Remote sensing estimation method and device for primary productivity of marine phytoplankton fraction

Abstract

The invention discloses a remote sensing estimation method and a remote sensing estimation device for the primary productivity of marine phytoplankton in size fraction, and the method comprises the steps: building an estimation function of the photosynthetic quantum yield of the phytoplankton in the corresponding size fraction based on the relation between the light intensity and the photosynthetic quantum yield of the phytoplankton in the single size fraction; calculating the net primary productivity of the phytoplankton of the corresponding size fraction based on the photosynthetic quantum yield and the average light absorption coefficient of the phytoplankton of the single size fraction under the scale of the light intensity duration and the ocean depth measurement range; and adding the net primary productivity of the phytoplankton in the plurality of size fractions, and calculating to obtain the total primary productivity of the phytoplankton. By adopting the technical scheme, in a framework of light absorption and conversion of the phytoplankton, the light absorption coefficient of the phytoplankton with different size fractions and the size fraction difference of the photosynthetic quantum yield are considered at the same time, an effective scheme for quantitative expression estimation of the phytoplankton is established, the estimation error of the primary productivity caused by neglecting the size fraction difference can be effectively reduced, and the estimation accuracy of the primary productivity is improved. And the estimation accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ocean remote sensing, and particularly relates to a method and device for estimating ocean phytoplankton size-fractioned primary productivity by remote sensing. BACKGROUND

[0002] Net primary productivity (NPP) of the ocean refers to the total amount of organic carbon fixed by autotrophs (mainly phytoplankton or terrestrial plants) through photosynthesis per unit of time, minus the amount of organic carbon remaining after autotrophs' own respiration consumption.

[0003] Phytoplankton is divided into three grades according to particle size, i.e., small-sized phytoplankton (micro, >20 µm), mainly including diatoms and dinoflagellates; micro-sized phytoplankton (nano, 2~20 µm), mainly composed of chrysophytes, cryptophytes and green algae; and pico-sized phytoplankton (pico, <2 µm), mainly including cyanobacteria. Different particle sizes of phytoplankton have obvious differences in response to environmental changes, and have obvious different physiological processes and ecological functions. Therefore, the primary productivity information of size-fractioned phytoplankton is of great significance.

[0004] In the prior art, there are various schemes for estimating ocean net primary productivity based on remote sensing data, but these schemes have similar problems. These schemes are mainly based on photosynthesis-related parameters for estimation, but do not consider the population differences of phytoplankton photosynthesis process, which undoubtedly causes errors in the estimation of primary productivity. More importantly, these models can only estimate the total primary productivity (i.e., the sum of primary productivity of all particle sizes of phytoplankton) at one time, and cannot estimate the size-fractioned primary productivity information more finely. SUMMARY

[0005] The present application provides a method and device for estimating ocean phytoplankton size-fractioned primary productivity by remote sensing, and aims to solve the technical problems in the prior art that the size-fractioned primary productivity information of phytoplankton cannot be estimated finely, and the estimation error is large.

[0006] Technical Solution: This invention provides a remote sensing estimation method for primary productivity of marine phytoplankton at different particle sizes, comprising: calculating the average light absorption coefficient of phytoplankton at a corresponding particle size using the specific absorption coefficient of a single particle size phytoplankton and the corresponding chlorophyll a concentration; establishing an estimation function for the photosynthetic quantum yield of phytoplankton at a corresponding particle size based on the relationship between light intensity and the photosynthetic quantum yield of a single particle size phytoplankton; obtaining the relationship between light intensity and the photosynthetic quantum yield of phytoplankton through fitting matched observation data and remote sensing data; calculating the net primary productivity of phytoplankton at a corresponding particle size based on the photosynthetic quantum yield and average light absorption coefficient of a single particle size phytoplankton at a scale of light intensity duration and ocean depth measurement range; and summing the net primary productivity of phytoplankton at multiple particle size sizes to calculate the total primary productivity of phytoplankton.

