-202110.pdf Cahoon, D. R., Hensel, P., Rybczyk, J., McKee, K. L., Proffitt, C. E., & Perez, B. C. (2003). Mass tree mortality leads to mangrove peat collapse at Bay Islands, Honduras after Hurricane Mitch. Journal of Ecology , 91 (6), 1093–1105. Ceccato, P., Flasse, S., Tarantola, S., Jacquemoud, S., & Grégoire, J.-M. (2001). Detecting vegetation leaf water content using reflectance in the optical domain. Remote Sensing of Environment , 77 (1), 22–33. Congalton, R. G. (1991). A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment , 37 (1), 35–46. https://doi.org/10.1016/0034-4257(91)90048-B Donato, D. C., Kauffman, J. B., Murdiyarso, D., Kurnianto, S., Stidham, M., & Kanninen, M. (2011). Mangroves among the most carbon-rich forests in the tropics. Nature Geoscience , 4 (5), 293–297. Drusch, M., Del Bello, U., Carlier, S., Colin, O., Fernandez, V., Gascon, F., Hoersch, B., Isola, C., Laberinti, P., & Martimort, P. (2012). Sentinel-2: ESA’s optical high-resolution mission for GMES operational services. Remote Sensing of Environment , 120 , 25–36. Duke, N. C. (2006). Australia’s mangroves: The authoritative guide to Australia’s mangrove plants . MER. Duke, N. C., Kovacs, J. M., Griffiths, A. D., Preece, L., Hill, D. J., Van Oosterzee, P., Mackenzie, J., Morning, H. S., & Burrows, D. (2017). Large-scale dieback of mangroves in Australia’s Gulf of Carpentaria: A severe ecosystem response, coincidental with an unusually extreme weather event. Marine and Freshwater Research , 68 (10), 1816–1829. Esri. (2025). ArcGIS Pro (Version 3.5) [Computer software]. Environmental Systems Research Institute. https://www.esri.com Farrugia, J. (2023). Debunking misconceptions about Boambee Creek Estuary. https://doi.org/10.5281/zenodo.18446367 Gao, B.-C. (1996). NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space. Remote Sensing of Environment , 58 (3), 257–266. Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment , 202 , 18–27. Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz-Sabater, J., Nicolas, J., Peubey, C., Radu, R., & Schepers, D. (2020). The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society , 146 (730), 1999–2049. Houston, W. A. (1999). Severe hail damage to mangroves at Port Curtis, Australia. Mangroves and Salt Marshes , 3 (1), 29–40. https://doi.org/10.1023/A:1009946809787 Hurlbert, S. H. (1984). Pseudoreplication and the design of ecological field experiments. Ecological Monographs , 54 (2), 187–211. Lewis, R. R. (2005). Ecological engineering for successful management and restoration of mangrove forests. Ecological Engineering , 24 (4), 403–418. https://doi.org/10.1016/j.ecoleng.2004.10.003 Lovelock, C. E., Cahoon, D. R., Friess, D. A., Guntenspergen, G. R., Krauss, K. W., Reef, R., Rogers, K., Saunders, M. L., Sidik, F., & Swales, A. (2015). The vulnerability of Indo-Pacific mangrove forests to sea-level rise. Nature , 526 (7574), 559–563. McKee, K. L. (1993). Soil physicochemical patterns and mangrove species distribution—Reciprocal effects? Journal of Ecology , 477–487. NSW Government Spatial Services. (2025). Elvis—Elevation and Depth—Foundation Spatial Data . https://elevation.fsdf.org.au/ O’Callaghan, J. F., & Mark, D. M. (1984). The extraction of drainage networks from digital elevation data. Computer Vision, Graphics, and Image Processing , 28 (3), 323–344. Pedregosa, F., Pedregosa, F., Varoquaux, G., Varoquaux, G., Org, N., Gramfort, A., Gramfort, A., Michel, V., Michel, V., Fr, L., Thirion, B., Thirion, B., Grisel, O., Grisel, O., Blondel, M., Prettenhofer, P., Prettenhofer, P., Weiss, R., Dubourg, V., … Cournapeau, D. (2011). Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research , 12 , 2825–2830. Radabaugh, K. R., Moyer, R. P., Chappel, A. R., Dontis, E. E., Russo, C. E., Joyse, K. M., Bownik, M. W., Goeckner, A. H., & Khan, N. S. (2020). Mangrove Damage, Delayed Mortality, and Early Recovery Following Hurricane Irma at Two Landfall Sites in Southwest Florida, USA. Estuaries and Coasts , 43 (5), 1104–1118. https://doi.org/10.1007/s12237-019-00564-8 Scholander, P., Van Dam, L., & Scholander, S. I. (1955). Gas exchange in the roots of mangroves. American Journal of Botany , 92–98. Steudle, E. (2001). The cohesion-tension mechanism and the acquisition of water by plant roots. Annual Review of Plant Biology , 52 (1), 847–875. Tam, N. F., & Wong, Y. S. (2000). Spatial variation of heavy metals in surface sediments of Hong Kong mangrove swamps. Environmental Pollution , 110 (2), 195–205. Tucker, C. J. (1979). Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing of Environment , 8 (2), 127–150. Underwood, A. (1994). On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecological Applications , 4 (1), 3–15. Wardrop, J., Butler, A., & Johnson, J. (1987). A field study of the toxicity of two oils and a dispersant to the mangrove Avicennia marina. Marine Biology , 96 (1), 151–156. Wilson, K. G., & Ralph, P. J. (2012). Laboratory testing protocol for the impact of dispersed petrochemicals on seagrass. Marine Pollution Bulletin , 64 (11), 2421–2427. https://doi.org/10.1016/j.marpolbul.2012.08.004 Zuur, A. F., Ieno, E. N., Walker, N., Saveliev, A. A., & Smith, G. M. (2009). Mixed effects models and extensions in ecology with R . Springer. https://doi.org/10.1007/978-0-387-87458-6 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 26 Apr, 2026 Reviews received at journal 16 Apr, 2026 Reviewers agreed at journal 16 Mar, 2026 Reviewers invited by journal 04 Mar, 2026 Editor assigned by journal 02 Mar, 2026 Submission checks completed at journal 02 Mar, 2026 First submitted to journal 20 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8929984","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":602389802,"identity":"098916c2-63c0-49d3-b7db-9aeae3f5dbad","order_by":0,"name":"Jade Farrugia","email":"data:image/png;base64,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","orcid":"","institution":"The University of Queensland","correspondingAuthor":true,"prefix":"","firstName":"Jade","middleName":"","lastName":"Farrugia","suffix":""}],"badges":[],"createdAt":"2026-02-21 02:23:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8929984/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8929984/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104216961,"identity":"40835b0f-4a5b-466d-a6d2-0b304793c529","added_by":"auto","created_at":"2026-03-09 09:21:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2867904,"visible":true,"origin":"","legend":"\u003cp\u003eLocation of the study area in Boambee Creek Estuary, Gumbaynggirr Country, near Coffs Harbour Airport, NSW, Australia. The map shows the Unhealthy Impact Site (Site 1, Orange) and the Healthy Control Site (Site 2, Pink) referenced by Benkendorff et al. (2025). The airport runway's close proximity to the estuary in the north supports the chemical contamination hypothesis tested in this study.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/9cc62891bb0cfdc8afe0a24c.png"},{"id":104405188,"identity":"31357d00-2420-4299-9c9b-1bbea51791b1","added_by":"auto","created_at":"2026-03-11 12:22:03","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":266922,"visible":true,"origin":"","legend":"\u003cp\u003eMeteorological signature of the Black Swan event on 20 October 2021. ERA5 reanalysis data show a statistically extreme precipitation anomaly (\u003cem\u003eZ\u003c/em\u003e \u0026gt; 4.10σ) coinciding with the onset of dieback. Note the clear absence of high wind gusts (\u0026lt;20 km/h) during the precipitation spike (Blue), confirming the event was not a wind-driven cyclone but rather a vertical convective downburst (hail/rain) event. This aligns with the observed pattern of standing dead damage.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/e236dae1ec0e18725b9417eb.png"},{"id":104216959,"identity":"259ccbe9-9296-48e3-ae35-35f5fb31e1ed","added_by":"auto","created_at":"2026-03-09 09:21:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":324824,"visible":true,"origin":"","legend":"\u003cp\u003eSpatiotemporal reconstruction of ecosystem health (2020–2024) using Sentinel-2 NDVI. The vertical dashed line indicates the hailstorm event (20 Oct 2021). Before the event, the Impact Site (Red) and Control Site (Blue) showed synchronous phenology. During the storm, Site 1 experienced an immediate structural collapse. A Welch’s t-test confirms the post-storm divergence is highly\u003cstrong\u003e \u003c/strong\u003esignificant (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), refuting the hypothesis of gradual chemical senescence.