Integrated prediction method and system for fire spread coupled with smoke and meteorological feedback
By coupling the WRF-Chem and WRF-Fire models, dynamic simulation of the meteorological feedback effect of smoke was achieved, which solved the problem of large fire spread prediction error and improved the accuracy of fire prediction and disaster prevention decision support.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-31
- Publication Date
- 2026-03-10
AI Technical Summary
Existing WRF-Fire models fail to effectively simulate the feedback effect of fire smoke on meteorological conditions, resulting in large errors in fire spread prediction and an inability to accurately predict extreme wildfire behavior.
By coupling the WRF-Chem and WRF-Fire models, fire emissions are estimated in real time and the bidirectional coupling process between smoke aerosols and meteorological fields is simulated. A dynamic meteorological field containing smoke radiation feedback is generated and input into the WRF-Fire model for fire spread calculation, thus establishing a closed-loop feedback mechanism of smoke radiation-meteorology-fire spread.
It improves the physical realism and accuracy of fire spread prediction, reveals the aerosol-dominated positive feedback chain, significantly improves the prediction of fire spread rate and burned area, and provides scientific support for disaster prevention decision-making and emergency response.
Smart Images
Figure CN121031138B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the interdisciplinary field of meteorology and fire science, specifically involving a coupled simulation system and method that combines atmospheric chemical transport models and wildfire spread models. Background Technology
[0002] Forest fires, one of the most destructive natural disasters globally, can have impacts that last for decades. They not only destroy decades of forest resources, causing both ecological and economic losses, but also lead to biodiversity loss. Studies confirm that approximately 20% of global carbon dioxide emissions originate from wildfires, and the resulting smoke pollutants (such as PM2.5) can spread hundreds of kilometers, triggering regional health crises. With climate change exacerbating the frequency and intensity of global wildfires, accurate prediction of fire spread has become a core requirement for disaster prevention and mitigation.
[0003] The WRF (Numerical Weather Prediction) model is a widely used numerical weather prediction model. While the most advanced WRF-Fire (Weather-Wildfire Coupled Simulation and Weather Research Forecasting Model) coupled model can simulate the two-way interaction between fire and atmosphere, it still has fundamental limitations: it does not consider the meteorological feedback effects caused by fire emissions (fire smoke, greenhouse gases, etc.). Wildfire-released aerosol components such as black carbon (BC) and organic carbon (OC) alter the atmospheric temperature field through radiative forcing, inducing local circulation reorganization; heat flux release disrupts boundary layer stability, forming a positive feedback loop of "fire-aerosol-meteorology." This effect is particularly pronounced in large pyrocumulus events, where smoke plumes can form self-sustaining meteorological systems, causing traditional models to have prediction errors of tens of times for extreme fires. Furthermore, the WRF-Fire fire spread simulation does not consider the feedback effects of smoke on meteorology and fire spread. WRF-Chem (numerical weather chemistry forecasting model), an online meteorological-chemical coupling model, can feed back the radiative effects of chemical components such as smoke and cloud interactions into meteorological simulations, but it lacks the ability to simulate fire spread. Therefore, the main problem this method aims to solve is to combine the advantages of WRF-Fire and WRF-Chem to consider the meteorological feedback effect of smoke in fire spread simulations. Summary of the Invention
[0004] This invention proposes an integrated method and system for predicting fire spread by coupling smoke and meteorological feedback, used to accurately predict extreme wildfire behavior and its interaction with the atmospheric environment. It is particularly suitable for forest fire early warning, emergency response decision support, and fire risk assessment in the context of climate change. Its core innovation lies in addressing the critical deficiency of existing technologies in simulating the feedback effect of smoke on fire spread. While traditional WRF-Fire models can simulate fire spread, they completely ignore the physical processes by which fire emissions (such as fire smoke) alter meteorological conditions and thus influence fire spread through radiation effects and cloud interactions. The main breakthrough of this method is that it achieves dynamic coupling simulation of the entire process from fire emission estimation to closed-loop feedback between smoke, meteorology, and fire behavior. This method first estimates the parameters of gaseous pollutants and key radioactive particulate matter emitted by forest fires in real time based on satellite remote sensing data, and converts them into high spatiotemporal resolution dynamic emission fluxes that can be identified by WRF-Chem. Precise matching with the meteorological and chemical grid is achieved through grid interpolation. Then, it innovatively uses WRF-Chem to simulate the bidirectional coupling process between smoke aerosols and the meteorological field. Relying on the MOSAIC (Aerosol Information Organization and Computation Model) aerosol module, it quantifies the radiative effects of black carbon / organic carbon, and combines it with the Fast-J (Fast Photolysis Scheme) photolysis module to generate a dynamic meteorological field that includes smoke radiation feedback, including key elements such as temperature field, wind field, and surface radiation flux. Finally, the corrected dynamic meteorological field is input into the WRF-Fire fire spread model in real time at 5-minute intervals, replacing its original meteorological driving data, thus constructing a closed-loop feedback mechanism of "smoke radiation - meteorological correction - fire spread enhancement". This coupled framework breaks through the limitation of ignoring the radiation effect of smoke in traditional fire spread simulation. For the first time in WRF-Fire, it realizes the key process by which smoke affects atmospheric radiation and thus feeds back into fire spread behavior, improving the physical realism and accuracy of wildfire spread prediction.