[0007] Specifically, the matching process between observational and remote sensing data includes: acquiring observational data including net primary productivity (NPMP) and total NMPMP of phytoplankton at multiple ocean depths and multiple particle sizes, as well as depth integral NMPMP; acquiring remote sensing data including ocean surface chlorophyll a concentration, phytoplankton light absorption coefficient, photosynthetically active radiation (PADR), and ocean water mixing layer depth; and matching observational and remote sensing data based on preset hourly time windows and pixel spatial windows, with the pixel spatial windows having a coefficient of variation threshold and an effective pixel count threshold.

[0008] Specifically, within the 400nm to 700nm wavelength range, the average light absorption coefficient is calculated based on the product of the specific absorption coefficient and the chlorophyll a concentration, using the following calculation formula: a' ph,i (z)=(∫ 400 700 Chla i (z)×a* ph,i (λ)dλ) / (700-400), Among them, a' ph,i (z) represents the average light absorption coefficient of phytoplankton of particle size i at ocean depth z, Chla i (z) represents the chlorophyll a concentration of phytoplankton of particle size i at ocean depth z, a* ph,i (λ) represents the specific absorption coefficient of phytoplankton of particle size i at wavelength λ.

[0009] Specifically, the chlorophyll a concentration is calculated as follows: the chlorophyll a concentration in the ocean surface layer, as well as the depth of the ocean euphotic layer and the depth of the ocean water mixing layer, are obtained to estimate the total chlorophyll a concentration at ocean depth z. Then, using a preset particle size division model, the chlorophyll a concentration of phytoplankton at a single particle size is obtained.

[0010] Specifically, based on the relationship between light intensity and phytoplankton photosynthetic quantum yield, as well as the maximum photosynthetic quantum yield of phytoplankton, an estimation function for the photosynthetic quantum yield of a single-particle-size phytoplankton is established.

[0011] Specifically, the estimation function for the photosynthetic quantum yield of phytoplankton at the corresponding particle size is as follows: φ i (z)=(K φ,i ×φ i,max ) / (K φ,i +E(z,t)), Where, φ i (z) represents the photosynthetic quantum yield of phytoplankton of particle size i at ocean depth z, K φ,i φ represents the light intensity required for phytoplankton of particle size i to reach half of its maximum photosynthetic quantum yield. i,max E(z,t) represents the maximum photosynthetic quantum yield of phytoplankton at particle size i, and E(z,t) represents the light intensity at ocean depth z at time t.

[0012] Specifically, particle size i includes small m, micro n, and picomotor p, where K φ,m K φ,n and K φ,p 3.5074Einm respectively -2 d -1 1.5619Einm -2 d -1 and 4.1241Einm -2 d -1 φ m,max φ n,max and φ p,max 0.0799 mol CEin -1 0.0760 mol CEin -1 and 0.0374 molCEin -1 .

[0013] Specifically, the net primary productivity of phytoplankton at the corresponding particle size is calculated based on the product of light intensity, photosynthetic quantum yield, and average light absorption coefficient.

[0014] Specifically, the total primary productivity of the phytoplankton is calculated using the following formula: NPP=Σ i=1 3 NPP i =∫ t=0 D ∫ z=0 Zeu E(z,t)a' ph,i (z)φ i (z)dzdt, Wherein, NPP represents total primary productivity of phytoplankton, NPP i denoted as net primary productivity of phytoplankton at particle size i, D represents the duration of light intensity, and Zeu represents the range of ocean depth measurement.

[0015] This invention also provides a remote sensing estimation device for phytoplankton primary productivity at different particle sizes, comprising: a light absorption coefficient calculation unit, a photosynthetic quantum yield estimation unit, and a phytoplankton primary productivity calculation unit at different particle sizes. The light absorption coefficient calculation unit calculates the average light absorption coefficient of phytoplankton at a given particle size using the specific absorption coefficient of a single particle size and the corresponding chlorophyll a concentration. The photosynthetic quantum yield estimation unit establishes an estimation function for the photosynthetic quantum yield of phytoplankton at a given particle size based on the relationship between light intensity and the photosynthetic quantum yield of a single particle size. The relationship between light intensity and the photosynthetic quantum yield of phytoplankton is obtained by fitting matched observation data and remote sensing data. The phytoplankton primary productivity calculation unit calculates the net primary productivity of phytoplankton at a given particle size, based on the photosynthetic quantum yield and average light absorption coefficient of a single particle size, within a range of light intensity duration and ocean depth measurement. The net primary productivity of phytoplankton at multiple particle sizes is then summed to calculate the total primary productivity of phytoplankton.

