32 results about "Climate pattern" patented technology

The invention discloses a staged design flood calculation method under the integrated climate model. By establishing the coupling model of GCM and VIC, the runoff process of the reservoir under the future climate change scenario is predicted, and the average value under the climate change scenario is obtained through the arithmetic mean method. The runoff process is analyzed by the probability change point method, and the main flood season and the non-main flood season are divided into main and non-main flood seasons. The calculation model of the most probable combination method for reservoir design flood peak volume is constructed through the Copula function, and the predicted runoff data of the reservoir under the future climate change scenario is used as Design the input of the flood peak combination method to calculate the model, and consider the combination of multi-variables with the same frequency and the most likely combination at the same time, and calculate the joint design value of the reservoir at different return period levels. The invention has a strong statistical basis and provides a new approach for the staged design flood calculation under the climate change scenario.

The present invention relates to a radiative transfer calculation method suitable for continuous changes in cloud microphysical properties. Based on a second-flow approximation scheme, the method includes the following steps: Step 1) Expressing the radiant flux with continuous changes in cloud microphysical properties as a constant term and a disturbance term and; step 2) calculate the first radiant flux according to the traditional second-flow radiative transfer equation, and set the first radiant flux as the constant item; step 3) use the perturbation method to combine the asymmetry factor g and the unit Substituting the parameterized form of the scattering albedo ω into the traditional two-flow radiation transfer intensity equation to form a non-uniformity-induced disturbance term equation group, and by solving the non-uniformity-induced disturbance term equation group, two micro-disturbance factors and the expression The second radiant flux; step 4) summing the first radiant flux and the second radiant flux to complete the solution of the radiant flux with continuously changing microphysical properties of the cloud. This method effectively solves the vertical inhomogeneity of optical properties inside clouds in climate models, thereby improving the accuracy of radiation calculations.

The invention discloses a multi-climate model output result set correction method, and proposes a classification method of the climate model: for example, according to longitude and latitude, according to landform type, and the like. According to the grids divided into the research area, it is judged which one belongs to the classification; combined with the proposed multi-climate model output result set calculation formula, the parameters of the formula are obtained by fitting with the measured values, and finally the corrected predicted values ​​are obtained. This correction method makes full use of data information and comprehensively considers the combination of different classification climate models. Compared with the simulation results of the previous single-optimized climate model for grids, the simulated value is closer to the measured value and more in line with the actual situation.

The invention discloses a short-term climate prediction method based on statistical downscaling technology, which includes performing singular value decomposition on my country's monthly average precipitation and large-scale predictors within the modeling fitting period, and extracting my country's monthly average precipitation and large-scale forecasting factors respectively. The main modes of the scale predictors are typically coupled with spatially heterogeneous correlation distribution spatial patterns and the corresponding principal component time series pairs; using stepwise regression analysis, the adaptive main modes of the monthly mean precipitation and large-scale predictors suitable for the forecast period are obtained. Modal typical coupled spatially heterogeneous correlation profiles and principal component time series pairs. The advantages are: integrated climate model and previous key factors, based on multiple processing methods such as EOF, SVD and multiple linear regression to extract the optimal coupling variation among multiple factors, developed multi-source information dynamic processing and extraction technology, and constructed optimal information Integrating self-adaptive downscaling technology to realize the gridded prediction of the monthly average precipitation in different regions of the country under the climatic conditions.