Marine environment observation data quality control method based on disposable temperature-salinity-depth instrument
By implementing multi-step quality control on marine environmental observation data from the disposable CTD (conductivity, temperature, depth) instrument and using satellite remote sensing and ocean physical laws to correct data deviations, the inaccuracy problem of the disposable CTD observation data has been solved, and the reliability and accuracy of the data have been improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE PLA NAVY SUBMARINE INST
- Filing Date
- 2026-04-09
- Publication Date
- 2026-05-12
AI Technical Summary
The marine environmental observation data from disposable temperature, salinity, and depth (TDT) instruments have biases, which affect the understanding of natural change patterns and the accuracy of marine environmental forecasts, and therefore quality control is necessary.
Using marine satellite remote sensing data, measured seabed topographic data, and high-resolution marine reanalysis datasets, combined with marine physical laws and statistical analysis methods, the seawater temperature, salinity, and depth observation data of the discardable CTD meter are corrected. This includes steps such as range correction, surface transient response correction, peak detection, vertical gradient detection, systematic depth deviation correction, and statistical analysis constraint correction.
This effectively improves the reliability of marine environmental observation results from the discardable temperature, salinity, and depth instrument (TDI), reduces data bias, and enhances data quality.
Smart Images
Figure CN122019984A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of marine observation technology, specifically to a method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth (TDT) instrument. Background Technology
[0002] Marine environmental observation plays a vital role in marine scientific research and environmental forecasting. On the one hand, observation is an important means of discovering marine physical phenomena, understanding ocean changes, and determining marine climate effects. On the other hand, observation plays a key role in ocean simulation and forecasting, providing important support for improving model performance and forecasting skills.
[0003] The disposable CTD (Conductivity, Temperature, Depth) is an advanced marine environmental observation device, primarily used on mobile platforms (such as ships and aircraft). It can efficiently acquire environmental information such as seawater temperature, salinity, and depth. Compared to traditional fixed-station observation methods (such as buoy and mooring observations), it features real-time speed, low cost, and ease of use. It is now widely used in marine environmental observation and surveys, numerical forecasting, and other fields. For example, data acquired based on the disposable CTD accounts for 35% of the total data in the World Ocean Database (WOD18), which is also one of the important data sources driving the U.S. Navy's operational forecasting system.
[0004] Although disposable CTDs have high application value, their observation results are subject to certain deviations due to factors such as instrument manufacturing, deployment process, and marine environment. These deviations will further lead to biases in the understanding of natural change patterns and significant errors in marine environmental forecasting. Therefore, before using marine environmental observation data from disposable CTDs, data quality control is necessary. Summary of the Invention
[0005] This invention provides a quality control method for marine environmental observation data based on a disposable CTD (conductivity, temperature, and depth) instrument. The invention aims to correct the seawater temperature, salinity, and depth observation results based on the disposable CTD instrument. It utilizes marine satellite remote sensing data, measured seabed topography data, high-resolution marine reanalysis datasets, and seabed topography datasets, employing a standardized observation data quality control process combined with marine physical laws and statistical analysis methods to correct the seawater temperature, salinity, and depth observation data from the disposable CTD instrument. This effectively improves the reliability of marine environmental observation results from the disposable CTD instrument.