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/4f8bebf4f9eb632f6600a1cd.png"},{"id":104405223,"identity":"1de796a1-5c54-41f3-b70d-86331abfcd0c","added_by":"auto","created_at":"2026-03-11 12:22:10","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":439667,"visible":true,"origin":"","legend":"\u003cp\u003eBiophysical state space analysis illustrating physiological decoupling. Pre-storm conditions (green) show a strong coupling (\u003cem\u003er\u003c/em\u003e=0.89) between canopy greenness (NDVI) and leaf moisture (NDMI), indicating a regulated hydraulic system. Post-storm conditions (red) at Site 1 reveal chaotic decoupling (r=0.38) and state collapse, a diagnostic marker of acute defoliation and xylem failure rather than chronic toxicity.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/09ab1ae0e5dd5201d1ae9cf6.png"},{"id":104404045,"identity":"ce3ba2aa-77f2-4b09-9018-657ef66d00d9","added_by":"auto","created_at":"2026-03-11 12:19:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":735432,"visible":true,"origin":"","legend":"\u003cp\u003eSpatial damage extent map (dNDVI) of the Boambee Creek Estuary. Classification relies on the difference between pre- and post-storm median Sentinel-2 composites. The Dead Zone (Dark Red, dNDVI \u0026gt; 0.5) is spatially coherent and geomorphically confined to the low-elevation basin at Site 1. In contrast, the Survivor Zone (Blue/White, dNDVI \u0026lt; 0.1) surrounding Site 2 indicates minimal canopy loss. This clear spatial boundary confirms that the impact was not diffuse (as expected from chemical drift) but physically and topographically constrained.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/c263756e329c244835cd633c.png"},{"id":104216963,"identity":"48888418-b265-43fe-bd59-84953f23bdd6","added_by":"auto","created_at":"2026-03-09 09:21:22","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":179563,"visible":true,"origin":"","legend":"\u003cp\u003eEstuary-wide logistic regression modelling of mortality probability (\u003cem\u003eN\u003c/em\u003e=1,306). The Death Curve (Red Line) shows the probability of mangrove mortality as a function of elevation (AHD). The model indicates a strongly non-linear relationship (\u003cem\u003ep\u003c/em\u003e \u0026lt; 10\u003csup\u003e-8\u003c/sup\u003e), with mortality risk approaching 100% as elevation drops below 1.5m. The vertical dashed line marks the 50% mortality threshold, confirming that low-elevation basins primarily determined survival.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/7694fea5493ba12c92f8bc28.png"},{"id":104216964,"identity":"1ccd13d6-9c0e-418f-89c2-838ceee7aa07","added_by":"auto","created_at":"2026-03-09 09:21:22","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":319542,"visible":true,"origin":"","legend":"\u003cp\u003eLinear regression of vegetation recovery rates vs. elevation. The analysis of post-storm recovery trends (NDVI/Year) reveals a distinct Bathtub Effect. Sites below 1.5m elevation (red dashed line) show zero or negative recovery, indicating ongoing stress (e.g., ponding). Sites above this threshold display a positive recovery trend, confirming that elevation is the main factor limiting ecosystem regeneration.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/6ce4ed3a43e9dbc61295799e.png"},{"id":104404862,"identity":"b872a5ad-35d1-4f79-a46f-10652abbc2ee","added_by":"auto","created_at":"2026-03-11 12:21:14","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1175168,"visible":true,"origin":"","legend":"\u003cp\u003eHydrological flow path modelling of the Boambee Creek Estuary catchment based on a 1-metre LiDAR DEM. The analysis shows that the main surface runoff routes (red lines) from the Coffs Harbour Airport runway flow directly into the Survivor Site 2 (Pink Dot). In contrast, Impact Site 1 (Orange Dot) is hydrologically separated from the main runway discharge. This separation questions the idea that runway runoff caused the mortality at Site 1.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/22376f5a43f719b69c9867f3.png"},{"id":104779754,"identity":"0c3e4b78-ddf0-48fa-9cda-a710a187188f","added_by":"auto","created_at":"2026-03-17 07:45:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6837696,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8929984/v1/e542663c-2c72-4d39-9935-4bc8389c9036.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multi-Sensor Geospatial Modelling to Address Complex Mangrove Dieback: Misattribution of Chemical Stressors Versus Physical Impact","fulltext":[{"header":"1. Introduction","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1 Global Context: Mangrove Resilience and Vulnerability\u003c/h2\u003e \u003cp\u003eMangrove ecosystems are among the most carbon-rich and biologically active environments on Earth, providing vital services from protecting coastlines to supporting fisheries (Donato et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Despite their adaptations to the fluctuating intertidal zones, like aerial root systems (pneumatophores) for gas exchange and salt-exclusion mechanisms, mangroves operate near their physiological limits (Lovelock et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). They are increasingly threatened by a combination of human-related stressors and worsening climate conditions. Understanding what causes mangrove death is essential for conservation efforts. Ecological disruptions are usually classified as either pulse events (sudden, short-term shocks such as cyclones, hailstorms, or tsunamis) or press events (ongoing, long-term stressors like sea-level rise, chemical pollution, or changes in hydrology) (Duke et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Although mangroves are evolutionarily adapted to recover from pulse events through epicormic resprouting and seedling recruitment, the combination of a physical pulse and persistent hydrological pressure can cause catastrophic ecosystem failure, known as peat collapse or drowning (Cahoon et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Differentiating these drivers is often difficult because of the diverse nature of estuarine environments. A single mortality event may seem to result from a visible, immediate cause (such as pollution sources), while the underlying reason (such as geomorphic vulnerability) remains hidden. This study investigates one such complex mortality event in the Boambee Creek Estuary, New South Wales, where competing ideas, chemical toxicity versus physical-hydrological failure, have significant implications for management.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e1.2 The Boambee Creek Mortality Event\u003c/h2\u003e \u003cp\u003eIn late 2021, a significant dieback event occurred in the mangrove forests of Boambee Creek, which includes \u003cem\u003eAvicennia marina\u003c/em\u003e and \u003cem\u003eAegiceras corniculatum\u003c/em\u003e, near the Coffs Harbour Regional Airport. The dieback was patchy, marked by a clear Dead Zone with no canopy, and a nearby Survivor Zone that showed resilience despite being in the same climate zone. The event happened alongside a severe weather event on 20 October 2021, featuring heavy rain and hailstorms (Bureau of Meteorology, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The ongoing dieback and failure to recover in some areas led to investigations into human-related causes. Initial studies indicated chemical contamination from the nearby airport was the main cause (Benkendorff et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, other interpretations pointed out that physical storm damage could be mistaken for toxicity (Farrugia, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e1.3 The Chemical Toxicity Hypothesis: A Critical Review\u003c/h2\u003e \u003cp\u003eA recent study by Benkendorff et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) examined this event and concluded that chemical contamination was the main cause of death. Their study suggested that hydrocarbon deposits from the nearby airport runway and flight path, possibly from fuel dumping or unburned Jet A1 residue, built up in the sediment, causing phytotoxicity. This finding was largely based on the detection of Total Petroleum Hydrocarbons (TPH) in sediment samples taken after the deaths. Although Benkendorff et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) provided important biochemical data, the idea that chemical toxicity caused the mortality faces several methodological and theoretical issues that need re-examination.