[0005] An integrated fire spread prediction method coupling smoke and meteorological feedback includes the following steps:
[0006] Step 1: Estimate forest fire emission parameters in real time based on satellite remote sensing data, including gaseous pollutants ( , ) and key radiation-active particulate matter (black carbon BC, organic carbon OC);
[0007] Step 2: Convert the emission parameters into spatiotemporal dynamic emission fluxes recognizable by WRF-Chem (numerical weather chemistry forecasting model), and achieve accurate matching with the WRF-Chem high-resolution meteorological chemistry grid through grid interpolation;
[0008] Step 3: The WRF-Chem simulation is used to simulate the bidirectional coupling process between smoke aerosols and meteorological fields. The radiation effect of BC / OC is quantified by the MOSAIC (Aerosol Information Organization and Computation Model) aerosol module and combined with the Fast-J (Fast Photolysis Scheme) photolysis module to generate a dynamic meteorological field (temperature field, wind field, and surface radiation flux) including smoke radiation feedback.
[0009] Step 4: Input the dynamic meteorological field output from Step 3 into the WRF-Fire (Weather-Wildfire Coupled Simulation and Weather Research Forecasting Model) model in real time at n-minute intervals, replacing its original meteorological field for fire spread calculation, and establishing a closed-loop feedback chain of "smoke radiation - meteorological correction - fire spread enhancement". Since the meteorological field simulated by WRF-Chem considers the feedback effect of smoke radiation, the fire spread simulation of WRF-Fire can couple the influence of smoke and meteorology, thereby realizing the consideration of the smoke radiation-meteorological coupling effect in fire spread prediction. Through this coupling mechanism, WRF-Fire has for the first time realized the simulation of the feedback effect of smoke on fire spread.
[0010] As described in step 4, the meteorological field output from WRF-Chem is input into WRF-Fire for fire spread simulation. Since the meteorological field of WRF-Chem considers the effect of smoke meteorological feedback, it can also be used in WRF-Fire fire spread simulation to reveal the impact of smoke meteorological feedback on fire spread. Through the dynamic coupling of WRF-Chem and WRF-Fire, the deficiency of traditional WRF-Fire fire spread simulation in failing to reveal the smoke meteorological feedback mechanism is solved.
[0011] The coupling mechanism described in step 4 is specifically manifested as follows: (1) Smoke radiation feedback effect: Aerosols absorb solar radiation to increase the boundary layer temperature, and the scattering effect changes the surface heat flux; (2) Meteorological dynamic correction: The corrected wind field enhances oxygen transport in front of the fire line, and the change in surface radiation flux accelerates the preheating of combustibles; (3) Time-dependent effect: The dynamic coupling mechanism captures the phased effects of organic carbon aerosols (initial suppression and later acceleration of fire spread); (4) Extreme fire behavior prediction: Fire spread rate prediction is achieved based on smoke feedback, solving the problem of underestimation of fire spread rate caused by neglecting smoke feedback in traditional models.
