16 results about "Surface air temperature" patented technology

The invention relates to a surface temperature prediction method and system based on multi-layer perception adaptation and prompt guidance, belongs to the technical field of space-time prediction, solves the problem of low prediction accuracy caused by the fact that inherent heterogeneity in space-time data is not considered in the prior art, and comprises the following steps: S1, obtaining multi-scale time priori knowledge; collecting historical spatio-temporal data in the spatial region to be predicted and processing the historical spatio-temporal data to obtain a standard data set; s2, constructing a prediction model based on multi-layer perception adaptation and prompt collaborative guidance, and generating space-time representation after space-time prompt guidance; s3, adopting the standard data set to train the constructed prediction model based on the multi-layer perception adaptation and prompt collaborative guidance to obtain a trained model; and S4, predicting future spatio-temporal data to obtain spatio-temporal representation after generating spatio-temporal prompt guidance, taking the spatio-temporal representation as a spatio-temporal prediction result, and providing the spatio-temporal prediction result to a surface air temperature monitoring process.

The invention discloses a remote sensing multi-parameter integrated inversion normal form method, system and equipment based on AI-Agent. According to the method, a deep learning neural network is dynamically driven through AI-Agent, a refining mechanism (RM)-Transformer-MoE size nested model, a physical method, a statistical method and expert knowledge are coupled, a DL-C-PSK normal form is constructed, a high-precision multi-source database is established based on the normal form, an appropriate radiation transfer equation is constructed through geophysical logical reasoning, and a high-precision multi-source database is established. And inversion of parameters such as surface temperature, surface emissivity, atmospheric water vapor content and near-surface air temperature is realized. According to a causal relationship between an input wave band and an output parameter, a direct synchronous inversion or iterative inversion mode is adopted to ensure multi-parameter high-precision synchronous inversion. Wherein the core of the deep learning neural network comprises RM logic derivation, SHAP model interpretation, Transform model architecture and a Transform-MoE size nested model, so that the interpretability, the adaptability and the precision of the model are improved. Through an AI-Agent driven RM-Transform-MoE nested model, deep coupling of physics-statistics-knowledge is realized, compared with a traditional SW method, the inversion precision is greatly improved, and verification shows that the technology is suitable for the fields of global climate observation, environment monitoring and the like.

The invention relates to an urban forest cooling function baseline construction and attenuation recovery evaluation method and system. The method comprises the following steps: acquiring day-by-day meteorological data of a research area to identify starting and ending dates of an extreme high temperature event; acquiring urban forest coverage distribution information in the research area and near-surface air temperature data or surface temperature data in a first date range, and calculating urban forest cooling efficiency data of each date in the first date range; calculating statistical characteristics of the urban forest cooling efficiency data of all dates of the beforehand reference period; determining the maximum attenuation value and date of the cooling efficiency during the extreme high-temperature event; and calculating the statistical characteristics of the urban forest cooling efficiency data of each sliding window by using the sliding windows, and calculating the recovery time and rate of urban forest cooling efficiency attenuation according to the baseline and a preset discrimination condition. It is ensured that attenuation is caused by the extreme high-temperature event, and the attenuation degree and the recovery process of the cooling function under the influence of the extreme high-temperature event are quantified.

The invention discloses a shipborne marine atmospheric boundary layer refractive index profile detection system and method, and relates to the technical field of marine atmospheric environment monitoring and electromagnetic wave propagation safeguard.The method comprises the steps that sea-air downlink infrared radiation spectrum data and sea surface air temperature, relative humidity and air pressure data are obtained; constructing a background atmosphere profile by using the reanalysis data, and correcting an initial profile bottom layer by using sea surface observation air temperature, relative humidity and air pressure data to form a prior state vector; constructing an inversion cost function including an observation residual term, a background prior term and a sea surface boundary constraint term, and performing inversion to obtain an atmospheric temperature profile and a relative humidity profile; and calculating an atmospheric refractive index profile according to the temperature profile, the relative humidity profile and the air pressure profile. According to the method, the marine atmospheric refractive index profile can be continuously obtained under the shipborne dynamic observation condition, and reliable data support is provided for marine electromagnetic propagation environment analysis and waveguide diagnosis.