\u003c/p\u003e \u003cp\u003eA key limitation of the previous analysis was its dependence on a binary comparison between two single locations (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2). In complex estuarine environments, a sample size of two is inadequate to distinguish site-specific geomorphic features, such as elevation and drainage, from larger landscape stressors. Without a wider spatial assessment, it is impossible to determine whether the Impact site failed due to pollution or simply because it functioned as a topographic sink. Drawing landscape-scale conclusions from such a small sample carries a high risk of selection bias, particularly Pseudoreplication, which can easily cause confusion between correlation and causation (Hurlbert, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1984\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDetecting hydrocarbons in dead mangrove stands does not necessarily prove they caused the death. Mangrove sediments are known to trap lipophilic contaminants (Tam \u0026amp; Wong, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2000\u003c/span\u003e), and \u003cem\u003eAvicennia marina\u003c/em\u003e is well known for tolerating moderate hydrocarbon levels (Wardrop et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1987\u003c/span\u003e). The Association is not Causation fallacy is especially relevant here; the high volatility of low-molecular-weight hydrocarbons in subtropical estuaries usually prevents them from persisting long in the water column. Linking mortality to acute water-column toxicity needs evidence of a concentration gradient matching the pattern of death, rather than just detecting a point source after the event.\u003c/p\u003e \u003cp\u003eThe chemical hypothesis assumes that runoff from the airport is the cause of toxicity. This creates a hydrological paradox. If the runway runoff is the source of the toxin, hydrological flow models suggest that the Survivor site, which is also near the runway catchment, should have been equally exposed. If the Control site receives the same or more runoff than the Impact site but remains unaffected, the idea of acute water-column toxicity is logically contradicted.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e1.4 The Physical-Hydrological Hypothesis\u003c/h2\u003e \u003cp\u003eAn alternative hypothesis proposed here is that the mortality resulted from a biophysical failure arising from the interaction between Acute Physical Trauma and Hydrological Incompetence. Mangroves depend on atmospheric oxygen diffusing through lenticels on their aerial roots to oxygenate their subterranean root systems (Scholander et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1955\u003c/span\u003e). This process is driven by transpiration, which creates a negative pressure gradient known as biological pumping (Steudle, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). Severe hailstorms can cause complete defoliation, instantly stopping transpiration. If this defoliation occurs in a basin that cannot drain (a hydrological trap), the loss of biological pumping, combined with static floodwaters, leads to rapid soil anoxia (McKee, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). Under these conditions, root drowning can occur within days to weeks, causing irreversible mortality regardless of chemical presence. This mechanism indicates that Elevation and Slope, rather than proximity to a flight path, are the primary factors influencing survival. Trees in low-lying basins drown after defoliation, while trees on slopes that drain survive.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e1.5 Objectives of the Forensic Reconstruction\u003c/h2\u003e \u003cp\u003eTo settle this debate, this study goes beyond the limitations of point sampling in the field to a landscape-scale forensic reconstruction (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;1,300). By combining multi-sensor satellite telemetry (Sentinel-1 SAR, Sentinel-2 MSI), climatological reanalysis (ERA5), and high-resolution LiDAR hydro-topographic modelling, we aim to uncover the physical and chemical factors behind the Boambee Creek dieback.\u003c/p\u003e \u003cp\u003eThis study pursues three main research goals. First, we assess the meteorological trigger to determine whether the event on 20 October 2021 was a black-swan shock capable of causing physical structural damage or simply a normal seasonal variation. Second, we use spatiotemporal diagnostics with Before-After-Control-Impact (BACI) analysis to establish if the dieback was instant, indicating physical trauma, or gradual, suggesting chronic chemical stress. Third, we examine the mechanism of mortality by modelling the statistical relationship between mortality, elevation, and drainage slope across the entire estuary. This includes explicitly mapping the airport runoff pathways to test the validity of the toxicity vector. This forensic approach offers a rigorous, statistically significant framework (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) for identifying the true cause of the ecosystem collapse.\u003c/p\u003e \u003c/div\u003e"},{"header":"2. Methodology","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study Area\u003c/h2\u003e \u003cp\u003eThe study examines the Boambee Creek Estuary (30.35\u0026deg;S, 153.10\u0026deg;E), a subtropical estuarine system on the mid-north coast of New South Wales, Australia. The system is mainly composed of Grey Mangrove (\u003cem\u003eAvicennia marina\u003c/em\u003e) and River Mangrove (\u003cem\u003eAegiceras corniculatum\u003c/em\u003e) (Duke, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). The site is bordered to the north by Coffs Harbour Regional Airport and experiences mixed semi-diurnal tides. To improve statistical accuracy, the Area of Interest (AOI) was defined not only by the two specific sites noted in previous research (Impact and Control) (Benkendorff et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) but also to encompass the entire connected mangrove area within the catchment, enabling stratified random sampling of over 1,300 pixels (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Meteorological Forensics (ERA5)\u003c/h2\u003e \u003cp\u003eTo objectively assess the severity of the alleged trigger event (20 October 2021), we used ECMWF ERA5-Land reanalysis data (Hersbach et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). We extracted hourly precipitation and wind gust vectors to characterise the storm's convective profile. A 20-year precipitation baseline (2000\u0026ndash;2021) was established to determine the climatological mean (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\mu\\:\\)\u003c/span\u003e\u003c/span\u003e) and standard deviation (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\sigma\\:\\)\u003c/span\u003e\u003c/span\u003e) for the area. The storm's intensity was then standardised to a Z-Score (\u003cem\u003eZ\u003c/em\u003e = (\u003cem\u003ex\u003c/em\u003e - \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\mu\\:\\)\u003c/span\u003e\u003c/span\u003e) / \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\sigma\\:\\)\u003c/span\u003e\u003c/span\u003e). This measure helps identify how rare the event is statistically, differentiating between typical seasonal rainfall and extreme shock events that can cause physical structural damage.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Spatiotemporal Reconstruction (BACI)\u003c/h2\u003e \u003cp\u003eWe used a Before-After-Control-Impact (BACI) design (Underwood, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1994\u003c/span\u003e) with satellite telemetry to reconstruct the collapse timeline. Sentinel-2 (Optical) imagery (Drusch et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) was accessed via Google Earth Engine (Gorelick et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) to create monthly median composites of the Normalised Difference Vegetation Index (NDVI) (Tucker, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1979\u003c/span\u003e) and the Normalised Difference Moisture Index (NDMI) (Gao, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1996\u003c/span\u003e) from 2020 to 2024. Cloud masking was applied using the QA60 band (cloud and cirrus bitmasks) to maintain radiometric integrity. A Welch\u0026rsquo;s t-test was conducted on the difference in vegetation indices (Δ\u003csub\u003e\u003cem\u003eNDVI\u003c/em\u003e\u003c/sub\u003e) before and after the storm to check for structural breaks in the time series.\u003c/p\u003e \u003cp\u003eWe examined the biophysical connection of the canopy. In healthy mangrove ecosystems, canopy greenness (NDVI) and leaf moisture (NDMI) are closely linked traits (Ceccato et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). We determined the Pearson correlation coefficient (\u003cem\u003er\u003c/em\u003e) between these two indices for both the pre-storm and post-storm periods. A notable decoupling of these indices was seen as a sign of severe defoliation and hydraulic failure.