[0012] In another aspect, the present invention also provides an integrated fire spread prediction system that couples smoke and meteorological feedback, comprising the following modules:
[0013] The parameter estimation module is used to estimate forest fire emission parameters in real time based on satellite remote sensing data.
[0014] The grid matching module is used to convert emission parameters into spatiotemporal dynamic emission fluxes that can be recognized by the numerical weather chemistry forecasting model WRF-Chem, and achieves matching with the WRF-Chem meteorological chemistry grid through grid interpolation;
[0015] The dynamic meteorological field generation module is used to generate a dynamic meteorological field that includes smoke radiation feedback by simulating the bidirectional coupling process between smoke aerosols and meteorological fields in WRF-Chem after mesh matching.
[0016] The fire spread prediction module is used to input the dynamic meteorological field into the weather-wildfire coupled simulation and weather research forecast model WRF-Fire in real time, replace its original meteorological field for fire spread calculation, establish a closed-loop feedback chain of smoke radiation-meteorological correction-fire spread enhancement, and realize the simulation of the feedback effect of smoke on fire spread.
[0017] The beneficial effects of this invention are:
[0018] (1) At the level of mechanism cognition, an aerosol-dominated positive feedback chain is revealed for the first time: fire emissions, through the synergistic effect of thermodynamic heating and radiative forcing, lead to an increase in wind speed at a height of 10 meters by 1-3 m / s. This self-reinforcing effect makes the fire spread rate in the "With Fire" scenario significantly higher than that in the "No Fire" scenario within 2.5 hours of ignition, and the burned area expands by 23-47 times. Of particular note is the two-phase effect of organic carbon (OC), which shows that the constant parameterization of the traditional model deviates significantly from reality: when the fire lasts for 100 minutes, the burned area in the no-OC scenario exceeds the baseline by 17.3%.
[0019] (2) In terms of application value, this technology provides scientific support for disaster prevention decision-making. A graded response strategy based on the OC dual-phase effect is implemented: in the early stages of a fire, priority is given to blocking OC-containing fuels (such as coniferous forests), while in the later stages, downwind protection is strengthened. The real-time risk map output by the system can guide evacuation route planning, improving emergency response efficiency. Regarding climate adaptation, the coupled model confirms that aerosol feedback will increase the frequency of extreme fires when global warming occurs. Attached Figure Description
[0020] Figure 1 This is a system architecture diagram of the present invention;
[0021] Figure 2 This refers to the forest fire risk simulation area considered in this invention;
[0022] Figure 3 A comparison diagram of the fire front's propagation trajectory;
[0023] Figure 4 The fire spread rate at different times under different configurations;
[0024] Figure 5The time-related fire area is set for different situations. Detailed Implementation
[0025] To better illustrate the invention and advantages of this project, the invention will be further explained below with reference to the accompanying drawings and examples;
[0026] Example 1:
[0027] This invention aims to overcome existing technological bottlenecks and provide an integrated method and system for predicting fire spread that couples smoke and meteorological feedback, such as... Figure 1 As shown. The core innovation lies in constructing a real-time interactive system between the WRF-Chem (numerical weather chemistry forecasting model) atmospheric chemistry model and the WRF-Fire fire spread model. By quantifying the correction of meteorological fields by aerosol radiation effects, the self-reinforcing mechanism of fire smoke is revealed. The specific technical solution includes three innovative aspects:
[0028] In the aerosol-meteorological feedback quantification stage, a high-resolution WRF-Chem model is used to simulate the feedback mechanism of smoke radiation effects on meteorology. WRF-Chem feeds back the optical properties of smoke, cloud interactions, and radiation effects to the simulation of temperature, wind field, etc., thereby simulating the impact of smoke on meteorological changes.
[0029] Since current WRF-Fire fire spread simulations do not consider the impact of smoke radiation feedback on fire spread, this method innovatively establishes a data exchange mechanism that iterates every 5 minutes in the dynamic coupling prediction stage. The corrected meteorological field output by WRF-Chem (including the temperature gradient, radiation flux, and three-dimensional wind field after aerosol perturbation) is input into the WRF-Fire model in real time, and atmospheric parameters are updated synchronously every 300 seconds, accurately capturing the transient coupling process of aerosol-meteorology-fire. This high-frequency interaction achieves, for the first time, a closed-loop feedback simulation from emission generation and atmospheric response to changes in fire behavior.