The invention relates to the technical field of health data, in particular to a coastal health care activity grading recommendation method and service system based on meteorological grid forecast, and the method comprises the steps: selecting a coastal health care activity place located on the seaside, obtaining the meteorological grid forecast content of the place, obtaining a deep sea area closest to the place, and obtaining the water temperature of the deep sea area; the meteorological grid forecast content comprises ultraviolet intensity, sea surface water temperature in the meteorological grid, ground temperature, land surface temperature, sea-land horizontal distance in the meteorological grid and wind speed; correcting the surface water temperature, the ground temperature and the wind speed according to the water temperature of the deep sea area; recommending health care activity items to health care personnel in a grading manner according to the ultraviolet intensity and the corrected parameters; by introducing the water temperature of the deep sea area and exploring the influence of the deep sea water on the water temperature, the air temperature and the wind speed of the surface layer sea surface, meteorological grid forecasting is enhanced, forecasting is more accurate, and the situation that the health care activity is proper due to inaccurate weather forecasting is avoided.

The invention discloses a near-surface air temperature data downscaling method driven by multi-source collaborative variables, which relates to the technical field of air temperature data processing and comprises the steps of data preprocessing, sample division, downscaling model construction and air temperature data processing. The method comprises the following steps: firstly, introducing cloud cover, long / short wave radiation, display / latent heat flux and other elements as collaborative variables based on a mode output statistical thought, generating a multi-dimensional multi-source variable data set, and performing preprocessing such as resampling to form a complete data set; carrying out sample division and balance by adopting K-fold cross validation and a first-time adaptive decline algorithm to obtain a total training set and an independent validation set; stable and effective features are screened, and then three types of models including an elastic network, a Bayesian optimization random forest and a neural architecture search-based long and short term memory network are constructed; the to-be-processed data is input into the model to generate high-resolution air temperature data, the data precision is ensured through multi-dimensional verification, systematic deviation is effectively restrained, nonlinear correlation and extreme air temperature correction capacity are enhanced, the space-time stability is improved, and the method is suitable for climatic analysis, ecological environment evaluation and other scenes.

This invention relates to the field of meteorological forecasting, providing a multimodal method for predicting Arctic climate change and analyzing its regulatory mechanisms. The aim is to predict Arctic climate change by understanding the relationship between NPOs, AMOCs, and Arctic near-surface temperatures (SATs). This method constructs and validates a large-sample dataset containing AMOC proxy data and historical Arctic SAT observation data to evaluate the performance and biases of the CMIP6 model in simulating AMOC and Arctic SAT changes. Next, it reveals the physical mechanisms by which AMOCs influence Arctic SATs and explores the regulatory role of NPO internal variability in the relationship between AMOCs and Arctic SATs. Based on this, it simulates future climate change scenarios, adjusts model parameters to accurately reflect the impacts of AMOCs and NPOs on Arctic SATs, and predicts future trends in Arctic SAT and AMOC changes to assess the reliability of the model predictions. This method, used to predict Arctic climate change, improves the accuracy of predicting future Arctic climate changes by gaining a deeper understanding of the interactions between different climate modes.

The present invention provides a method and system for predicting available water volume, relating to the field of artificial intelligence technology. The method includes: fitting a hook-shaped response function that characterizes the relationship between the predicted value of the available water volume in the target basin and the near-surface air temperature based on the available water volume in the target basin during the first time period; determining the preliminary predicted value of the available water volume in the target basin during the second time period corresponding to each global climate model based on the air temperature data of the target basin output by multiple global climate models and the hook-shaped response function; constructing an emergence constraint model according to the air temperature change trend of each global climate model during the third time period and the corresponding preliminary predicted value of the available water volume; and substituting the air temperature change trend of the target basin during the fourth time period into the emergence constraint model to obtain the predicted value of the available water volume in the target basin during the second time period. The present invention predicts the available water volume in the target basin in the future period by fitting the hook-shaped response function and combining the emergence constraint method.

