A Remote Sensing Inversion Method for Ocean Heat Content Combining Remote Sensing and Buoy Data

Abstract

The invention relates to a remote sensing inversion method of ocean heat content combined with remote sensing and buoy data, comprising the following steps: obtaining global ocean multi-source sea surface remote sensing observation data and Argo measured grid data and performing preprocessing to obtain sea surface data, time and space parameters and ocean interior heat content data; according to Argo’s measured coordinate data, the ocean interior heat content data of each grid point is in one-to-one correspondence with the sea surface remote sensing and spatiotemporal parameters input into the model, which are respectively marked as feature matrix X and label matrix Y; X and Y are vertically merged and divided into training data set, verification data set and test data set according to time series; the training data set is trained based on the long short-term memory neural network LSTM time series deep learning method, and the loss function of the data set The optimal network depth and parameters are selected, and the inversion model is established; the historical long-term time-series feature matrix X' is obtained, and used as the input data of the model to further reconstruct the ocean internal heat content OHC data set.

Description

technical field [0001] The invention relates to the field of remote sensing information processing and application, in particular to a remote sensing inversion method for ocean heat content combined with remote sensing and buoy data. Background technique [0002] In the past ten years, remote sensing technology has been applied in the ocean field with remarkable results. However, remote sensing observations are limited to the ocean surface, and cannot directly detect key dynamic information inside the ocean. At the same time, the internal dynamical process of the ocean is complex, with multi-dimensional, multi-scale, dynamic, and time-series dependencies. Ocean interior observation plays an important role in understanding the dynamic change mechanism of the ocean system and analyzing the heat change and redistribution of the global ocean. Although the existing internal ocean buoy observation data Argo data has high accuracy, it is sparse and limited, the global distribution...

AI Technical Summary

Problems solved by technology

Although the existing internal ocean buoy observation data Argo data has high accuracy, it is sparse and limited, with less than 4000 global distributions, and the temporal and spatial resolution is low, which is far from meeting the needs of multi-dimensional and multi-scale ocean observations

Images