[0030] This invention proposes a wildfire prediction method based on a dynamic coupling framework of smoke-meteorology and fire spread, specifically including the following steps:
[0031] Step 1: Estimate forest fire emission parameters in real time based on satellite remote sensing data, including gaseous pollutants ( , ) and key radiation-active particulate matter (black carbon BC, organic carbon OC);
[0032] Step 2: Convert the emission parameters into spatiotemporal dynamic emission fluxes recognizable by WRF-Chem (Chemical Weather Research and Forecasting Model), and achieve accurate matching with the WRF-Chem high-resolution meteorological and chemical grid through grid interpolation;
[0033] Step 3: The WRF-Chem simulation is used to simulate the bidirectional coupling process between smoke aerosols and meteorological fields. The radiation effect of BC / OC is quantified by the MOSAIC (Aerosol Information Organization and Computation Model) aerosol module and combined with the Fast-J (Fast Photolysis Scheme) photolysis module to generate a dynamic meteorological field (temperature field, wind field, and surface radiation flux) including smoke radiation feedback.
[0034] Step 4: Input the dynamic meteorological field output from Step 3 into the WRF-Fire (Weather-Wildfire Coupled Simulation and Weather Research Forecasting Model) model in real time at n-minute intervals, replacing its original meteorological field for fire spread calculation, and establishing a closed-loop feedback chain of "smoke radiation - meteorological correction - fire spread enhancement". Since the meteorological field simulated by WRF-Chem considers the feedback effect of smoke radiation, the fire spread simulation of WRF-Fire can couple the influence of smoke and meteorology, thereby realizing the consideration of the smoke radiation-meteorological coupling effect in fire spread prediction. Through this coupling mechanism, WRF-Fire has for the first time realized the simulation of the feedback effect of smoke on fire spread.
[0035] As described in step 1, satellite remote sensing data is used to monitor smoke emissions, including... , Other particulate matter, such as black carbon and organic carbon, is included in the emissions calculations to ensure that smoke emissions reflect actual conditions. This allows for a comprehensive assessment of the impact of wildfire emissions and their components. By excluding certain smoke components from emissions calculations, the effects of smoke on meteorology and fire spread can be explored. These sensitivity experiments can separate the individual and combined effects of different fire emission species and their radiative interactions.
[0036] As shown in step 2, WRF-Chem simulation can couple smoke emissions into meteorological simulation online. First, the smoke emissions need to be converted into the form required by WRF-Chem, and then grid interpolation is used to achieve accurate matching with the WRF-Chem high-resolution meteorological and chemical grid.
[0037] As described in step 3, this invention develops an innovative coupled modeling framework for studying the impact of meteorological feedback caused by fire emissions on wildfire spread dynamics. Since WRF-Fire does not consider the influence of smoke radiation and other effects on the meteorological field in wildfire spread simulation, this method first simulates the impact of smoke on meteorology based on WRF-Chem, and then uses the meteorological field output by WRF-Chem for fire spread simulation, thereby ensuring that the smoke-fire spread simulation can explore the meteorological feedback effect of smoke.
[0038] As described in step 4, the meteorological field output from WRF-Chem is input into WRF-Fire for fire spread simulation. Since the meteorological field of WRF-Chem considers the effect of smoke meteorological feedback, it can also be used in WRF-Fire fire spread simulation to reveal the impact of smoke meteorological feedback on fire spread. Through the dynamic coupling of WRF-Chem and WRF-Fire, the deficiency of traditional WRF-Fire fire spread simulation in failing to reveal the smoke meteorological feedback mechanism is solved.
[0039] The coupling mechanism described in step 4 is specifically manifested as follows: (1) Smoke radiation feedback effect: aerosols absorb solar radiation to increase the boundary layer temperature, and the scattering effect changes the surface heat flux; (2) Meteorological dynamic correction: the corrected wind field enhances oxygen transport in front of the fire line, and the change in surface radiation flux accelerates the preheating of combustibles; (3) Time-dependent effect: the dynamic coupling mechanism captures the phased effects of organic carbon aerosols (initial suppression and later acceleration of fire spread); (4) Extreme fire behavior prediction: solves the problem of underestimation of fire spread speed caused by neglecting smoke feedback in traditional models.