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Hydro-Topographic Auditing\u003c/h2\u003e \u003cp\u003eTo test the Hydrological Trap hypothesis, a 1-metre LiDAR-derived Digital Elevation Model (DEM) (NSW Government Spatial Services, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) was used to extract geomorphic variables for each estuary pixel.\u003c/p\u003e \u003cp\u003eWe conducted an estuary-wide logistic regression to model the likelihood of mortality (\u003cem\u003eP\u003c/em\u003e\u003csub\u003e\u003cem\u003edeath\u003c/em\u003e\u003c/sub\u003e) as a function of elevation. This involved sampling \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,306 pixels across the mangrove area, classifying them as Dead (dNDVI \u0026lt; -0.15) or Alive, and fitting a sigmoid probability curve (Zuur et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). This method advances beyond anecdotal site comparisons to establish a statistical rule at the landscape scale.\u003c/p\u003e \u003cp\u003eGeomorphic drainage potential was evaluated by calculating the slope (degrees) and topographic curvature. Hydrological routing was performed in ArcGIS Pro (Esri, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) using the D8 flow direction algorithm (O\u0026rsquo;Callaghan \u0026amp; Mark, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1984\u003c/span\u003e) to map surface runoff pathways from the airport runway. This enabled verification that the Impact site was the primary recipient of runway runoff, thereby testing the validity of the chemical vector hypothesis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Evaluation of Classification Accuracy\u003c/h2\u003e \u003cp\u003eTo validate the spatial extent of the damage, a change detection map (dNDVI) was created by subtracting the post-storm median (Nov 2021\u0026ndash;Jan 2022) from the pre-storm median (Jul\u0026ndash;Sep 2021). The accuracy of this classification was confirmed using a pixel-level confusion matrix (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;114) (Congalton, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). Ground truth labels were obtained from high-resolution aerial imagery and the known status of the Impact and Control sites. Sensitivity, specificity, and overall accuracy were calculated to ensure that the identified Dead Zone represented a distinct population rather than random noise.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Statistical Analysis\u003c/h2\u003e \u003cp\u003eAll statistical analyses were conducted within the Python environment (pandas, scipy.stats, sklearn) (Pedregosa et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Differences in mean elevation and slope between the Impact and Control sites were evaluated using independent t-tests. The significance of the divergence in the time series was analysed with Welch\u0026rsquo;s t-test for unequal variances. The relationship between elevation and mortality probability was examined using logistic regression and Point-Biserial correlation. All tests were considered significant at the ⍺ = 0.05 level, with high-significance thresholds set at \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.1 The Meteorological Trigger: A Statistical Black Swan\u003c/h2\u003e \u003cp\u003eTo validate the hypothesis of acute physical trauma, the meteorological conditions of the alleged trigger event (20 October 2021) were analysed against a 20-year climatological baseline. The ERA5 reanalysis data (Hersbach et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) confirmed that the Boambee Creek catchment experienced a statistically extreme hydrometeorological anomaly. The storm event registered a precipitation Z-Score of 4.10σ relative to the historical mean (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In standard normal distribution terms, a deviation of this magnitude corresponds to a probability of occurrence of less than 0.01%, classifying the storm as a shock event exceeding the 1-in-100-year recurrence interval. This event was characterised by low surface wind speeds (\u0026lt;\u0026thinsp;20 km/h) but extreme vertical precipitation intensity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This sets the event apart from a wind-driven cyclonic disturbance. The combination of low horizontal wind shear and high-intensity precipitation aligns with a quasi-stationary convective cell capable of producing large-diameter hail. This explains the specific damage pattern observed: trees were not uprooted by wind, but rather defoliated and debarked on site by the vertical kinetic energy of falling hydrometeors.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of the forensic meteorological and geomorphic review for the Boambee Creek mortality event. Meteorological data is sourced from ERA5 reanalysis (2000\u0026ndash;2021 baseline). Geomorphic data is obtained from a 1-metre LiDAR DEM analysis.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParameter\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eClassification/Interpretation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMeteorological Trigger (ERA5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEvent Date\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e20 October 2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHailstorm/Severe Storm\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrecipitation Z-Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.10σ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExtreme Shock Event (\u0026gt;\u0026thinsp;1-in-100 Year Recurrence)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGeomorphic Drainage (LiDAR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eImpact Site (Site 1) Slope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.99\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFlat/Stagnant Basin (Hydrological Trap)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eControl Site (Site 2) Slope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.09\u0026deg;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDrainage Slope (Effective De-Watering)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDrainage Ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.6:1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSite 2 is 2.6x steeper than Site 1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Spatiotemporal Forensics: Instantaneous Ecosystem Collapse\u003c/h2\u003e \u003cp\u003eThe reconstruction of the ecosystem's physiological trajectory using Sentinel-2 telemetry uncovers a clear temporal signature of collapse. The Before-After-Control-Impact (BACI) analysis shows that before the storm event, the Impact Site (Site 1) and the Control Site (Site 2) had synchronised phenological cycles, with no significant statistical difference in mean Normalised Difference Vegetation Index (NDVI). This synchrony suggests that both sites faced similar environmental conditions and had comparable baseline health.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCoi\u003c/strong\u003e \u003cp\u003encident with the storm date, the time series shows an immediate structural break. The NDVI at Site 1 plummeted sharply, diverging from the Control site\u0026rsquo;s trajectory straight after the event (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). A Welch\u0026rsquo;s t-test on the pre- and post-storm mean differences (\u003csub\u003e\u003cem\u003eNDVI\u003c/em\u003e\u003c/sub\u003e) confirms this divergence is highly statistically significant (\u003cem\u003et\u003c/em\u003e\u0026thinsp;=\u0026thinsp;22.88, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The rapid occurrence of this break, within a single satellite capture interval, rules out the hypothesis of chemical senescence. Phytotoxicity caused by soil hydrocarbons usually appears as a slow physiological decline (chlorosis) over months or years. The sudden, steep drop in vegetation indices is a clear spectral sign of acute physical defoliation.