[0006] To achieve the above objectives, the technical solution of the present invention is as follows: A method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge includes the following steps: Step 1: Range correction of observation data; Based on the actual observation time and location of the disposable CTD (Conductivity, Temperature, Depth) instrument, and high-resolution ocean reanalysis dataset, determine the theoretical observation range of temperature and salinity observation data. Combined with the designed measurement range of temperature and salinity of the disposable CTD instrument, determine the effective range of temperature and salinity observation data. Based on the measured seabed topography data or high-resolution seabed topography dataset at the actual observation location of the disposable CTD instrument, determine the theoretical maximum value of depth observation data. Combined with the maximum measurement depth of the disposable CTD instrument, determine the effective range of depth observation data. Observation results exceeding the above temperature, salinity, and depth observation data ranges are considered outliers and are removed. Step 2: Correct the surface transient response of the observation data; set a depth threshold for surface transient response correction, and consider the observation data below the depth of the threshold as surface transient response outliers and remove them; Step 3: Perform peak detection correction on the observation data; calculate the peak detection value of the observation profile and set the peak detection threshold. Observation data with a depth higher than the threshold are regarded as peak detection outliers and removed. Step 4: Perform vertical gradient detection and correction on the observation data; calculate the vertical gradient of the observation profile and set the vertical gradient detection threshold. Observation data at depths higher than the threshold are regarded as vertical gradient detection outliers. Combine the vertical structure of temperature and salinity observation profiles corresponding to the marine environmental gridded climatological data at the same observation time and similar observation locations to remove vertical gradient detection outliers except for temperature and salinity strata. Step 5: Perform systematic depth bias correction on the observation data; based on the maximum measurement depth of the discardable temperature, salinity and depth meter, select the corresponding depth bias correction parameter scheme to eliminate systematic bias in the depth observation; Step 6: Perform marine physical law constraint correction on the observation data; based on the temperature, salinity, and depth data obtained from the discardable temperature, salinity, and depth instrument and the observation location information, calculate the density difference between adjacent water layers in the vertical direction, and consider the observation data with the upper layer density greater than the lower layer density as outliers and remove them; Step 7: Perform statistical analysis and constraint correction on the observation data; perform three-dimensional interpolation to align the temperature, salinity, and depth data obtained from the disposable CTD data with the marine environmental gridded climatological data at the same observation month and similar observation locations, calculate the correlation coefficient between the disposable CTD observation profile and the climatological data profile, and consider the observation data with an absolute value of the correlation coefficient below the threshold as outliers and remove them; further calculate the temperature and salinity anomalies and standard deviations of the disposable CTD data at adjacent times and locations at the same depth, and consider the observation data that deviates from the mean state by three times the standard deviation as outliers and remove them.
[0007] Preferably, in step 1, the marine satellite remote sensing data used includes sea surface temperature data (SST) and sea surface salinity data (SSS). The high-resolution marine reanalysis dataset used is one of the Chinese CORA2, American HYCOM, or European GLORYS12v1 marine reanalysis datasets. The measured seabed topography data used comes from a shipborne multibeam echo sounder, and the high-resolution seabed topography dataset used is the American GEBCO seabed topography dataset.
[0008] Preferably, in step 2, the depth threshold for correcting the surface transient response is set to 4 meters, and observations shallower than 4 meters are discarded.
[0009] Preferably, in step 3, the peak detection value is calculated as follows: ; In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths This represents the peak detection value calculated from temperature or salinity observations at three adjacent depth layers, with the depth difference between adjacent depth layers taken as 1 meter.
[0010] Preferably, in step 4, the vertical gradient is specifically calculated as follows: ; In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths Representing the The depth at each depth layer The vertical gradient is calculated from temperature or salinity observations at two adjacent depth layers; the depth difference between adjacent depth layers is taken as 3 meters; the gridded climatological data for the marine environment is selected from the World Oceans Dataset 2023.
[0011] Preferably, in step 5, the systematic bias of depth observation is corrected using the following equation: ; ; ; In the formula, This represents the observation time after the disposable temperature, salinity, and depth instrument (TDI) is submerged in water. This represents the depth observation results of the discardable temperature, salinity, and depth instrument after bias correction. This represents the average value of temperature profiles shallower than 500 meters obtained from discarded temperature, salinity, and depth gauge (TDT) observations. , , and These are the depth deviation correction parameters.
[0012] Preferably, in step 6, the density is calculated using the international thermodynamic seawater equation of state TEOS-10 standard. Specifically, the conservative temperature CT, absolute salinity SA, and pressure P are converted and calculated based on the temperature, salinity, and depth data and observation location information obtained from the discarded temperature, salinity, and depth instrument. The above conversion calculation results are then substituted into the thermodynamic seawater equation of state based on the TEOS-10 standard to calculate the seawater density.
