A method for ocean temperature and salinity horizontal gridding considering continuous variation of element field

Abstract

The application discloses a kind of ocean temperature and salt horizontal gridding method considering continuous change of element field, comprising the following steps: step one: search the observation data of corresponding vertical level in a certain space-time range for each horizontal grid in the vertical each standard layer of whole sea area;Step two: calculate the mean of temperature and salinity;Step three: calculate the standard deviation of temperature and salinity;The beneficial effects of the application are that: the application considers the continuous change factor of element field into the process of horizontal gridding calculation, and adjusts the search time range when searching observation data, ensures that deep gridding calculation can also have as many observation data as possible, which is beneficial to the gridding calculation of temperature and salinity elements.

Description

Technical Field

[0001] This invention belongs to the field of continuous variation of element fields, and specifically relates to a gridding method for ocean temperature and salinity levels that takes into account continuous variation of element fields. Background Technology

[0002] With the advancement of science and technology, international marine meteorological and hydrological environmental data have increased significantly, enabling the development of higher-resolution global climate statistical analysis products. Moreover, previous statistical analysis results can no longer accurately reflect the true state of the current climate cycle. Furthermore, with the upgrading of computer hardware, computing power has been greatly enhanced, enabling the application of the continuous changing characteristics of element fields in statistical analysis research. Summary of the Invention

[0003] The purpose of this invention is to provide a horizontal gridding method for ocean temperature and salinity that takes into account continuous changes in the element field. This method incorporates the continuous changes in the element field into the horizontal gridding calculation process and adjusts the search time range with reference to depth when searching for observation data, so as to ensure that deep gridding calculations can also obtain as much observation data as possible.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for gridding ocean temperature, salinity, and horizontal parameters considering continuous changes in the element field, comprising the following steps:

[0005] Step 1: Search for observation data of the corresponding vertical layers within a certain time and space range for each standard layer of each horizontal grid in the entire sea area.

[0006] Step 2: Calculate the average values ​​of temperature and salinity;

[0007] Step 3: Calculate the standard deviation of temperature and salinity.

[0008] As a preferred technical solution of the present invention, when searching for observation data in the time dimension, the search time range is different at different depths.

[0009] As a preferred technical solution of the present invention, the number of observations in each depth time window is balanced, and the near-surface time window is defined as 22.5 days before and after the center day of each month.

[0010] As a preferred technical solution of the present invention, in step two, the minimum slope technique is adopted, which obtains the temperature and salinity result of the horizontal grid point vertically to a specific standard layer by minimizing the following interpolation equation containing data error terms and relaxation terms.

[0011] As a preferred technical solution of the present invention, in step three, the calculated monthly average climate values ​​of temperature and salinity are interpolated to the time and location of the observation to obtain the climate average value of the observation point. The observed value is then subtracted from the interpolated climate average value of the observation point to obtain the observed outlier value.

[0012] As a preferred technical solution of the present invention, mesh values ​​are generated on a constant depth surface.

[0013] Compared with the prior art, the beneficial effects of the present invention are:

[0014] This invention incorporates continuously changing factors of the element field into the horizontal gridded calculation process, and adjusts the search time range with reference to depth when searching for observation data, so as to ensure that deep gridded calculations can also have as much observation data as possible, which is beneficial to the gridded calculation of temperature and salinity elements. Attached Figure Description

[0015] Figure 1 This is a schematic diagram showing the location of the analysis grid and surrounding observations in this invention;

[0016] Figure 2 This is a flowchart of the meshing method of the present invention;

[0017] Figure 3 This is a schematic diagram of the monthly average distribution structure of surface ocean temperature in the Northwest Pacific Ocean in January, according to the present invention.

[0018] Figure 4 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 10 meters in January in the Northwest Pacific Ocean region, as per the present invention.

[0019] Figure 5 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 50 meters in January in the Northwest Pacific Ocean region according to the present invention.

[0020] Figure 6 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 100 meters in January in the Northwest Pacific Ocean region, according to the present invention.

[0021] Figure 7 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 200 meters in January in the Northwest Pacific Ocean region, according to the present invention.

[0022] Figure 8 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 500 meters in January in the Northwest Pacific Ocean region, according to the present invention.

[0023] Figure 9 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 1000 meters in January in the Northwest Pacific Ocean region, according to the present invention.

[0024] Figure 10This is a schematic diagram of the monthly average distribution structure of ocean temperature at 2000 meters in January in the Northwest Pacific Ocean region, according to the present invention.