[0016] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: It can obtain the primary productivity of small, micro and picoplankton, respectively, enriching the connotation information of primary productivity products; within the framework of phytoplankton light absorption and transformation, it simultaneously considers the particle size differences of light absorption coefficient and photosynthetic quantum yield of phytoplankton of different particle sizes, and establishes an effective scheme for their quantitative expression and estimation, which can effectively reduce the estimation error of primary productivity caused by ignoring particle size differences and improve the estimation accuracy. Attached Figure Description

[0017] Figure 1 A schematic diagram illustrating the framework of the remote sensing estimation method for phytoplankton fractional primary productivity provided by this invention; Figure 2 This is a comparative diagram of the observation data and model estimation provided by the present invention, with the comparison object being primary productivity at different granularities; Figure 3 This is a comparative diagram of the observation data and model estimation provided by the present invention, with the comparison object being total primary productivity; Figure 4 This is a comparative diagram of the observation data and model estimation provided by the present invention, with the comparison object being the total primary productivity after the cross-sectional integration. Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0019] See Figure 1 This is a schematic diagram of the framework of the remote sensing estimation method for phytoplankton fractional primary productivity provided by the present invention.

[0020] In this embodiment of the invention, observational data obtained using multiple cruise observation data and primary productivity observation data at ocean depth z from a fixed observation station includes: net primary productivity (NPP) of phytoplankton at multiple ocean depths and total net primary productivity (Total NPP), as well as net primary productivity (INPP) after depth integration within the euphotic layer. This allows for the construction of a multi-source dataset. The multi-source dataset can be divided into a fitting set (80%) and a validation set (20%), used for parameter fitting and accuracy verification and evaluation of the subsequent model, respectively.

[0021] In practice, the depth integral INPP is the integration of NPP(z) at depth z, and the depth of integration is from 0 meters above the surface to the depth Zeu of the true light layer.

[0022] In this embodiment of the invention, the acquired remote sensing data includes: the concentration of chlorophyll a in the ocean surface layer, and the light absorption coefficient of phytoplankton (typically the light absorption coefficient a of phytoplankton at a wavelength of 443 nm). ph (443)), photosynthetically active radiation (PAR) (photons with wavelengths in the 700 nm range, which is called photosynthetically active radiation), and ocean water mixing layer depth. Ocean water mixing layer depth is usually reanalysis data. Ocean water mixing layer depth refers to the depth of a layer of water in the ocean surface where the physical properties of the water, such as temperature, salinity, and density, are relatively uniform in the vertical direction due to the strong turbulent mixing caused by physical processes such as wind, waves, and convection.

[0023] In this embodiment of the invention, based on a preset hourly time window (usually ±12-hour time window) and a pixel spatial window (usually 3×3 pixel spatial window), and with the pixel spatial window having a coefficient of variation threshold and an effective pixel number threshold, the thresholds can be set according to the actual application scenario, thereby performing quality control on the matching results, matching the observed data and remote sensing data to obtain the measured dataset.

[0024] In this embodiment of the invention, the average light absorption coefficient of phytoplankton of a corresponding particle size is calculated by using the specific absorption coefficient of a single particle size phytoplankton and the corresponding chlorophyll a concentration.

[0025] In practical implementation, by combining existing research on the specific absorption coefficient of phytoplankton, the specific absorption coefficient a* for phytoplankton of different particle sizes can be established. ph,i (λ). The specific absorption coefficient of phytoplankton is an optical parameter that describes the ability of a unit mass of phytoplankton to absorb light at a specific wavelength. It is a key parameter that connects phytoplankton biomass (such as chlorophyll a concentration) with the inherent optical properties of water.

[0026] In practice, chlorophyll a concentration can characterize the biomass of phytoplankton of different particle sizes, while the specific absorption coefficient can reflect the light absorption capacity of phytoplankton of the corresponding particle size. Therefore, the average light absorption coefficient (average light absorption coefficient in a specific wavelength range, mainly 400nm to 700nm) of phytoplankton of the corresponding particle size can be estimated using these two parameters.