[0013] Preferably, in step 7, the marine environmental gridded climatological data uses the World Oceans Dataset 2023 version, and the correlation coefficient calculation standard uses the Pearson correlation coefficient. The formula for calculating the correlation coefficient is as follows: ; In the formula, and These represent the observation profile results from a discarded temperature, salinity, and depth instrument (TTIMA) and the climatological data profile results, respectively. This represents the number of vertical depth layers, with the depth difference between adjacent depth layers set to 1 meter.
[0014] The marine environmental observation data quality control method based on the discardable temperature, salinity, and depth gauge of this invention has the following beneficial effects: This invention aims to correct the seawater temperature, salinity, and depth observation results based on a disposable CTD (conductivity, temperature, and depth) instrument. It utilizes marine satellite remote sensing data, measured seabed topography data, high-resolution marine reanalysis datasets, and seabed topography datasets, along with standardized observation data quality control procedures, combined with marine physical laws and statistical analysis methods, to correct the seawater temperature, salinity, and depth observation results of the disposable CTD instrument. This effectively improves the reliability of the marine environmental observation results from the disposable CTD instrument. Attached Figure Description
[0015] Figure 1 This is a schematic diagram of the implementation process of the present invention.
[0016] Figure 2 This is a graph showing the quality control results of temperature observation data in an embodiment of the present invention. Detailed Implementation
[0017] The following is a detailed description of the embodiments of the present invention in a step-by-step manner. The description takes the quality control process of environmental observation data from a discarded temperature, salinity, and depth instrument in a certain sea area in May 2024 as an example. The described embodiments are only some embodiments of the present invention, and not all embodiments.
[0018] Example 1: A method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth (TDT) instrument, such as... Figure 1 As shown, it includes the following steps: Step 1: Range correction of observation data; Based on the actual observation time and location of the disposable CTD (Conductivity, Temperature, Depth) instrument, and high-resolution ocean reanalysis dataset, determine the theoretical observation range of temperature and salinity observation data. Combine this with the designed measurement range of temperature and salinity of the disposable CTD instrument to determine the effective range of temperature and salinity observation data. Based on the measured seabed topography data or high-resolution seabed topography dataset at the actual observation location of the disposable CTD instrument, determine the theoretical maximum value of depth observation data. Combine this with the maximum measurement depth of the disposable CTD instrument to determine the effective range of depth observation data. Observation results exceeding the above temperature, salinity, and depth observation data ranges are considered outliers and are removed. Step 2: Correct the surface transient response of the observation data; Since there is a difference between the storage temperature of the disposable CTD (concentration, temperature, and depth) instrument before deployment and the surface seawater temperature, the instrument cannot fully adapt to the environmental changes in the early stage of water entry, and there is a certain response time, which leads to certain deviations in the temperature and salinity data measured in the short period of time after water entry; Set a surface transient response correction depth threshold, and regard the observation data below the depth of the threshold as surface transient response anomalies and remove them; Step 3: Perform peak detection correction on the observation data; During the use of the disposable temperature, salinity and depth meter, there may be problems with poor wire insulation, which will lead to current leakage and cause obvious unreasonable data abrupt changes at a certain depth of the observation profile, i.e., peaks; Calculate the peak detection value of the observation profile and set the peak detection threshold. Observation data at depths higher than the threshold are regarded as peak detection anomalies and are removed. Step 4: Perform vertical gradient detection and correction on the observation data; Since discardable temperature, salinity, and depth meters may malfunction during use, resulting in significant depth differences between adjacent vertical observation points, this will cause drastic changes in the temperature and salinity observation results of adjacent vertical observation points; calculate the vertical gradient of the observation profile and set a vertical gradient detection threshold. Observation data at depths higher than this threshold are considered vertical