[0025] Figure 11 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 3600 meters in the Northwest Pacific Ocean in January, according to the present invention.

[0026] Figure 12 This is a schematic diagram of the monthly average distribution structure of ocean temperature at 5000 meters in the Northwest Pacific Ocean in January, according to the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Please see Figure 1 and Figure 2 This invention provides a method for gridding ocean temperature and salinity levels that considers continuous changes in the element field, comprising the following steps:

[0029] Step 1: Search for observational data of the corresponding vertical layers within a certain spatiotemporal range for each horizontal grid and standard layer across the entire sea area; denot the central day of each month as d. m d m = (m-0.5)*365.25 / 12, m=1,2,...,12; Table 1 lists the central day of each month;

[0030] Table 1. Central day of each month

[0031] 2 45.7 6 167.4 10 289.2 3 76.1 7 197.8 11 319.6 4 106.5 8 228.3 12 350.0

[0032] When searching for observational data in the time dimension, the search time range varies at different depths. To ensure that more observational data are collected at deeper depths, the time window increases with increasing depth. The number of observational data decreases with increasing depth. To balance the number of observations in the time windows at different depths, the near-surface time window is defined as 22.5 days before and after the center day of each month. The deeper you go, the fewer the observations, and the larger the time window becomes. See Table 2 for details. Observational data is searched within a 0.25°×0.25° square area on the horizontal plane.

[0033] Table 2. Time windows (days) for search observations at different depths.

[0034] 1 0-200 <![CDATA[45 days, i.e., [d m -22.5, d m +22.5] days]]> 2 220-400 <![CDATA[60 days, i.e., [d m -30, d m +30] days]]> 3 500-800 <![CDATA[90 days, i.e., [d m -45, d m +45] days]]> 4 900-1200 <![CDATA[120 days, i.e., [d m -60, d m +60] days]]> 5 1300-6600 <![CDATA[366 days, i.e., [d m -183, d m +183] days]]>

[0035] Step 2: Calculate the mean values ​​of temperature and salinity; employ the minimum slope technique, which obtains the temperature and salinity results of a specific standard layer vertically from a given horizontal grid point by minimizing the following interpolation equation (objective functional) containing data error terms and relaxation terms (slope terms):

[0036]

[0037] In the summation formula on the right, the first term within the curly braces is the square of the first derivative of the grid value along the latitude direction, and the second term is the square of the first derivative of the grid value along the longitude direction; these two terms are relaxation terms. The third term is the square of the difference between the observed value and the grid value, i.e., the error term. (n, m) is the horizontal grid index, where n is the east-west index and m is the south-north index; T n，m This refers to the grid temperature at position (n, m) (hereafter, temperature will be used as an example; the salinity is handled in the same way, simply by converting all temperature variables T to salinity variables), θ n，m，k It is the k-th temperature observation within a range of 0.5 grid lengths around the (n, m) grid point; the meridional length of the grid is Δy (derived from the product resolution, Δy = 0.25°, i.e., 27.7 km), and the zonal length is Δx. m =Δycos(λ) m ), λ m It refers to latitude in radians, Δx m It varies with latitude, becoming shorter closer to the poles; F′ represents a factor balancing smoothness and resolution. Figure 1 A schematic diagram showing the location of the analysis grid relative to surrounding observations is provided;

[0038] Each term in equation (1) is multiplied by (Δy). 2 This yields the objective functional ∫ that needs to be optimized and solved subsequently.

[0039] J = J′ × Δy 2 F = F′ × Δy 2

[0040]

[0041] Here, we take F = 1;

[0042] In order to obtain T in equation (2) that minimizes J, n，m The analytical value, let The solution is obtained using the Gauss-Seidel iterative method.

[0043] The Gauss-Seidel iteration is an iterative method in numerical linear algebra used to solve approximate values ​​of systems of linear equations. Similar to the Jacobi method, the Gauss-Seidel iteration is based on the principle of matrix factorization. Its solution steps are as follows:

[0044] Assume the system of linear equations is

[0045] a i1 x1+a i2 x2 + ... + a in x n =b i (i = 1, 2, ..., n)

[0046] The iterative formula for the Gauss-Seidel iterative method is as follows:

[0047]

[0048] Here it is assumed that a ii ≠0. In many cases, the Gauss-Seidel iterative method converges faster than the simple iterative method. Its difference from the simple method lies in the computation... At that time, the newly iterated version was utilized. When the value of is strictly diagonally dominant or symmetric positive definite, the Gauss-Seidel iteration method will definitely converge; at the land boundary, a 0-gradient boundary condition is used to eliminate the landing point grid and the cross-land grid, but the influence of the observation points around the boundary is still considered; the so-called 0-gradient boundary condition means that there is no extraction at the region boundary.