[0040] Through this tightly coupled framework, this invention examines how smoke-induced surface heat flux, boundary layer stability, and wind pattern variations influence fire behavior in WRF-Fire, including fire intensity, spread rate, and fuel consumption patterns. A carefully chosen 5-minute coupling interval captures both the rapidly evolving fire-atmosphere interactions and ensures computational feasibility. This method enables a systematic study of the entire feedback loop from fire emissions to atmospheric changes and then to the dynamics of fire spread. This approach provides new insights into the predictability of fire behavior, particularly for extreme wildfire events where these feedback mechanisms dominate.
[0041] Example 2:
[0042] In another aspect, the present invention also provides an integrated fire spread prediction system that couples smoke and meteorological feedback, comprising the following modules:
[0043] The parameter estimation module is used to estimate forest fire emission parameters in real time based on satellite remote sensing data.
[0044] The grid matching module is used to convert emission parameters into spatiotemporal dynamic emission fluxes that can be recognized by the numerical weather chemistry forecasting model WRF-Chem, and achieves matching with the WRF-Chem meteorological chemistry grid through grid interpolation;
[0045] The dynamic meteorological field generation module is used to generate a dynamic meteorological field that includes smoke radiation feedback by simulating the bidirectional coupling process between smoke aerosols and meteorological fields in WRF-Chem after mesh matching.
[0046] The fire spread prediction module is used to input the dynamic meteorological field into the weather-wildfire coupled simulation and weather research forecast model WRF-Fire in real time, replace its original meteorological field for fire spread calculation, establish a closed-loop feedback chain of smoke radiation-meteorological correction-fire spread enhancement, and realize the simulation of the feedback effect of smoke on fire spread.
[0047] Figure 2 The wind field variation characteristics under different WRF-Chem simulation settings are presented. Compared with the "no smoke emission" scenario, the "smoke emission" scenario shows a significantly accelerated wind speed change. This acceleration may be driven by fire emissions, which enhances regional meteorological feedback through the following mechanisms: (1) the heat flux caused by the fire destabilizes the boundary layer and promotes turbulent mixing; (2) the aerosol-radiation effect alters the local pressure gradient, thereby enhancing near-surface wind speed. The "no black carbon", "no organic carbon", and "no aerosol radiation feedback" scenarios show similar regional variation patterns to the simulation containing complete fire emissions, indicating that although individual aerosol components contribute to wind modulation, their absence alone does not significantly change the overall spatial distribution.
[0048] Figure 3 The wind field changes shown demonstrate the impact of forest fire emissions on fire line characteristics. The results indicate that forest fire emissions significantly alter fire line propagation patterns by changing regional climate conditions through multiple pathways. Compared to the baseline scenario of "no smoke emissions," the fire spread rate in the "smoke emissions" scenario is significantly faster, which can be attributed to three key mechanisms: (1) the heat flux generated by fires enhances atmospheric instability and convective activity; (2) aerosol emissions alter boundary layer dynamics through radiation absorption effects; and (3) fire-generated pyrocumulus clouds influence local wind patterns.
[0049] Temporal analysis of the aerosol effect revealed a more complex nonlinear relationship. In the initial phase (01:30:00, August 16, 2015), the “no organic carbon” scenario (excluding the “no smoke” scenario) exhibited the slowest spread rate, consistent with findings that organic carbon enhanced the coupling between fire and atmosphere through radiative heating in the early morning. However, this pattern dramatically reversed within just one hour (to 02:30:00), with the “no organic carbon” scenario showing the fastest spread rate. The fire spread rates of the “no black carbon” and “no aerosol radiative feedback” scenarios also showed temporal fluctuations. These variations are likely due to significant temporal heterogeneity in the interaction between aerosols and fire, influenced by factors including: (i) varying solar altitude angles affecting radiative absorption efficiency; (ii) varying atmospheric stability conditions; and (iii) nonlinear feedback between fire intensity and aerosol generation. The observed reversal of the “no organic carbon” effect, in particular, highlights the importance of considering time scales when assessing the impact of aerosols on wildfire behavior, as demonstrated by observational studies using coupled fire-atmosphere models.
