On the variations of sea surface pCO2 in the northern South China Sea: A remote sensing based neural network approach

Jo, Young‐Heon; Dai, Minhan; Zhai, Weidong; Yan, Xiao-Hai; Shang, Shaoling

doi:10.1029/2011jc007745

Cited by 32 publications

(36 citation statements)

References 49 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…대륙주변해는 대기, 해양, 대륙을 잇는 주요 탄소 저장고로서 (Omar et al, 2007), Tsunogai et al(1999) (Takahashi et al, 2002;Takahashi et al, 2009;Friedrich and Oschlies, 2009b) 동 해에서 표층 pCO 2 를 계산하는 모델 구축 연구도 진행되어 왔다 (Oh et al, 1999;Kang, 1999;Choi et al, 2011;Kim et al, 2014 (Lefèvre et al, 2005;Friedrich and Oschlies, 2009a;Telszewski et al, 2009;Jo et al, 2012;Landschützer et al, 2013). 신경망 모델은 모델 디자인 과정에서 입력자료와 결과값 사이 의 비선형적인 관계를 스스로 만들어 낼 수 있다.…”

unclassified

“…신경망 모델은 모델 디자인 과정에서 입력자료와 결과값 사이 의 비선형적인 관계를 스스로 만들어 낼 수 있다. 따라서 현장관측자 료를 바탕으로 pCO 2 값을 계산할 수 있으며, 2차원의 위성관측자 료를 활용하여 현장관측에서 빠진 해역을 대상으로 pCO 2 분포도를 생성할 수 있다 (Jo et al, 2012;Landschützer et al, 2013 (Jo et al, 2012 (Fig. 3).…”

unclassified

“…3(b)). 현장관측값과 비교하여 ±20 Jo et al, 2012). NN model consists of three layers (input layer, hidden layer and output layer).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Sea Surface pCO2 and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Park¹,

Lee

Jo³

2016

The Sea

Self Cite

View full text Add to dashboard Cite

Currently available surface seawater partial pressure carbon dioxide (pCO 2 ) data sets in the East Sea are not enough to quantify statistically the carbon dioxide flux through the air-sea interface. To complement the scarcity of the pCO 2 measurements, we construct a neural network (NN) model based on satellite data to map pCO 2 for the areas, which were not observed. The NN model is constructed for the Ulleung Basin, where pCO 2 data are best available, to map and estimate the variability of pCO 2 based on in situ pCO 2 for the years from 2003 to 2012, and the sea surface temperature (SST) and chlorophyll data from the MODIS (Moderate-resolution Imaging Spectroradiometer) sensor of the Aqua satellite along with geographic information. The NN model was trained to achieve higher than 95% of a correlation between in situ and predicted pCO 2 values. The RMSE (root mean square error) of the NN model output was 19.2 µatm and much less than the variability of in situ pCO 2 . The variability of pCO 2 with respect to SST and chlorophyll shows a strong negative correlation with SST than chlorophyll. As SST decreases the variability of pCO 2 increases. When SST is lower than 15 o C, pCO 2 variability is clearly affected by both SST and chlorophyll. In contrast when SST is higher than 15 o C, the variability of pCO 2 is less sensitive to changes in SST and chlorophyll. The mean rate of the annual pCO 2 increase estimated by the NN model output in the Ulleung Basin is 0.8 µatm yr -1 from 2003 to 2014. As NN model can successfully map pCO 2 data for the whole study area with a higher resolution

show abstract

unclassified

See 1 more Smart Citation

Sea Surface pCO2 and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Park¹,

Lee

Jo³

2016

The Sea

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the field of remote sensing, recently adopted machine learning approaches or nonlinear analyses, may be able to effectively model the pressure of ocean carbon dioxide. These techniques include various neural networks including self-organizing maps (SOM) [6,14,15,20,32,34,35], mechanistic nonlinear models [13], principal component analysis [9], mechanistic semi-analytic algorithms [7], and quadratic polynomial regression [9]. However, generalization of the empirical modeling approaches from one area to different areas is still challenging.…”

Section: Introductionmentioning

confidence: 99%

Estimation of Fugacity of Carbon Dioxide in the East Sea Using In Situ Measurements and Geostationary Ocean Color Imager Satellite Data