[0027] In this embodiment of the invention, the average light absorption coefficient is calculated based on the product of the specific absorption coefficient and the chlorophyll a concentration within the 400nm to 700nm wavelength range, specifically using the following calculation formula: a' ph,i (z)=(∫ 400 700 a ph,i (λ,z)×E(λ,z)dλ) / (∫ 400 700 E(λ,z)dλ) ≈(∫ 400 700 Chla i (z)×a* ph,i (λ)dλ) / (700-400), Among them, a' ph,i (z) represents the average light absorption coefficient of phytoplankton of particle size i at ocean depth z in the 400 to 700 nm wavelength range, Chla i (z) represents the chlorophyll a concentration of phytoplankton of particle size i at ocean depth z, a* ph,i (λ) represents the specific absorption coefficient of phytoplankton of particle size i at wavelength λ; a ph,i (λ,z) ​​represents the light absorption coefficient of phytoplankton of size i at ocean depth z at wavelength λ, which is also the specific absorption coefficient a* of phytoplankton of size i. ph,i E(λ) is the product of the chlorophyll a concentration at the corresponding particle size; E(λ,z) ​​represents the light irradiance (light intensity) at wavelength λ at ocean depth z.

[0028] In practical implementation, the above formula uses an equivalent average absorption coefficient to represent the overall absorption capacity of phytoplankton to actual light intensity in the 400-700 nm wavelength range. Specifically, the absorption coefficient α at each wavelength is first... ph,i (λ,z) ​​is weighted by the light intensity E(λ,z) ​​at that wavelength, and integrated over the range of 400 to 700 nm to obtain the total absorption in the same wavelength band; then the total absorption in the same wavelength band is calculated using the total light intensity (∫) 400 700 E(λ,z) ​​is normalized to obtain the weighted average form a' ph,i (z). Furthermore, assuming that E(λ,z) ​​varies negligibly with wavelength in the 400–700 nm band (or can be approximated by the band average), the weighting term can be extracted from the integral and canceled out in the numerator and denominator, reducing the expression to a... ph,i (z) band arithmetic mean. On the other hand, a ph,i (z) can be expressed as the chlorophyll a concentration Chla i (z) and specific absorption coefficient a* ph,i The product of (λ) gives the corresponding approximate expression.

[0029] In practical implementation, photons in the 400nm to 700nm wavelength range are the main photons involved in phytoplankton photosynthesis, i.e., photosynthetic quanta. Chlorophyll a, b and other auxiliary pigments (such as carotenoids and phycobiliproteins) mainly absorb light energy in this wavelength range to drive photochemical reactions. Therefore, by performing integral calculations on the 400nm to 700nm wavelength range, the light absorption coefficients of phytoplankton at three particle sizes can be quantitatively estimated.

[0030] In this embodiment of the invention, the concentration of chlorophyll a in the ocean surface layer, the depth of the ocean euphotic layer (usually the depth where light intensity decays to 1% of the sea surface) and the depth of the ocean water mixing layer are obtained, and the total chlorophyll a concentration at ocean depth z is calculated. Using a preset particle size division model (an empirical model can be used), the chlorophyll a concentration of phytoplankton at a single particle size is obtained.

[0031] In practical implementation, to estimate the total chlorophyll a concentration at ocean depth z, the depth of the euphotic layer and the mixed layer of the ocean in a certain area can be used to determine whether the vertical distribution of chlorophyll in that area is uniformly mixed or Gaussian. If the mixed layer depth is greater than the euphotic layer depth, it is determined to be uniformly mixed, and therefore the chlorophyll concentration at any depth z is equal to the chlorophyll a concentration at the ocean surface. In other words, obtaining the chlorophyll concentration at the sea surface yields the chlorophyll concentration distribution of the entire vertical profile. If the mixed layer depth is less than the euphotic layer depth, it is determined to be Gaussian. In this case, based on the surface chlorophyll concentration and the euphotic layer depth, a preset chlorophyll concentration vertical profile distribution model (an empirical Gaussian function model can be used) is needed to estimate the chlorophyll a concentration at ocean depth z.

[0032] In this embodiment of the invention, an estimation function for the photosynthetic quantum yield of phytoplankton of a corresponding size is established based on the relationship between light intensity and the photosynthetic quantum yield of a single phytoplankton.