[0050] This comprehensive analysis reveals that while fire emissions generally accelerate fire spread rates through thermodynamic and kinetic pathways, the specific effects of individual aerosol components (BC, OC) and their radiative feedback are highly dependent on time and meteorological context. These findings have significant implications for fire modeling, suggesting that parameterization schemes should simultaneously consider direct thermal effects and time-varying aerosol-atmosphere interactions to improve prediction accuracy.
[0051] Figure 4 Fire spread rates under different experimental configurations were demonstrated. The results showed that the fire spread rate in the "no fire, no smoke" scenario was significantly lower than in other scenarios, confirming that forest fire emissions significantly increased the fire spread rate. This acceleration effect is consistent with existing research findings that fire emissions promote fire-atmosphere coupling through the following mechanisms: (1) radiative heating of smoke aerosols, (2) enhanced convective circulation, and (3) preheating of fuel by thermal radiation. The similar distribution patterns observed in the "no black carbon," "no organic carbon," and "no aerosol radiative feedback" scenarios indicate that these components have comparable radiative effects.
[0052] Figure 5 The temporal evolution of burned area under different experimental configurations was shown, revealing key information about fire-atmosphere interactions. The burned area in the "no smoke emission" scenario was significantly smaller than that in the "smoke emission" scenario, with the latter exceeding the burned area by tens of times within just 2.5 hours after ignition. This huge difference quantitatively confirms previous findings that forest fire emissions enhance fire spread through multiple mechanisms: (1) aerosol-induced radiative heating pre-treats the fuel, (2) smoke-driven convective feedback loops, and (3) altered atmospheric boundary layer dynamics.
[0053] In the initial stages of a fire (<1 hour), the differences between the scenarios of "smoke emission," "no black carbon," "no organic carbon," and "no aerosol radiative feedback" are minimal. However, as the fire progresses, a clear differentiation pattern emerges: until +2 hours after ignition, the "no organic carbon" scenario develops to have the largest burned area, followed by the "no black carbon" and "no aerosol radiative feedback" scenarios. This evolution may be due to the hierarchical nature of aerosol effects, where the absence of organic carbon has the greatest impact on the normal optical properties of smoke, while black carbon and radiative feedback act through different but interconnected pathways. The amplification of these differences over time highlights the cumulative nature of aerosol-fire interactions, where initial microphysical effects gradually manifest as macroscopic changes in fire behavior.
[0054] This invention addresses a key gap in wildfire modeling, focusing on the meteorological feedback effect of fire emissions on fire spread dynamics—a phenomenon long overlooked in current operational forecasting systems. By innovatively coupling the WRF-Chem atmospheric chemistry model with the WRF-Fire fire spread model, we have developed a new framework capable of fully capturing the feedback loop of "fire emissions-meteorological field changes-fire behavior changes."
[0055] Based on five sensitivity experimental designs, this invention reveals three core findings: (1) Emission-driven wind speed enhancement mechanism: Fire emissions enhance wind speed through the synergistic effect of thermodynamic and aerosol radiation effects, forming a self-reinforcing fire spread trend; (2) Dramatic fire behavior effect: Coupled system shows that aerosol emissions significantly accelerate the initial spread rate, and the burned area is tens of times larger than that in the no-emission scenario within 2.5 hours; (3) Time-varying interactive characteristics of aerosols: There is a complex time-varying interaction between smoke chemical processes and fire dynamic behavior.
[0056] These findings deepen our understanding of the fire-atmosphere-chemical coupling process, demonstrating that emission feedback is not merely a secondary factor, but a key driving mechanism for extreme fire behavior. The results provide a physical basis for improving operational fire models, particularly by enhancing the characterization of aerosol-meteorological interactions, effectively addressing the underestimation of the spread potential of catastrophic fires by traditional models. This study's wildfire prediction framework comprehensively considers the fire emission-meteorological feedback mechanism, providing direct guidance for emergency response planning and climate change adaptation strategies in high-risk fire areas.