gradient detection outliers. Combine the vertical structure of the temperature and salinity observation profiles corresponding to the marine environmental gridded climatological data at the same observation time and similar observation locations, and remove vertical gradient detection outliers other than temperature and salinity strata; Step 5: Perform systematic depth bias correction on the observation data; Since most disposable CTDs do not have pressure sensors, their depth observation results are calculated by estimating the descent rate equation based on empirical parameters, which leads to a certain systematic bias in the results; Based on the maximum measurement depth of the disposable CTD, select the corresponding depth bias correction parameter scheme to eliminate the systematic bias in the depth observation. Step 6: Perform marine physical law constraint correction on the observation data; Ocean density is related to temperature, salinity, and pressure, mainly showing that the density increases with increasing depth. Based on the temperature, salinity, and depth data and observation location information obtained from the discardable temperature, salinity, and depth meter, calculate the density difference between adjacent water layers vertically. Observation data with the upper layer density greater than the lower layer density are regarded as outliers and removed. Step 7: Perform statistical analysis and constraint correction on the observation data. Ocean temperature and salinity profiles under similar observation time and location conditions have a certain similarity. The temperature, salinity, and depth data obtained by the disposable CTD (Conductivity, Temperature, Depth) instrument are three-dimensionally interpolated and aligned with the marine environmental gridded climatological data at the same observation month and similar observation location. The correlation coefficient between the disposable CTD observation profile and the climatological data profile is calculated. Observation data with an absolute value of the correlation coefficient below the threshold are regarded as outliers and removed. Further, the anomalies and standard deviations of temperature and salinity observed by disposable CTD instruments at adjacent times and locations at the same depth are calculated. Observation data that deviate from the mean by three times the standard deviation are regarded as outliers and removed.
[0019] Example 2: Based on Example 1, this example discloses that: in step 1, the discardable temperature, salinity, and depth (TDT) meter used is the XCTD-4 from TSK Corporation of Japan, with a maximum measurement depth of 1850 meters. The actual observation time was May 12, 2024, and the actual observation location was the central sea area of a certain sea area. The marine satellite remote sensing data used included sea surface temperature data (SST) from OSTIA and sea surface salinity data (SSS) from SMOS. The high-resolution ocean reanalysis dataset used was the European GLORYS12v1 ocean reanalysis dataset. The measured seabed topography data used came from the shipborne multibeam echo sounder on the research vessel "Xiangyanghong 52". The high-resolution seabed topography dataset used was the US GEBCO_2023 seabed topography dataset.
[0020] Example 3: Based on Example 1 or Example 2, this example discloses that in step 2, the surface transient response correction depth threshold is set to 4 meters, and observation results shallower than 4 meters are discarded.
[0021] Example 4: Based on Example 1, Example 2, or Example 3, this example discloses that in step 3, the temperature peak detection threshold is set to 2℃ / m and the salinity peak detection threshold is set to 0.3psu / m. Observation results with peak detection values exceeding these thresholds are discarded. The specific calculation method for the peak detection value is as follows: ; In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths This represents the peak detection value calculated from temperature or salinity observations at three adjacent depth layers; the depth difference between adjacent depth layers is taken as 1 meter.
[0022] Example 5: Based on Example 1, Example 2, Example 3, or Example 4, this example discloses that in step 4, the temperature vertical gradient detection threshold is set to 0.7℃ / m, and the salinity vertical gradient detection threshold is set to 9.0 psu / m; the temperature and salinity vertical gradients observed by the discardable temperature-salinity-depth meter are calculated, and combined with the gridded climatological data of the marine environment, outliers in the vertical gradient detection, except for the mesospheric layer, are removed. The specific calculation method for the vertical gradient is as follows: ; In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths Representing the The depth at each depth layer The vertical gradient is calculated from temperature or salinity observations at two adjacent depth layers, with a depth difference of 3 meters between adjacent depth layers; the gridded climatological data for the marine environment are selected from the World Oceans Dataset 2023.