[0033] In practical implementation, based on the measured dataset established by this invention, the differences in photosynthetic quantum yield of phytoplankton of different sizes and their response to environmental changes can be analyzed, clarifying the intrinsic relationship between photosynthetic quantum yield of phytoplankton of different sizes and marine environmental factors. This invention further quantifies the photosynthetic quantum yield of phytoplankton of different sizes into a model that varies with light intensity (light intensity varies with depth), thereby realizing the estimation of photosynthetic quantum yield of phytoplankton of three sizes.

[0034] In this embodiment of the invention, an estimation function for the photosynthetic quantum yield of a single-particle-size phytoplankton is established based on the relationship between light intensity and phytoplankton photosynthetic quantum yield, as well as the maximum photosynthetic quantum yield of phytoplankton.

[0035] In this embodiment of the invention, the empirical estimation function for the photosynthetic quantum yield of the corresponding particle size phytoplankton is as follows: φ i (z)=(K φ,i ×φ i,max ) / (K φ,i +E(z,t)), Where, φ i (z) represents the photosynthetic quantum yield of phytoplankton of particle size i at ocean depth z, K φ,i φ represents the light intensity required for phytoplankton of particle size i to reach half of its maximum photosynthetic quantum yield. i,max E(z,t) represents the maximum photosynthetic quantum yield of phytoplankton at particle size i, and E(z,t) represents the light intensity at ocean depth z at time t.

[0036] In practical application, the above formula signifies that the number of times phytoplankton can receive light and perform carbon fixation per unit time is limited. When the actual ambient light intensity is below the upper limit of light that phytoplankton can utilize, the photosynthetic quantum yield is relatively high; conversely, when the actual light intensity is above the upper limit of light that phytoplankton can utilize, a photoinhibition effect occurs, and the photosynthetic quantum yield decreases. Therefore, overall, the photosynthetic quantum yield is inversely proportional to the light intensity.

[0037] In this embodiment of the invention, particle size i includes micro, nano, and pico, wherein K φ,m K φ,n and K φ,p 3.5074Einm respectively -2 d -1 1.5619Einm -2 d -1 and 4.1241Einm -2 d -1 φ m,max φ n,max and φ p,max 0.0799 mol CEin -1 0.0760 mol CEin -1 and 0.0374 molCEin -1 .

[0038] In this embodiment of the invention, the relationship between light intensity and phytoplankton photosynthetic quantum yield is obtained by fitting matched observation data and remote sensing data.

[0039] In practice, based on the pre-established measured dataset, a Gaussian process is used as a surrogate model during the fitting process, and the root mean square error is minimized as the objective function. The final model parameters K are obtained. φ,m K φ,n and K φ,p , and φ m,max φ n,max and φ p,max Specifically, based on the measured total primary productivity (NPP(z)) data during the cultivation period at depth z, and the average light absorption coefficient (a') of phytoplankton of different particle sizes at the corresponding depth, the data were analyzed. ph,i The data of illumination (E(z)) and light intensity (E(z)) are used to apply the formula NPP(z)=Σ i=1 3 NPP i (z)=E(z)a' ph,i (z)φ i (z), K is obtained by fitting. φ,i and φ i,max .

[0040] In this embodiment of the invention, the net primary productivity of phytoplankton at the corresponding particle size is calculated based on the photosynthetic quantum yield and average light absorption coefficient of a single particle size phytoplankton, within the scale of light intensity duration and ocean depth measurement range.

[0041] In this embodiment of the invention, the net primary productivity of phytoplankton at the corresponding particle size is calculated based on the product of light intensity, photosynthetic quantum yield, and average light absorption coefficient.

[0042] In practice, the parameters for assessing primary productivity are derived from the framework of light absorption and transformation, based on light intensity, photosynthetic quantum yield, and average light absorption coefficient. Light intensity and average light absorption coefficient are mainly related to light absorption, while photosynthetic quantum yield is mainly related to transformation. The primary productivity of a single-size phytoplankton is calculated by taking both aspects into account.

[0043] In this embodiment of the invention, the net primary productivity of phytoplankton at multiple particle sizes is added together to calculate the total primary productivity of phytoplankton.