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThis structural collapse was further supported by an analysis of the biophysical state space. In the pre-storm phase, the ecosystem maintained a tightly linked relationship between canopy greenness (NDVI) and leaf moisture content (NDMI), with a Pearson correlation coefficient of \u003cem\u003er\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.89. This strong link indicates a functional hydraulic system where chlorophyll content increases linearly with turgor pressure. In the post-storm phase, this relationship decoupled considerably (\u003cem\u003er\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.38) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), suggesting a chaotic breakdown of physiological regulation consistent with the severing of the hydraulic continuum due to extensive leaf loss and subsequent xylem cavitation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Spatial Extent and Classification Accuracy\u003c/h2\u003e \u003cp\u003eThe spatial analysis confirms that the mortality was not diffuse or random, as might be expected from a volatile chemical plume, but was instead spatially coherent and geomorphically confined. The change detection map (dNDVI) delineated a specific Dead Zone characterised by a mean dNDVI of -0.51, indicating total loss of photosynthetic biomass. The surrounding Survivor Zones maintained a mean dNDVI of -0.07 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), representing only minor canopy thinning consistent with recoverable storm damage.\u003c/p\u003e \u003cp\u003eThe pixel-level accuracy assessment (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;114) confirmed this zonation with an Overall Classification Accuracy of 93.9%. The confusion matrix showed a sensitivity of 100% for the Impact zone, meaning that every sampled pixel within the geomorphic basin of Site 1 was correctly identified as Severely Damaged. The high specificity (87.5%) of the Control zone indicates that, while the Survivor site experienced some peripheral stress, it remained a statistically distinct population from the Dead Zone. This precise spatial delineation argues against a gradient-based toxicity model and instead suggests a threshold-based driver of mortality.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Topographic Determinism: The Death Curve\u003c/h2\u003e \u003cp\u003eThe most decisive outcome of this forensic audit is the identification of elevation as the key variable governing survival. The estuary-wide logistic regression (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,306) uncovered a strong, non-linear relationship between elevation and mortality. The resulting probability curve (The Death Curve) shows a sigmoid pattern, with the likelihood of mortality approaching 100% as elevation falls below 1.5 metres relative to the Australian Height Datum (AHD)(Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The statistical significance of this link is undeniable (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;1.6 ✕ 10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e), establishing elevation as a universal predictor of survival throughout the entire estuary.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA comparative analysis of the specific study sites explains the difference in their outcomes. Site 1 (The Dead Zone) has a mean elevation of 0.86 m, placing it firmly within the high-risk bathtub zone identified by the logistic model. Site 2 (The Survivor Zone) has a mean elevation of 2.86 m and a vertical difference of 2.00 m. Geomorphic slope analysis shows a key difference in drainage potential. Site 1 has a very gentle slope of 1.99\u0026deg;, forming a flat, stagnant basin. Site 2 has a mean slope of 5.09\u0026deg;, which is 2.6 times steeper than the impact site. This topographic setup confirms that Site 1 acted as a hydrological sink, holding floodwaters and runoff, while Site 2 served as a drainage slope, allowing quick dewatering after the storm surge. This topographic control is further supported by the linear regression of post-storm recovery rates (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e), which demonstrates a strong positive correlation between elevation and vegetation regrowth.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.5 The Runoff Paradox\u003c/h2\u003e \u003cp\u003eThe hydrological routing analysis carried out in ArcGIS Pro challenges the chemical toxicity hypothesis. The flow accumulation modelling shows that the main surface runoff pathways from the Coffs Harbour Airport runway and tarmac surfaces flow directly into the catchment of Site 2 (The Survivor) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Site 1 remains hydrologically separated from direct runway runoff, receiving mainly runoff from the nearby non-industrial forested catchment.\u003c/p\u003e \u003cp\u003eThis finding presents a Runoff Paradox for the chemical hypothesis. If the runoff contained lethal levels of hydrocarbons or unburnt fuel, Site 2, which directly receives the hydrological flow, should have experienced equal or higher mortality than Site 1. Site 2 showed significant resilience and fully recovered. The site most exposed to the supposed pollution source's runoff empirically indicates that the runoff water quality was sublethal. The mortality at Site 1 cannot be credited to the water source (airport runoff), but must be due to the water's residence time, which is entirely determined by the basin's geomorphology.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.1 The Mechanism of Mortality: The Hydrological Trap\u003c/h2\u003e \u003cp\u003eThe results of this forensic reconstruction offer compelling evidence that the mangrove dieback at Boambee Creek was caused by a biophysical failure best described as a Hydrological Trap. While the initial trigger was certainly the physical trauma from the 20 October 2021 hailstorm, confirmed by the 4.10σ precipitation Z-Score (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) and the immediate structural break in the Sentinel-2 time series (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), the specific morphology of the damage is crucial. The meteorological data confirm a low-wind, high-intensity vertical event (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), which explains why the trees were defoliated in situ rather than uprooted. The ongoing dieback at Site 1 was entirely determined by basin geomorphology.\u003c/p\u003e \u003cp\u003eThe logistic regression analysis establishes a statistically significant connection (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e) between elevation and survival, identifying a mortality threshold of approximately 1.5 metres AHD (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). This finding aligns with the theoretical framework of peat collapse and drowning described in mangrove ecophysiology (Cahoon et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Mangroves depend on a delicate balance between hydroperiod and aeration; the aerial root systems must be exposed to the atmosphere for a critical part of the tidal cycle to enable oxygen diffusion to the rhizosphere (Scholander et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e1955\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe 2.00m elevation deficit and negligible slope (\u0026lt;\u0026thinsp;2\u0026deg;) at Site 1 characterise it as a topographic depression with limited drainage capacity (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Following the catastrophic defoliation by hail, the cessation of canopy transpiration, the primary mechanism for pumping water out of the soil profile (Steudle, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2001\u003c/span\u003e), combined with the stagnant topography, led to a state of permanent inundation. Similar mechanisms of delayed mortality due to ponding have been observed in Florida following Hurricane Irma (Radabaugh et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Lacking the gravitational head to drain (unlike the steeper Site 2), the soil at Site 1 likely transitioned rapidly to severe anoxia (McKee, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). Under such conditions, root respiration ceases, toxic sulphides accumulate, and mortality becomes irreversible within weeks, regardless of any chemical contaminants.\u003c/p\u003e \u003cp\u003eThis Hydrological Trap mechanism explains why the dieback was spatially variable despite the uniform delivery of the meteorological shock. The storm did not target Site 1 because it was near a flight path; it targeted Site 1 because it was the only area where the physical removal of the canopy resulted in a fatal hydrological failure.