This invention discloses a high-resolution spatialization method for near-surface air temperature in glacial regions. This method integrates measured data from high-altitude meteorological stations, satellite remote sensing data, and machine learning-optimized spatial interpolation techniques to construct a continuous temperature field under complex terrain conditions. First, by fusing station-measured and satellite remote sensing data, a data complementarity method suitable for sparsely observed areas is established. Based on a machine learning optimization algorithm, spatialized glacial temperature records from meteorological station observations are used as training data, while satellite remote sensing data with broader coverage is employed. Surface temperature retrieved through a radiative transfer model is used as input features to correct the surface temperature of the study area, generating a spatial distribution of air temperature that better matches ground observations. Finally, by fusing altitude and vertical temperature lapse rate as core constraint variables and combining benchmark verification based on station observation data, the accuracy of near-surface air temperature spatialization calculations in high-altitude complex terrain areas is improved.

The present invention discloses a three-stage high-precision near-surface temperature remote sensing estimation method, comprising: performing temporal normalization on remotely sensed surface temperature data to eliminate imaging time differences; clustering a study area into multiple sub-areas based on differences in natural conditions; and using four machine learning models to estimate the temperature in each sub-area based on the temporally normalized surface temperature data and spatial auxiliary variables to obtain spatially determined temperature estimates; and integrating the temperature estimation results of the four machine learning models using a generalized additive model to obtain a more accurate near-surface temperature. This method not only effectively eliminates the uncertainty caused by temporal differences in surface temperature but also considers the differences in model adaptability in different geographical environments. By integrating multiple single machine learning models based on an integrated approach, the accuracy of remote sensing temperature estimation is improved.

The application discloses a shipborne marine atmospheric boundary layer refractive index profile detection system and method, relates to the technical field of marine atmospheric environment monitoring and electromagnetic wave propagation guarantee, and comprises the following steps: acquiring sea-air downward infrared radiation spectrum data and sea surface air temperature, relative humidity and air pressure data; constructing a background atmospheric profile by using reanalysis data, and correcting the initial profile bottom layer by using the sea surface observation air temperature, relative humidity and air pressure data to form a prior state vector; constructing an inversion cost function containing an observation residual term, a background prior term and a sea surface boundary constraint term, and inverting to obtain an atmospheric temperature profile and a relative humidity profile; and calculating an atmospheric refractive index profile according to the temperature profile, the relative humidity profile and the air pressure profile. The application can continuously acquire the marine atmospheric refractive index profile under the shipborne dynamic observation condition, and provides reliable data support for marine electromagnetic propagation environment analysis and waveguide diagnosis.

A snow depth and snow water equivalent inversion method based on a passive microwave load comprises the following steps: step 1, simulating a microwave load snow surface brightness temperature by adopting a microwave snow radiation transmission model, and obtaining an effective snow particle size by iterating the snow particle size and minimizing a model simulation brightness temperature difference and a load actual brightness temperature difference under a corresponding condition, the load brightness temperature and the snow depth are combined to complete the construction of a snow depth machine learning model data set; step 2, adopting adaptive LASSO regression to improve an adaptive penalty weight, constructing a machine learning model and solving an inversion formula coefficient by combining the machine learning model data set constructed in the step S1 and adopting a load multi-channel brightness temperature difference and an effective accumulated snow particle size; and step 3, calculating the snow density by adopting near-surface air temperature and wind speed data, and multiplying the snow density by the snow depth obtained by inversion in the step 2 to obtain the snow water equivalent. According to the invention, the snow depth and snow water equivalent inversion algorithm which is universally applicable to the satellite pixel scale can be provided.