[0044] In this embodiment of the invention, the total primary productivity of the phytoplankton is calculated using the following formula: NPP=Σ i=1 3 NPP i =∫ t=0 D ∫ z=0 Zeu E(z,t)a' ph,i (z)φ i (z)dzdt, Wherein, NPP represents total primary productivity of phytoplankton, NPP i Let i represent the primary productivity of phytoplankton at particle size i, D represent the duration of light intensity, and Zeu represent the ocean depth range for which the primary productivity of phytoplankton is estimated.

[0045] In practical implementation, the above formula is based on the light absorption and transformation process of phytoplankton, and has three key parameters: light intensity, average light absorption coefficient of phytoplankton, and photosynthetic quantum yield. At time t and depth z, E(z,t) represents the available light intensity, and a' ph,i (z) represents the light absorption capacity of phytoplankton at depth z in the 400-700 nm wavelength range, i.e., how much light intensity can be absorbed by phytoplankton of particle size i, while φ i(z) represents the ability of absorbed light intensity to be converted into organic matter within phytoplankton (because absorbed light energy has three possible fates: conversion into organic matter, dissipation as heat, and release as fluorescence). Finally, a double integral is performed from the sea surface to the euphotic zone depth and from time 0 to the duration of daylight to obtain the diurnal scale of phytoplankton primary productivity at the i-th particle size from the sea surface to the euphotic zone depth. Finally, the three particle sizes are summed to obtain the total primary productivity.

[0046] Previous studies have shown that the specific absorption coefficient α of phytoplankton of different particle sizes... * ph (λ) can differ by nearly one order of magnitude, while its maximum photosynthetic quantum yield φ max Significant differences also exist in the dominant marine environments at different particle sizes. Therefore, neglecting the differences in these photosynthetic parameters (absorption coefficient and photosynthetic quantum yield) across different particle sizes will inevitably lead to substantial errors in the estimation of phytoplankton primary productivity at different particle sizes. This invention, based on multi-source observation data and starting from both the light absorption process and photon yield, constructs a quantitative expression model for the light absorption coefficient and photosynthetic quantum yield of phytoplankton at different particle sizes. Furthermore, it establishes a particle-scale primary productivity estimation method based on the light absorption and transformation framework, achieving a more accurate remote sensing estimation of phytoplankton primary productivity at different particle sizes and enabling a more precise calculation of total primary productivity.

[0047] In practical implementation, the accuracy of the estimation method provided by this invention can be verified and evaluated based on the online multi-source dataset validation set. This invention directly compares and verifies the granular-level primary productivity estimation results (based on the method provided by this invention) using measured granular-level primary productivity data. The verification scatter plot is shown below. Figures 2 to 4 As shown.

[0048] As can be seen from the scatter plot, for the granular primary productivity depth z (NPP) i Verification status of (z) Figure 2 The model exhibits good estimation accuracy, with a root mean square error (RMSE) of only 1.91 mgCm. -3 d -1 The bias is only 0.55 mg Cm -3 d -1 R 2 The coefficient of determination represents the regression relationship between measured primary productivity and estimated primary productivity; n represents the total number of samples, i.e., 44 Pico phytoplankton, 44 Micro phytoplankton, and 44 Nano phytoplankton. Simultaneously, the model was indirectly validated at the level of total NPP (NPP(z)) at depth z. Firstly, the granular-level primary productivity model established in this invention was used to estimate the granular-level NPP.i (z), then three-level NPP i (z) was summed to obtain NPP(z), and compared with the measured NPP(z) on site. The results are as follows: Figure 3 As shown, the model also exhibits good accuracy, with the scatter points evenly distributed on both sides of the 1:1 line, and the RMSE is only 64.39 mgCm. -3 d -1 The bias was also as low as -2.94 mgCm. -3 d -1 HOT represents a long-term primary productivity observation station in a certain region, where HOT indicates the primary productivity measured at that station. Similarly, BATS represents another long-term primary productivity measurement station in a certain region, where BATS indicates the primary productivity measured at that station. The Test Set represents the 20% test dataset previously divided from the measured data. Finally, the estimated NPP(z) was depth-integrated to obtain the depth-integrated primary productivity (INPP), which was compared with the field-measured INPP. The results are as follows: Figure 4 As shown, the model estimation results also exhibit a high degree of consistency with the measured data, with an RMSE of 282.50 mgCm. -2 d -1 The bias was 5.81 mg Cm. -2 d -1 "Others" represents the total primary productivity data (INPP) after depth integration, excluding the HOT and Bats observation stations. The validation results at these three levels demonstrate that the granular primary productivity estimation method provided by this invention has good estimation accuracy.