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Global Parallels: Pulse-Press Interactions in Mangrove Diebacks\u003c/h2\u003e \u003cp\u003eThe interaction between an acute pulse disturbance (hail) and a chronic press stressor (poor drainage) observed in Boambee Creek reflects mechanisms found in major mangrove diebacks worldwide. The large 2015 dieback of \u003cem\u003eAvicennia marina\u003c/em\u003e in the Gulf of Carpentaria, Australia, was linked to a similar combination of climatic stress (El Ni\u0026ntilde;o-driven drought) and local hydrological failure, where lower sea levels disconnected the mangroves from their water source (Duke et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). After Hurricane Irma (2017) in Florida, patchy mortality was associated with ponding in microtopographic depressions where storm surge water couldn't recede, leading to drowning in a manner very similar to the Boambee Creek event (Radabaugh et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese global case studies reaffirm the validity of the physical-hydrological hypothesis. They show that when the biophysical limits of mangrove resilience are exceeded by a pulse event, the recovery or collapse path is almost always shaped by the local hydrological regime (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The Boambee Creek event adds to this global knowledge by illustrating that hailstorms, often overlooked in favour of cyclones, can generate enough kinetic energy to trigger this peat collapse cascade in vulnerable, low-lying basins (Houston, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1999\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003e4.3 The Runoff Paradox: Challenging the Chemical Excuse\u003c/h2\u003e \u003cp\u003eThe spatial routing of airport runoff offers the strongest rebuttal to the chemical toxicity hypothesis. The fact that the main runway discharge flows into the remaining forest (Site 2) presents a Runoff Paradox that cannot be explained by a toxicity model (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e). In ecotoxicology, the dose-response relationship states that the ecosystem receiving the highest level of a contaminant should show the most severe symptoms. The survival and quick recovery of Site 2 clearly show that the airport runoff did not contain phytotoxic levels of hydrocarbons or herbicides sufficient to cause death.\u003c/p\u003e \u003cp\u003eThis conclusion is further supported by the survival of the seagrass (\u003cem\u003eZostera muelleri\u003c/em\u003e) bio-indicators near the impact zone. Seagrasses are widely regarded as indicators of estuarine water quality because they are highly sensitive to water-column turbidity and dissolved hydrocarbons (Wilson \u0026amp; Ralph, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Their continued survival indicates that the water remained chemically benign throughout the event. The selective death of the mangroves at Site 1, therefore, points to mechanisms that affect trees but not submerged grasses: aerial physical trauma and root-zone anoxia.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Methodological Strengths: From Anecdote to Audit\u003c/h2\u003e \u003cp\u003eThe main strength of this study is its shift from the anecdotal scale of field observation (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2) to the statistical robustness of landscape-scale remote sensing (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;1,300). Previous evaluations were limited by selection bias, where the choice of sampling sites influenced the outcomes, a common issue of pseudoreplication (Hurlbert, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1984\u003c/span\u003e). By performing a stratified random audit across the entire estuary, this study removed such bias and uncovered landscape-scale laws (e.g., the Death Curve) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) that were hidden in point-based sampling.\u003c/p\u003e \u003cp\u003eThe application of a BACI design using high-frequency satellite telemetry enabled precise temporal isolation of the trigger event. While field sampling conducted months after the event can only infer the cause of death from residual evidence (often leading to the Association is not Causation error regarding hydrocarbons), the Sentinel-2 time series provided a real-time physiological trace of the collapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This ability to identify the exact week of the change in state (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) was crucial in confirming the hailstorm as the key factor.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e4.5 Limitations\u003c/h2\u003e \u003cp\u003eDespite the strength of the geospatial approach, this study recognises certain limitations. The spatial resolution of Sentinel-2 (10m) and the SRTM/LiDAR DEMs (1-30m) inherently generalise micro-topographic features. While adequate for landscape-scale modelling, sub-metre variations in topography (crab-hole micro-relief) that affect seedling recruitment cannot be resolved. Although the Bathtub Effect hypothesis is strongly supported by the convergence of topographic, spectral, and global comparative evidence (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e), no real-time soil redox potential sensors were active in the basin during the storm. The diagnosis of anoxia is inferred from geomorphic boundary conditions rather than direct measurement. While the chemical hypothesis is statistically refuted by the spatial patterns of survival, this study did not perform new chemical assays; it relied on the falsification of the vector hypothesis (the Runoff Paradox) to dismiss toxicity (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e4.6 Future Directions and Management Implications\u003c/h2\u003e \u003cp\u003eThe findings of this study indicate a shift in management strategy for Boambee Creek and similar urbanised estuaries. Remediation efforts centred on chemical cleanup or altering flight paths are unlikely to prevent future diebacks, as the cause is geomorphic, not toxicological (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Future research should focus on installing in situ hydrological monitoring stations (piezometers) to measure the hydroperiod and drainage rates in low-lying basins.\u003c/p\u003e \u003cp\u003eThis forensic protocol, which combines BACI satellite analysis with hydrotopographic auditing, should be adopted as a standard first-response tool for investigating coastal vegetation diebacks. By quickly ruling out or confirming physical causes, managers can avoid costly and ineffective interventions based on speculative chemical links. For Boambee Creek, the priority must be restoring hydrological connectivity to the Site 1 basin, possibly through engineering drainage channels, to prevent water stagnation during future extreme weather events (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e), a strategy aligned with the principles of Ecological Mangrove Restoration (Lewis, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThe investigation into the 2021 mangrove dieback at Boambee Creek Estuary is a key case for applying forensic geospatial science in coastal ecology. By shifting the analytical approach from a small field comparison (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2) to a landscape-scale audit (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,306), this study has fundamentally redefined the event from a localised chemical spill to a system-wide geomorphic failure. The evidence indicates that the Chemical Toxicity hypothesis is unlikely and supports a Physical-Hydrological model of mortality.\u003c/p\u003e \u003cp\u003eThe meteorological reconstruction identifies the trigger mechanism as an acute atmospheric shock rather than a chronic pollutant accumulation. The storm of 20 October 2021 was a statistically extreme anomaly (Z\u0026thinsp;=\u0026thinsp;4.10σ), representing a\u0026thinsp;\u0026gt;\u0026thinsp;1-in-100-year recurrence event. The kinetic energy associated with this event was sufficient to cause catastrophic canopy defoliation, an impact confirmed by the instantaneous structural break observed in the Sentinel-2 time series (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e \u003cp\u003eThis study explains the deadly physiological process that occurred after this physical trauma. Mangrove species like \u003cem\u003eAvicennia marina\u003c/em\u003e are obligate halophytes that depend on active leaf excretion to control internal salt levels. The near-total defoliation at the Impact Site severely compromised this osmoregulatory ability. Without leaves, the trees lost their primary means of removing salt. The loss of canopy transpiration disrupted the biological pump that maintains root aeration.\u003c/p\u003e \u003cp\u003eThe persistence of this failure was due to the basin's Hydrological Trap. The logistic regression model (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e) and topographic analysis show that the Impact Site is at a mean elevation of 0.86m with a minimal slope of 1.99\u0026deg;, effectively serving as a stagnant basin. In this topographic depression, the halt in transpiration caused rapid ponding and soil anoxia. The mangroves faced a double impact: severe salt toxicity from their inability to excrete saline loads, and root asphyxiation from their inability to drain floodwaters. The Survivor Site (Site 2), located 2.00m higher and on a slope 2.6 times steeper, maintained enough drainage capacity to avoid ponding, enabling the trees to sustain osmoregulation through residual foliage and quick epicormic recovery.