The present invention relates to the field of health data technology, and particularly to a method and service system for graded recommendation of coastal health care activities based on meteorological grid forecasts. The method comprises selecting a location for coastal health care activities located on the seaside, obtaining the meteorological grid forecast content of the location, the deep sea area closest to the location, and obtaining the water temperature of the deep sea area; the meteorological grid forecast content includes ultraviolet intensity, sea surface water temperature within the meteorological grid, ground air temperature, land surface temperature, horizontal distance between sea and land within the meteorological grid, and wind speed; the surface water temperature, ground air temperature and wind speed are corrected according to the water temperature of the deep sea area; health care activity projects are graded and recommended to health care personnel according to the ultraviolet intensity and the corrected parameters; the present invention introduces the water temperature of the deep sea area, explores the influence of deep sea water on surface sea water temperature, air temperature and wind speed, strengthens the meteorological grid forecast, makes the forecast more accurate, and avoids the health care activities being counterproductive due to inaccurate weather forecasts.

The application provides a ground weather situation recognition method based on a convolutional neural network, relates to the field of weather situation recognition, and comprises the following steps: establishing a polar coordinate grid system with a target city as an initial original point and performing pressure interpolation to construct an initial ground pressure matrix; correcting the polar coordinate grid system and calculating offset information; performing pressure interpolation and temperature interpolation on each grid point in the corrected coverage area to construct an interpolation uncertainty matrix; calculating the symmetric pressure difference of the corrected original point to obtain a pressure difference matrix; obtaining a multi-channel meteorological feature image based on the ground pressure matrix, the ground temperature matrix, the interpolation uncertainty matrix and the pressure difference matrix, constructing a position offset feature vector based on the offset information, and inputting the trained convolutional neural network to obtain a weather situation category recognition result. The application realizes simultaneous multi-category ground weather situation recognition and improves the recognition stability under complex observation conditions.

The application discloses a kind of near-surface air temperature direct reduction rate generation method based on MODIS data, belong to remote sensing near-surface air temperature direct reduction rate generation technical field.The present application directly generates near-surface air temperature data using MODIS data, reduces the uncertainty brought by many auxiliary parameters in the generation process of near-surface air temperature;Using moving window convolution, strictly control the data in moving window, and then generate high temporal resolution, spatially continuous near-surface air temperature direct reduction rate product.The present application reduces the dependence on a variety of auxiliary ground variables and meteorological station measured values in the traditional near-surface air temperature direct reduction rate estimation method;The present application judges whether the field pixel of remote sensing data meets the requirements pixel by pixel, changes the size of moving window according to the judgment result, strictly controls the input data of model, to obtain credible daily spatially continuous SATLR;The near-surface air temperature direct reduction rate produced by the present application has the characteristics of high temporal resolution, spatially continuous, etc.

The invention relates to a rice field near-surface air temperature estimation method which comprises the following steps: firstly, determining latitude and longitude coordinates of a to-be-estimated region A, then determining a to-be-estimated time period according to the growth period of rice in the region, collecting latitude and longitude coordinates of a surrounding weather station B, and calculating the distance between the region A and the region B as well as azimuth angles, wind directions and wind speed weight factors of the region A and the region B; according to the positions of A and B, calculating the sun-by-sun surface water index (SWI) and leaf area index (LAI) values of the central position of the to-be-estimated region and the weather station from related data in a satellite remote sensing database; and finally, establishing a rice field near-surface air temperature estimation model according to the data, and further calculating the near-surface air temperature of the to-be-estimated region hour by hour or day by day in the to-be-estimated time period according to the model. Quantitative estimation and monitoring of the near-surface air temperature of the rice field are realized, so that the dynamic change of the near-surface air temperature of the rice field can be monitored, and the estimation precision can be improved; the method has an important guiding value for deeply understanding the change rule of the near-surface air temperature of the rice field and realizing accurate management of the rice field.