[0049] This invention also provides a remote sensing estimation device for phytoplankton primary productivity at different particle sizes, comprising: a light absorption coefficient calculation unit, a photosynthetic quantum yield estimation unit, and a phytoplankton primary productivity calculation unit at different particle sizes, wherein: the light absorption coefficient calculation unit is used to calculate the average light absorption coefficient of phytoplankton at a corresponding particle size by using the specific absorption coefficient of a single particle size phytoplankton and the corresponding chlorophyll a concentration; the photosynthetic quantum yield estimation unit is used to establish an estimation function for the photosynthetic quantum yield of phytoplankton at a corresponding particle size based on the relationship between light intensity and the photosynthetic quantum yield of a single particle size phytoplankton; the relationship between light intensity and phytoplankton photosynthetic quantum yield is obtained by fitting matched observation data and remote sensing data; the phytoplankton primary productivity calculation unit is used to calculate the net primary productivity of phytoplankton at a corresponding particle size based on the photosynthetic quantum yield and average light absorption coefficient of a single particle size phytoplankton at a scale of light intensity duration and ocean depth measurement range; and the net primary productivity of phytoplankton at multiple particle sizes is added together to calculate the total primary productivity of phytoplankton.

[0050] In specific implementations, the remote sensing estimation device for phytoplankton fractional primary productivity provided by this invention, wherein the execution unit performs the methods, steps, or functions, which can refer to the remote sensing estimation method for phytoplankton fractional primary productivity provided by this invention.

Claims

1. A remote sensing method for estimating the primary productivity of marine phytoplankton at different particle sizes, characterized in that, include: The average light absorption coefficient of phytoplankton of a certain size was calculated by using the specific absorption coefficient of a single phytoplankton and the corresponding chlorophyll a concentration. Based on the relationship between light intensity and the photosynthetic quantum yield of a single phytoplankton, an estimation function for the photosynthetic quantum yield of phytoplankton at the corresponding particle size is established; the relationship between light intensity and phytoplankton photosynthetic quantum yield is obtained by fitting matched observation data and remote sensing data. Based on the photosynthetic quantum yield and average light absorption coefficient of individual phytoplankton, the net primary productivity of phytoplankton at the corresponding particle size is calculated within the scale of light intensity duration and ocean depth measurement range. The total primary productivity of phytoplankton is calculated by summing the net primary productivity of multiple particle sizes.

2. The remote sensing estimation method for the primary productivity of marine phytoplankton at different particle sizes according to claim 1, characterized in that, The matching process between the observation data and the remote sensing data includes: The acquired observational data include: net primary productivity and total net primary productivity of phytoplankton at multiple ocean depths and multiple particle sizes, as well as depth integral net primary productivity; the acquired remote sensing data include: ocean surface chlorophyll a concentration, phytoplankton light absorption coefficient, photosynthetically active radiation, and ocean water mixing layer depth. Based on preset hourly time windows and pixel spatial windows, and with the pixel spatial windows having a coefficient of variation threshold and an effective pixel count threshold, the observed data and remote sensing data are matched.

3. The remote sensing estimation method for primary productivity of marine phytoplankton at different particle sizes according to claim 1, characterized in that, The calculation yields the average light absorption coefficient of phytoplankton at the corresponding particle size, including: Within the 400nm to 700nm wavelength range, the average light absorption coefficient is calculated based on the product of the specific absorption coefficient and the chlorophyll a concentration, using the following formula: and' ph,i (z)=(∫ 400 700 Chill i (z)×a* ph,i (λ)dλ) / (700-400), Among them, a' ph,i (z) represents the average light absorption coefficient of phytoplankton of particle size i at ocean depth z, Chla i (z) represents the chlorophyll a concentration of phytoplankton of particle size i at ocean depth z, a* ph,i (λ) represents the specific absorption coefficient of phytoplankton of particle size i at wavelength λ.