\u003c/p\u003e \u003cp\u003eThe Runoff Paradox offers a strong rebuttal to the airport contamination theory. The hydrological modelling shows that the main runoff pathways from the runway flow directly into the Survivor Site. The resilience of the forest that receives the direct hydrological load clearly demonstrates that the runoff was not phytotoxic. The survival of the nearby seagrass bioindicators further indicates that there was no water-column toxicity.\u003c/p\u003e \u003cp\u003eThe Dead Zone at Boambee Creek is not a chemical hotspot but a geomorphic vulnerability revealed by a stochastic weather event. This finding serves as a warning for estuarine management: blaming mortality on convenient human-made sources (such as airports) without thorough geospatial validation can mask the true physical causes. Future conservation efforts should focus on restoring hydrological connections to low-lying basins to reduce the risk of drowning during the increasingly unpredictable storm events forecast under climate change.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding Source\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research did not receive any specific funding from public, commercial, or not-for-profit agencies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of Competing Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author states that they have no known conflicts of interest or personal relationships that might have influenced the work reported in this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCRediT Author Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJade Farrugia:\u003c/strong\u003e Conceptualisation, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing, Visualisation, Project administration.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of generative AI and AI-assisted technologies in the manuscript preparation process\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the preparation of this work, the author used Gemini (Google) to enhance language, readability, and structure. After utilising this tool, the author reviewed and edited the content as necessary and accepts full responsibility for the published article.\u003cbr\u003e \u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBenkendorff, K., Briggs, R., Caraco, S., Shilling, J., Islami, M., Davey, A., Boland-Hoskins, E., \u0026amp; Dowell, A. (2025). Investigation of Bara-Baruga (mangrove) ecosystem recovery after a hail storm in Boambee Creek, Gumbaynggirr, NSW. \u003cem\u003eMarine and Freshwater Research\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e(6), NULL-NULL.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBureau of Meteorology. (2021). \u003cem\u003eMonthly Weather Review\u0026mdash;Australia: October 2021\u003c/em\u003e. Australian Government. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bom.gov.au/climate/mwr/aus/mwr-aus-202110.pdf\u003c/span\u003e\u003cspan address=\"https://www.bom.gov.au/climate/mwr/aus/mwr-aus-202110.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCahoon, D. R., Hensel, P., Rybczyk, J., McKee, K. L., Proffitt, C. E., \u0026amp; Perez, B. C. (2003). Mass tree mortality leads to mangrove peat collapse at Bay Islands, Honduras after Hurricane Mitch. \u003cem\u003eJournal of Ecology\u003c/em\u003e, \u003cem\u003e91\u003c/em\u003e(6), 1093\u0026ndash;1105.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCeccato, P., Flasse, S., Tarantola, S., Jacquemoud, S., \u0026amp; Gr\u0026eacute;goire, J.-M. (2001). Detecting vegetation leaf water content using reflectance in the optical domain. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e77\u003c/em\u003e(1), 22\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCongalton, R. G. (1991). A review of assessing the accuracy of classifications of remotely sensed data. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(1), 35\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0034-4257(91)90048-B\u003c/span\u003e\u003cspan address=\"10.1016/0034-4257(91)90048-B\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDonato, D. C., Kauffman, J. B., Murdiyarso, D., Kurnianto, S., Stidham, M., \u0026amp; Kanninen, M. (2011). Mangroves among the most carbon-rich forests in the tropics. \u003cem\u003eNature Geoscience\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(5), 293\u0026ndash;297.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDrusch, M., Del Bello, U., Carlier, S., Colin, O., Fernandez, V., Gascon, F., Hoersch, B., Isola, C., Laberinti, P., \u0026amp; Martimort, P. (2012). Sentinel-2: ESA\u0026rsquo;s optical high-resolution mission for GMES operational services. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e120\u003c/em\u003e, 25\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuke, N. C. (2006). \u003cem\u003eAustralia\u0026rsquo;s mangroves: The authoritative guide to Australia\u0026rsquo;s mangrove plants\u003c/em\u003e. MER.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDuke, N. C., Kovacs, J. M., Griffiths, A. D., Preece, L., Hill, D. J., Van Oosterzee, P., Mackenzie, J., Morning, H. S., \u0026amp; Burrows, D. (2017). Large-scale dieback of mangroves in Australia\u0026rsquo;s Gulf of Carpentaria: A severe ecosystem response, coincidental with an unusually extreme weather event. \u003cem\u003eMarine and Freshwater Research\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e(10), 1816\u0026ndash;1829.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsri. (2025). \u003cem\u003eArcGIS Pro (Version 3.5) [Computer software]. Environmental Systems Research Institute.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.esri.com\u003c/span\u003e\u003cspan address=\"https://www.esri.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarrugia, J. (2023). \u003cem\u003eDebunking misconceptions about Boambee Creek Estuary.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5281/zenodo.18446367\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.18446367\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, B.-C. (1996). NDWI\u0026mdash;A normalized difference water index for remote sensing of vegetation liquid water from space. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(3), 257\u0026ndash;266.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., \u0026amp; Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e202\u003c/em\u003e, 18\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHersbach, H., Bell, B., Berrisford, P., Hirahara, S., Hor\u0026aacute;nyi, A., Mu\u0026ntilde;oz-Sabater, J., Nicolas, J., Peubey, C., Radu, R., \u0026amp; Schepers, D. (2020). The ERA5 global reanalysis. \u003cem\u003eQuarterly Journal of the Royal Meteorological Society\u003c/em\u003e, \u003cem\u003e146\u003c/em\u003e(730), 1999\u0026ndash;2049.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHouston, W. A. (1999). Severe hail damage to mangroves at Port Curtis, Australia. \u003cem\u003eMangroves and Salt Marshes\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(1), 29\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/A:1009946809787\u003c/span\u003e\u003cspan address=\"10.1023/A:1009946809787\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHurlbert, S. H. (1984). Pseudoreplication and the design of ecological field experiments. \u003cem\u003eEcological Monographs\u003c/em\u003e, \u003cem\u003e54\u003c/em\u003e(2), 187\u0026ndash;211.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLewis, R. R. (2005). Ecological engineering for successful management and restoration of mangrove forests. \u003cem\u003eEcological Engineering\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(4), 403\u0026ndash;418. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecoleng.2004.10.003\u003c/span\u003e\u003cspan address=\"10.1016/j.ecoleng.2004.10.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovelock, C. E., Cahoon, D. R., Friess, D. A., Guntenspergen, G. R., Krauss, K. W., Reef, R., Rogers, K., Saunders, M. L., Sidik, F., \u0026amp; Swales, A. (2015). The vulnerability of Indo-Pacific mangrove forests to sea-level rise. \u003cem\u003eNature\u003c/em\u003e, \u003cem\u003e526\u003c/em\u003e(7574), 559\u0026ndash;563.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcKee, K. L. (1993). Soil physicochemical patterns and mangrove species distribution\u0026mdash;Reciprocal effects? \u003cem\u003eJournal of Ecology\u003c/em\u003e, 477\u0026ndash;487.