Jang

Park

et al. 2017

Remote Sensing

View full text Add to dashboard Cite

Abstract:The ocean is closely related to global warming and on-going climate change by regulating amounts of carbon dioxide through its interaction with the atmosphere. The monitoring of ocean carbon dioxide is important for a better understanding of the role of the ocean as a carbon sink, and regional and global carbon cycles. This study estimated the fugacity of carbon dioxide (ƒCO 2 ) over the East Sea located between Korea and Japan. In situ measurements, satellite data and products from the Geostationary Ocean Color Imager (GOCI) and the Hybrid Coordinate Ocean Model (HYCOM) reanalysis data were used through stepwise multi-variate nonlinear regression (MNR) and two machine learning approaches (i.e., support vector regression (SVR) and random forest (RF)). We used five ocean parameters-colored dissolved organic matter (CDOM; <0.3 m −1 ), chlorophyll-a concentration (Chl-a; <21 mg/m 3 ), mixed layer depth (MLD; <160 m), sea surface salinity (SSS; 32-35), and sea surface temperature (SST; 8-28 • C)-and four band reflectance (Rrs) data (400 nm-565 nm) and their ratios as input parameters to estimate surface seawater ƒCO 2 (270-430 µatm). Results show that RF generally performed better than stepwise MNR and SVR. The root mean square error (RMSE) of validation results by RF was 5.49 µatm (1.7%), while those of stepwise MNR and SVR were 10.59 µatm (3.2%) and 6.82 µatm (2.1%), respectively. Ocean parameters (i.e., sea surface salinity (SSS), sea surface temperature (SST), and mixed layer depth (MLD)) appeared to contribute more than the individual bands or band ratios from the satellite data. Spatial and seasonal distributions of monthly ƒCO 2 produced from the RF model and sea-air CO 2 flux were also examined.

show abstract

“…Remote sensing pCO 2 data are used to calculate the spatial variability and the correlation among observations in individual grid boxes. Given the rapidly growing applications of remote sensing techniques in mapping pCO 2 and constraining pCO 2 spatial variability [e.g., Stephens et al, 1995;Nelson et al, 2001;Ono et al, 2004;Sarma et al, 2006;Zhu et al, 2009;Hales et al, 2012;Jo et al, 2012], our approach to estimating uncertainty in gridded pCO 2 data should have numerous applications.…”

Section: Introductionmentioning

confidence: 99%

Quantifying uncertainty sources in the gridded data of sea surface CO₂ partial pressure

et al. 2014

Self Cite

View full text Add to dashboard Cite

The bulk uncertainty in the gridded sea surface pCO 2 data is crucial in assessing the reliability of the CO 2 flux estimated from measurements of air-sea pCO 2 difference, because atmospheric pCO 2 are relatively homogeneous and well defined. The bulk uncertainty results from three different sources: analytical error (E m ), spatial variance (r 2 s ), and the bias from undersampling (r 2 u ). Common uncertainty quantification by standard deviation may mix up the different sources of uncertainty. We have established a simple procedure to determine these three sources of uncertainty using remote sensing-derived and field-measured pCO 2 data. E m is constrained by the analytical method and data reduction procedures. r 2 s is derived from the remotely sensed pCO 2 field. r 2 u is determined by spatial variance and the effective number of observations, considering, for the first time, the geometric bias introduced by pCO 2 sampling. This approach is applied to 1 3 1 gridded pCO 2 data collected from the East China Sea. We demonstrate that the spatial distribution of these biases is uneven and that none of them follow the same spatial trend as the standard deviation. r 2 s contributes the most to the uncertainty in gridded pCO 2 data over those grid boxes with good sampling coverage, while r 2 u dominates the total uncertainty in the grid boxes with poor sampling coverage. Application of this procedure to other parts of the global ocean will help to better define the inherent spatial variability of the pCO 2 field and thus better interpolate and/or extrapolate pCO 2 data, and eventually better constrain air-sea CO 2 fluxes.

show abstract

On the variations of sea surface pCO₂ in the northern South China Sea: A remote sensing based neural network approach

Cited by 32 publications

References 49 publications

Sea Surface pCO<sub>2</sub> and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Sea Surface pCO<sub>2</sub> and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Estimation of Fugacity of Carbon Dioxide in the East Sea Using In Situ Measurements and Geostationary Ocean Color Imager Satellite Data

Quantifying uncertainty sources in the gridded data of sea surface CO₂ partial pressure

Contact Info

Product

Resources

About

On the variations of sea surface pCO2 in the northern South China Sea: A remote sensing based neural network approach

Cited by 32 publications

References 49 publications

Sea Surface pCO<sub>2</sub> and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Sea Surface pCO<sub>2</sub> and Its Variability in the Ulleung Basin, East Sea Constrained by a Neural Network Model

Estimation of Fugacity of Carbon Dioxide in the East Sea Using In Situ Measurements and Geostationary Ocean Color Imager Satellite Data

Quantifying uncertainty sources in the gridded data of sea surface CO2 partial pressure

Contact Info

Product

Resources

About

On the variations of sea surface pCO₂ in the northern South China Sea: A remote sensing based neural network approach

Quantifying uncertainty sources in the gridded data of sea surface CO₂ partial pressure