4. The remote sensing estimation method for primary productivity of marine phytoplankton at different particle sizes according to claim 3, characterized in that, The chlorophyll a concentration was calculated as follows: The concentration of chlorophyll a in the ocean surface layer, as well as the depth of the ocean euphotic layer and the depth of the ocean water mixing layer, are obtained. The total chlorophyll a concentration at ocean depth z is calculated. Using a preset particle size division model, the chlorophyll a concentration of phytoplankton at a single particle size is obtained.

5. The remote sensing estimation method for the fractional primary productivity of marine phytoplankton according to claim 3, characterized in that, The establishment of the estimation function for the photosynthetic quantum yield of phytoplankton at the corresponding particle size includes: Based on the relationship between light intensity and phytoplankton photosynthetic quantum yield, as well as the maximum photosynthetic quantum yield of phytoplankton, an estimation function for the photosynthetic quantum yield of a single-size phytoplankton is established.

6. The remote sensing estimation method for the fractional primary productivity of marine phytoplankton according to claim 5, characterized in that, The estimation function for the photosynthetic quantum yield of the corresponding particle size phytoplankton is as follows: φ i (z)=(K φ,i ×φ i,max ) / (K φ,i +E(z,t)), Where, φ i (z) represents the photosynthetic quantum yield of phytoplankton of particle size i at ocean depth z, K φ,i φ represents the light intensity required for phytoplankton of particle size i to reach half of its maximum photosynthetic quantum yield. i,max E(z,t) represents the maximum photosynthetic quantum yield of phytoplankton at particle size i, and E(z,t) represents the light intensity at ocean depth z at time t.

7. The remote sensing estimation method for the fractional primary productivity of marine phytoplankton according to claim 6, characterized in that, Particle size i includes small m, micro n, and pico p, where K φ,m K φ,n and K φ,p 3.5074Einm respectively -2 d -1 1.5619Einm -2 d -1 and 4.1241Einm -2 d -1 φ m,max φ n,max and φ p,max 0.0799 mol CEin -1 0.0760 mol CEin -1 and 0.0374 molCEin -1 .

8. The remote sensing estimation method for the fractional primary productivity of marine phytoplankton according to claim 6, characterized in that, The calculation yields the net primary productivity of phytoplankton at the corresponding particle size, including: The net primary productivity of phytoplankton at the corresponding particle size is calculated by multiplying light intensity, photosynthetic quantum yield, and average light absorption coefficient.

9. The remote sensing estimation method for primary productivity of marine phytoplankton at different particle sizes according to claim 8, characterized in that, The total primary productivity of the phytoplankton is calculated using the following formula: NPP=Σ i=1 3 NPP i =∫ t=0 D ∫ z=0 Zeu E(z,t)a' ph,i (z)φ i (z)dzdt, Wherein, NPP represents total primary productivity of phytoplankton, NPP i denoted as net primary productivity of phytoplankton at particle size i, D represents the duration of light intensity, and Zeu represents the range of ocean depth measurement.

10. A remote sensing estimation device for the primary productivity of marine phytoplankton at different particle sizes, characterized in that, include: The system includes a light absorption coefficient calculation unit, a photosynthetic quantum yield estimation unit, and a phytoplankton fractional primary productivity calculation unit, among which: The light absorption coefficient calculation unit is used to calculate the average light absorption coefficient of phytoplankton of a corresponding particle size by using the specific absorption coefficient of a single particle size phytoplankton and the corresponding chlorophyll a concentration. The photosynthetic quantum yield estimation unit is used to establish an estimation function for the photosynthetic quantum yield of phytoplankton of a corresponding size based on the relationship between light intensity and the photosynthetic quantum yield of a single phytoplankton; the relationship between light intensity and phytoplankton photosynthetic quantum yield is obtained by fitting matched observation data and remote sensing data; The phytoplankton fractional primary productivity calculation unit is used to calculate the net primary productivity of phytoplankton at the corresponding fractional size based on the photosynthetic quantum yield and average light absorption coefficient of a single fractional size of phytoplankton, within the scale of light intensity duration and ocean depth measurement range; and to calculate the total primary productivity of phytoplankton by adding the net primary productivity of multiple fractional sizes.

Images