[0023] Example 6, based on Example 1, Example 2, Example 3, Example 4, or Example 5, discloses that in step 5, the systematic bias of depth observation is corrected using the following equation: ; ; ; In the formula, This represents the observation time after the disposable temperature, salinity, and depth instrument (TDI) is submerged in water. This represents the depth observation results of the discardable temperature, salinity, and depth instrument after bias correction. This represents the average value of temperature profiles shallower than 500 meters obtained from discarded temperature, salinity, and depth gauge (TDT) observations. , , and These are depth deviation correction parameters; the discardable temperature, salinity, and depth gauge used (TSK XCTD-4, Japan) has a maximum measurement depth of 1850 meters. , , and The values are 3.418, 1.10e-3, 2.63e-4, and 2.31e-6, respectively.
[0024] Example 7: Based on Example 1, Example 2, Example 3, Example 4, Example 5, or Example 6, this example discloses that: in step 6, the density is calculated using a new, high-precision international thermodynamic seawater equation of state (TEOS-10) standard. Specifically, the conservative temperature (CT), absolute salinity (SA), and pressure (P) are calculated based on the temperature, salinity, and depth data and observation location information obtained from the discarded temperature, salinity, and depth instrument. The results of the above conversion calculations are then substituted into the thermodynamic seawater equation of state based on the TEOS-10 standard to calculate the seawater density. The relevant calculations of the TEOS-10 standard are implemented through a software function library of standard algorithms (such as the Gibbs SeaWater function library).
[0025] Example 8, based on Example 1, Example 2, Example 3, Example 4, Example 5, Example 6, or Example 7, discloses that: in step 7, the latest World Oceans Dataset 2023 (WOA23) is used for the gridded climatological data of the marine environment; the Pearson correlation coefficient is used as the standard for calculating the correlation coefficient; the absolute value threshold of the correlation coefficient is set to 0.8; and the formula for calculating the correlation coefficient is: ; In the formula, and These represent the observation profile results from a discarded temperature, salinity, and depth instrument (TTIMA) and the climatological data profile results, respectively. The vertical depth stratification is determined by the actual depth range of the sea area observed by the discardable CTD (conductivity, temperature, depth) instrument (32 to 48 meters), with each stratification having a resolution of 1 meter.
[0026] Figure 2 The results of temperature observation data quality control in the embodiments of the present invention are presented. The results show that: using the observation data quality control method proposed in this invention, the uncertainty of the ocean surface temperature observation results caused by the transient response after the instrument enters the seawater is well corrected, and the false results of the ocean bottom temperature observation caused by the continued increase of the depth observation results after the instrument contacts the seabed are also well corrected; in addition, the observation results are consistent with the laws of ocean physics and statistical analysis characteristics, such as the high temperature of the upper ocean layer, the low temperature of the lower ocean layer, and the depth of the ocean temperature jump layer at the observation location is about 15 to 20 meters.
Claims
1. A method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth (TDT) instrument, characterized by: Includes the following steps: Step 1: Range correction of observation data; Based on the actual observation time and location of the disposable CTD (Conductivity, Temperature, Depth) instrument, and high-resolution ocean reanalysis dataset, determine the theoretical observation range of temperature and salinity observation data. Combined with the designed measurement range of temperature and salinity of the disposable CTD instrument, determine the effective range of temperature and salinity observation data. Based on the measured seabed topography data or high-resolution seabed topography dataset at the actual observation location of the disposable CTD instrument, determine the theoretical maximum value of depth observation data. Combined with the maximum measurement depth of the disposable CTD instrument, determine the effective range of depth observation data. Observation results exceeding the above temperature, salinity, and depth observation data ranges are considered outliers and are removed. Step 2: Correct the surface transient response of the observation data; set a depth threshold for surface transient response correction, and consider the observation data below the depth of the threshold as surface transient response outliers and remove them; Step 3: Perform peak detection correction on the observation data; calculate the peak detection value of the observation profile and set the peak detection threshold. Observation data with a depth higher than the threshold are regarded as peak detection outliers and removed. Step 4: Perform vertical gradient detection and correction on the observation data; calculate the vertical gradient of the observation profile and set the vertical gradient detection threshold. Observation data at depths higher than the threshold are regarded as vertical gradient detection outliers. Combine the vertical structure of temperature and