Claims
1. A method for integrated prediction of fire spread coupled with smoke and meteorological feedbacks, characterized in that, The method comprises the following steps: Step 1: Real-time estimation of forest fire emission parameters based on satellite remote sensing data; Step 2: Convert the emission parameters into a spatio-temporal dynamic emission flux that can be recognized by the numerical weather chemistry prediction model WRF-Chem, and realize the matching with the WRF-Chem meteorological chemistry grid through grid interpolation; Step 3: After the meteorological chemistry grid matching, based on the WRF-Chem simulation of the two-way coupling process of smoke aerosol and meteorological field, a dynamic meteorological field containing smoke radiation feedback is generated; Step 4: Input the dynamic meteorological field into the weather-fire coupling simulation and the weather research and forecast model WRF-Fire to realize the simulation of the feedback effect of smoke on fire spread, which is specifically realized as follows: With n The dynamic weather field is input into the weather-fire coupling simulation in real time at intervals of 1 minute to replace the original weather field for fire spread calculation, to establish a smoke radiation-weather correction-fire spread enhancement closed-loop feedback chain, and to realize the simulation of the feedback effect of smoke on fire spread.
2. The method of claim 1, wherein the method further comprises: The forest fire emission parameters include gaseous pollutants and key radiation active particles. The gaseous pollutants include CO2 and SO2; the key radiation active particles include black carbon BC and organic carbon OC.
3. The method of claim 1, wherein the method further comprises: The specific implementation process of step 3 is as follows: the WRF-Chem simulation of the two-way coupling process of smoke aerosol and meteorological field is adopted, the radiation effect of BC / OC is quantified through the aerosol information organization and calculation model MOSAIC aerosol module, and the dynamic meteorological field containing smoke radiation feedback is generated in combination with the Fast-J photolysis module. The dynamic meteorological field includes temperature field, wind field and surface radiation flux.
4. The method of claim 1, wherein the method further comprises: The coupling in step 4 is specifically as follows: Smoke radiation feedback effect: aerosol absorption of solar radiation increases the boundary layer temperature, and scattering effect changes the surface heat flux; Meteorological dynamic correction: the corrected wind field enhances the oxygen transport in front of the fire line, and the change of surface radiation flux accelerates the preheating of combustible materials; Time-dependent effect: the dynamic coupling mechanism captures the phased action of organic carbon aerosol, realizes the inhibition in the early stage and accelerates the fire spread in the later stage; Extreme fire behavior prediction: the fire spread speed is predicted according to the smoke feedback.
5. The method of claim 4, wherein the method further comprises: The corrected meteorological field obtained after the meteorological dynamic correction contains the temperature gradient, radiation flux and three-dimensional wind field after the disturbance of aerosol.
6. The method of claim 4, wherein the fire spread integrated prediction method coupled with the smoke and weather feedback effect is characterized by, In the process of coupling, it also includes the radiation heating of smoke aerosol, the enhanced convective circulation and the preheating of fuel by thermal radiation.
7. A fire spread integrated prediction system coupling smoke and meteorological feedbacks for implementing the fire spread integrated prediction method according to any one of claims 1 to 6, characterized in that, The method comprises the following modules: A parameter estimation module for real-time estimation of forest fire emission parameters based on satellite remote sensing data; A grid matching module for converting the emission parameters into a spatio-temporal dynamic emission flux that can be recognized by the numerical weather chemistry prediction model WRF-Chem, and realizing the matching with the WRF-Chem meteorological chemistry grid through grid interpolation; A dynamic meteorological field generation module for generating a dynamic meteorological field containing smoke radiation feedback through the two-way coupling process of smoke aerosol and meteorological field simulated by the grid-matched WRF-Chem; A fire spread prediction result module for real-time input of the dynamic meteorological field into the weather-fire coupling simulation and the weather research and forecast model WRF-Fire to replace the original meteorological field for fire spread calculation, establish a smoke radiation-meteorological correction-fire spread enhancement closed loop feedback chain, and realize the simulation of the feedback effect of smoke on fire spread.