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNSW Government Spatial Services. (2025). \u003cem\u003eElvis\u0026mdash;Elevation and Depth\u0026mdash;Foundation Spatial Data\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://elevation.fsdf.org.au/\u003c/span\u003e\u003cspan address=\"https://elevation.fsdf.org.au/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Callaghan, J. F., \u0026amp; Mark, D. M. (1984). The extraction of drainage networks from digital elevation data. \u003cem\u003eComputer Vision, Graphics, and Image Processing\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(3), 323\u0026ndash;344.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePedregosa, F., Pedregosa, F., Varoquaux, G., Varoquaux, G., Org, N., Gramfort, A., Gramfort, A., Michel, V., Michel, V., Fr, L., Thirion, B., Thirion, B., Grisel, O., Grisel, O., Blondel, M., Prettenhofer, P., Prettenhofer, P., Weiss, R., Dubourg, V., \u0026hellip; Cournapeau, D. (2011). Scikit-learn: Machine Learning in Python. \u003cem\u003eJournal of Machine Learning Research\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 2825\u0026ndash;2830.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRadabaugh, K. R., Moyer, R. P., Chappel, A. R., Dontis, E. E., Russo, C. E., Joyse, K. M., Bownik, M. W., Goeckner, A. H., \u0026amp; Khan, N. S. (2020). Mangrove Damage, Delayed Mortality, and Early Recovery Following Hurricane Irma at Two Landfall Sites in Southwest Florida, USA. \u003cem\u003eEstuaries and Coasts\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(5), 1104\u0026ndash;1118. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12237-019-00564-8\u003c/span\u003e\u003cspan address=\"10.1007/s12237-019-00564-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScholander, P., Van Dam, L., \u0026amp; Scholander, S. I. (1955). Gas exchange in the roots of mangroves. \u003cem\u003eAmerican Journal of Botany\u003c/em\u003e, 92\u0026ndash;98.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteudle, E. (2001). The cohesion-tension mechanism and the acquisition of water by plant roots. \u003cem\u003eAnnual Review of Plant Biology\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(1), 847\u0026ndash;875.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTam, N. F., \u0026amp; Wong, Y. S. (2000). Spatial variation of heavy metals in surface sediments of Hong Kong mangrove swamps. \u003cem\u003eEnvironmental Pollution\u003c/em\u003e, \u003cem\u003e110\u003c/em\u003e(2), 195\u0026ndash;205.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTucker, C. J. (1979). Red and photographic infrared linear combinations for monitoring vegetation. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e(2), 127\u0026ndash;150.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnderwood, A. (1994). On beyond BACI: sampling designs that might reliably detect environmental disturbances. \u003cem\u003eEcological Applications\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(1), 3\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWardrop, J., Butler, A., \u0026amp; Johnson, J. (1987). A field study of the toxicity of two oils and a dispersant to the mangrove Avicennia marina. \u003cem\u003eMarine Biology\u003c/em\u003e, \u003cem\u003e96\u003c/em\u003e(1), 151\u0026ndash;156.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilson, K. G., \u0026amp; Ralph, P. J. (2012). Laboratory testing protocol for the impact of dispersed petrochemicals on seagrass. \u003cem\u003eMarine Pollution Bulletin\u003c/em\u003e, \u003cem\u003e64\u003c/em\u003e(11), 2421\u0026ndash;2427. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.marpolbul.2012.08.004\u003c/span\u003e\u003cspan address=\"10.1016/j.marpolbul.2012.08.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZuur, A. F., Ieno, E. N., Walker, N., Saveliev, A. A., \u0026amp; Smith, G. M. (2009). \u003cem\u003eMixed effects models and extensions in ecology with R\u003c/em\u003e. Springer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-0-387-87458-6\u003c/span\u003e\u003cspan address=\"10.1007/978-0-387-87458-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"environmental-monitoring-and-assessment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emas","sideBox":"Learn more about [Environmental Monitoring and Assessment](http://link.springer.com/journal/10661)","snPcode":"10661","submissionUrl":"https://submission.nature.com/new-submission/10661/3","title":"Environmental Monitoring and Assessment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Mangrove dieback, Forensic ecology, Sentinel-2, Hydrological trap, Boambee Creek, Restoration","lastPublishedDoi":"10.21203/rs.3.rs-8929984/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8929984/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAccurately identifying the causes of mangrove dieback is essential for global coastal management, yet visually similar dieback events can result in incorrect attribution of causes. A significant mangrove dieback in Boambee Creek Estuary (2021) was initially linked to chemical contamination from nearby Coffs Harbour Airport. This study uses a forensic geospatial reconstruction to test the validity of this toxicity hypothesis against a physical-hydrological alternative. Going beyond the limitations of site-specific field sampling (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2), we employed multi-sensor satellite telemetry (Sentinel-2), LiDAR-based geomorphic modelling, and ERA5 climatological reanalysis to assess the ecosystem at the landscape level (\u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1,306). The investigation pinpoints a statistically extreme hailstorm on 20 October 2021 as the trigger event, representing a\u0026thinsp;\u0026gt;\u0026thinsp;1-in-100-year anomaly (\u003cem\u003eZ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.10σ). Time-series diagnostics confirm an immediate structural collapse (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) coinciding with the storm, ruling out the signature of gradual chemical aging. Mortality followed a Death Curve, where the likelihood of death neared 100% at elevations below 1.5m AHD (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;10\u003csup\u003e\u0026minus;\u0026thinsp;8\u003c/sup\u003e) within stagnant topographic basins (\u0026lt;\u0026thinsp;2\u0026deg; slope). Hydrological routing shows that the main runoff from the airport flows directly into the remaining forest, creating a Runoff Paradox that statistically discredits the chemical vector hypothesis. We conclude that the dieback resulted from a Hydrological Trap; sudden physical defoliation stopped canopy transpiration, causing rapid soil anoxia and root drowning in geomorphically unstable basins. Future management should focus on restoring hydrological connectivity rather than chemical remediation. This study highlights the vital need to incorporate landscape-scale multi-sensor remote sensing (Optical, Radar, and LiDAR) for validating localized field sampling and accurately diagnosing heterogeneous dieback events worldwide.\u003c/p\u003e","manuscriptTitle":"Multi-Sensor Geospatial Modelling to Address Complex Mangrove Dieback: Misattribution of Chemical Stressors Versus Physical Impact","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-09 09:21:17","doi":"10.21203/rs.3.rs-8929984/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-26T21:54:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-16T11:02:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"199567358479731881791012762427254856742","date":"2026-03-16T15:15:24+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-04T16:40:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-02T06:20:35+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-02T06:18:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Environmental Monitoring and Assessment","date":"2026-02-21T02:11:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"environmental-monitoring-and-assessment","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"emas","sideBox":"Learn more about [Environmental Monitoring and Assessment](http://link.springer.com/journal/10661)","snPcode":"10661","submissionUrl":"https://submission.nature.com/new-submission/10661/3","title":"Environmental Monitoring and Assessment","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"991906e2-293c-4078-85b0-92e90065c425","owner":[],"postedDate":"March 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-11T16:23:27+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-09 09:21:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8929984","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8929984","identity":"rs-8929984","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}