salinity observation profiles corresponding to the marine environmental gridded climatological data at the same observation time and similar observation locations to remove vertical gradient detection outliers except for temperature and salinity strata. Step 5: Perform systematic depth bias correction on the observation data; based on the maximum measurement depth of the discardable temperature, salinity and depth meter, select the corresponding depth bias correction parameter scheme to eliminate systematic bias in the depth observation; Step 6: Perform marine physical law constraint correction on the observation data; based on the temperature, salinity, and depth data obtained from the discardable temperature, salinity, and depth instrument and the observation location information, calculate the density difference between adjacent water layers in the vertical direction, and consider the observation data with the upper layer density greater than the lower layer density as outliers and remove them; Step 7: Perform statistical analysis and constraint correction on the observation data; perform three-dimensional interpolation to align the temperature, salinity, and depth data obtained from the disposable CTD data with the marine environmental gridded climatological data at the same observation month and similar observation locations, calculate the correlation coefficient between the disposable CTD observation profile and the climatological data profile, and consider the observation data with an absolute value of the correlation coefficient below the threshold as outliers and remove them; further calculate the temperature and salinity anomalies and standard deviations of the disposable CTD data at adjacent times and locations at the same depth, and consider the observation data that deviates from the mean state by three times the standard deviation as outliers and remove them.
2. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that: In step 1, the ocean satellite remote sensing data used include sea surface temperature data (SST) and sea surface salinity data (SSS). The high-resolution ocean reanalysis dataset used is one of the Chinese CORA2, American HYCOM, or European GLORYS12v1 ocean reanalysis datasets. The measured seabed topography data used comes from a shipborne multibeam echo sounder, and the high-resolution seabed topography dataset used is the American GEBCO seabed topography dataset.
3. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 2, the depth threshold for correcting the transient response of the surface layer is set to 4 meters, and observations shallower than 4 meters are discarded.
4. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 3, the peak detection value is calculated as follows: ； In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths This represents the peak detection value calculated from temperature or salinity observations at three adjacent depth layers, with the depth difference between adjacent depth layers taken as 1 meter.
5. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 4, the vertical gradient is specifically calculated as follows: ； In the formula, Represents the location of the depth layer. Representing the Temperature or salinity observations at different depths Representing the The depth at each depth layer The vertical gradient is calculated from temperature or salinity observations at two adjacent depth layers; the depth difference between adjacent depth layers is taken as 3 meters; the gridded climatological data for the marine environment is selected from the World Oceans Dataset 2023.
6. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 5, the systematic bias of the depth observation is corrected using the following equation: ； ； ； In the formula, This represents the observation time after the disposable temperature, salinity, and depth instrument (TDI) is submerged in water. This represents the depth observation results of the discardable temperature, salinity, and depth instrument after bias correction. This represents the average value of temperature profiles shallower than 500 meters obtained from discarded temperature, salinity, and depth gauge (TDT) observations. , , and These are the depth deviation correction parameters.
7. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 6, the density is calculated using the international thermodynamic seawater equation of state TEOS-10 standard. Specifically, the conservative temperature CT, absolute salinity SA, and pressure P are calculated based on the temperature, salinity, and depth data and observation location information obtained from the discarded temperature, salinity, and depth instrument. The results of the above conversion calculations are then substituted into the thermodynamic seawater equation of state based on the TEOS-10 standard to calculate the seawater density.
8. The method for quality control of marine environmental observation data based on a discardable temperature, salinity, and depth gauge as described in claim 1, characterized in that, In step 7, the marine environmental gridded climatological data uses the World Oceans Dataset 2023 version, and the correlation coefficient is calculated using the Pearson correlation coefficient. The formula for calculating the correlation coefficient is as follows: ； In the formula, and These represent the observation profile results from a discarded temperature, salinity, and depth instrument (TTIMA) and the climatological data profile results, respectively. This represents the number of vertical depth layers, with the depth difference between adjacent depth layers set to 1 meter.