[0049] The third step is to calculate the standard deviation of temperature and salinity, using a method similar to that used for calculating the mean. First, the calculated monthly average climatic mean of temperature and salinity is interpolated to the time and location of the observation to obtain the climatic mean at the observation point. The interpolated climatic mean at the observation point is then subtracted from the observed value to obtain the outlier. Finally, the minimum slope technique is applied to solve for σ. 2 The cost function is

[0050]

[0051] Here, F = 1 / 15. It is the climate mean at the time and location of the observation point, interpolated from the monthly average grid climate field. In order to obtain σ in equation (3) that minimizes J, 2 ,make You can get Finally, calculate the standard deviation.

[0052] To make full use of the observation data (especially the large number of XBT profiles with only temperature observations) and to achieve gridding in shallow water areas, the scheme chooses to generate grid values ​​on a steady depth surface.

[0053] To address the issue that the amount of observational data decreases with increasing depth, different search time ranges are selected for different depths when searching for observational data. Starting from the surface and moving downwards, the time window is larger as the depth increases, thus ensuring that there is still sufficient observational data at greater depths. During horizontal gridded calculations, factors balancing smoothness and resolution are set based on the distribution of observation points, taking into account the continuous changes in the element field into the horizontal gridded calculations, and solving for temperature and salinity elements using the minimum slope technique.

[0054] The ocean temperature gridded product for the Northwest Pacific Ocean developed according to this invention has a resolution of 0.25° and can still show relatively continuous performance in ocean areas with greater depths. Figures 3-12 It displays the monthly average climatological distribution of ocean temperature in the Northwest Pacific Ocean in January at depths of 10 meters, 50 meters, 100 meters, 200 meters, 500 meters, 1000 meters, 2000 meters, 3600 meters, and 5000 meters. Compared with previous atlas products, the climatological ocean temperature product for the Northwest Pacific Ocean developed according to this invention can still show a relatively fine spatial distribution of ocean temperature below 200 meters in water depth.

[0055] Although embodiments of the invention have been shown and described in detail above, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for ocean temperature and salinity horizontal gridding considering continuous variation of element field, characterized in that: Includes the following steps: Step 1: Search for observation data of the corresponding vertical layer within a certain spatiotemporal range for each standard layer of each horizontal grid in the entire sea area; when searching for observation data in the time dimension, the search time range is different at different depths; balance the number of observations in the time window at each depth, and define the near-surface time window as 22.5 days before and after the center day of each month; Step 2: Calculate the mean values ​​of temperature and salinity; use the minimum slope technique, which obtains the temperature and salinity results of the horizontal grid point vertically to a specific standard layer by minimizing the following interpolation equation containing data error terms and relaxation terms; ， In the summation formula on the right, the first term within the curly braces is the square of the first derivative of the grid value along the latitude direction, and the second term is the square of the first derivative of the grid value along the longitude direction; these two terms are relaxation terms. The third term is the square of the difference between the observed value and the grid value, i.e., the error term. (n, m) is the horizontal grid index, where n is the east-west index and m is the south-north index. This refers to the grid temperature at position (n, m). We will use temperature as an example from now on. The salinity is handled in the same way; simply convert all temperature variables T to salinity variables. It is the k-th temperature observation within a range of 0.5 grid lengths around the (n, m) grid point; the meridional length of the grid is Based on product resolution =0.25°, which is 27.7km. The latitudinal length is taken as... , It refers to the radians of latitude. It varies with latitude; the closer to the poles, the shorter the distance. A factor representing the balance between smoothness and resolution; Step 3: Calculate the standard deviation of temperature and salinity; use the calculated monthly average climate values ​​of temperature and salinity to interpolate them to the time and location of the observation to obtain the climate mean of the observation point; subtract the interpolated climate mean of the observation point from the observed value to obtain the observed outlier value.

2. The ocean temperature, salinity, and horizontal gridding method considering continuous changes in the element field according to claim 1, characterized in that: Generate mesh values ​​on a constant depth surface.

Images