Traditional oceanography has relied on field surveys, but after the successful launch of the world’s first ocean color satellite, the Coastal Zone Color Scanner (CZCS), in 1978, the marine environment began to be observed from spatial, global, and long-term perspectives. However, since the launch of the world’s first geostationary satellite, the Geostationary Ocean Color Imager (GOCI), the realm of ocean color satellites has expanded in short-term, regional, and application aspects. The GOCI was successfully launched on June 27, 2010, and after trial operations, it began regular observations in April 2011, performing its mission stably for 10 years and 6 months until October 2021. Its successor, GOCI-II, was launched on February 19, 2020, continuing the GOCI’s mission of marine environmental monitoring and marine disaster surveillance without any gap. GOCI-II provides more precise information than the GOCI through improved spatial and spectral resolutions, supplying data to the government, related organizations, and users in need (Choi et al., 2021; Lee et al., 2021b; Park et al., 2021; Ruddick et al., 2014; Ryu et al., 2012).

The main purpose of the GOCI was near real-time monitoring of large-scale red tides occurring in the waters around the Korean Peninsula. However, monitoring tasks have been conducted for transboundary floating algae, such as green tides, which have occurred frequently since 2008, and brown tides (Sargassum horneri), which have significantly affected the Korean Peninsula since 2013 (Lee and Lee, 2012; Son et al., 2015). The GOCI provides 13 types of products, including remote reflectance, phytoplankton concentration, and suspended sediment concentration. In contrast, GOCI-II offers 26 types of products, doubling the number of products. Specifically, detection technologies for low-salinity water, floating algae, and sea fog have been enhanced to more accurate levels based on technologies developed during the operation of the GOCI. Furthermore, the number of utilized products has increased not only in the marine sector but also in the land, atmosphere, and meteorological fields, thereby expanding the utilization of the GOCI series in various areas (Ryu and Ishizaka, 2012; Choi et al., 2021).

The observation area of the GOCI local area(LA) mode includes the Yellow Sea, East China Sea, East Sea/Sea of Japan, and parts of the Pacific Ocean to the south of Japan (Fig. 1). The region encompasses Northeast Asia, extending from Mongolia and Russia in the north to northern Taiwan in the south, and from eastern China in the west to the entire region of Japan in the east. Covering an area of 2,500 × 2,500 km, the area accounts for only 1.2% of the Earth’s surface but is a region with a very high population density, a significant economic scale, substantial maritime traffic, and considerable political and military importance. As a consequence of human activities in the region, there have been significant environmental changes, and substantial variability due to climate change is to be expected. Furthermore, the Full Disk mode, which has been recently incorporated into GOCI-II, offers a comprehensive global perspective from the Korean Peninsula, encompassing approximately one-third of the Earth’s surface. Indonesia is located in the nadir direction, and the coverage includes Southeast Asia and Australia, which span parts of the Indian and Pacific Oceans.

Figure 1. Spatial coverage of (a) the GOCI and (b) the GOCI-II in Local Area (LA) mode. The 16 slots and 12 slots in the observation sequence of the GOCI series are indicated respectively. (c) Spatial coverage of the GOCI-II Full Disk (FD) mode.

From the perspective of ocean optics, the coverage area in LA mode includes very turbid waters from the Yangtze River Basin; China’s coastal waters, such as the Bohai Sea; Case-2 waters of the Yellow Sea coast around the Korean Peninsula; and Case-1 waters of the East Sea and the Pacific Ocean to the east of Japan. Therefore, atmospheric correction and ocean analysis algorithms must be developed while considering the characteristics of both regions. Considering the need for various technologies utilized for analysis due to these ocean optical reasons and geopolitical factors, as well as for user convenience, the GOCI Data Processing System (GDPS) for the GOCI and the GOCI-II Ground System (G2GS) for GOCI-II have been developed and implemented (Han et al., 2010; Moon et al., 2010; Ahn et al., 2016). These systems include Level-1B data corresponding to atmospheric correction and technologies for Level-2 analysis for ocean analysis and utilization. The Korea Ocean Satellite Center has continuously improved the accuracy and enhanced algorithms through ongoing research and version updates (Ahn et al., 2012; Choi et al., 2012; Ahn et al., 2015; Ahn and Park, 2020; Choi et al., 2021).

The enhancements in GOCI and GOCI-II’s observational capabilities and data processing systems have markedly influenced the expansion of research utilizing these satellites. Over the past 15 years, this research has resulted in the publication of 578 papers, reflecting the increasing interest in geostationary ocean color satellite data. Initially, studies were focused on data processing, but since 2012, applications in ocean and atmospheric sciences have expanded. Moreover, the integration of diurnal information in 27% of studies and the growing use of artificial intelligence, applied in over 50% of papers by 2022, highlights advancements in the field. This bibliometric analysis of GOCI and GOCI-II research provides insights into current trends and offers a foundation for future studies, particularly in further refining artificial intelligence (AI) applications and enhancing diurnal analysis for environmental monitoring.

From 2005 to December 2023, we investigated SCI/SCIE, SCOPUS, Korea Citation Index (KCI), and nonindexed papers. Using the Web of Science (https://apps.webofknowledge.com/), we searched for papers on the subject of “GOCI” or “Geostationary Ocean Color Imager” and found 400 papers. The search fields included article titles, abstracts, author keywords, and system keywords. As of December 31, 2023, we had also searched KCI papers in the Korea Citation Index (https://www.kci.go.kr/kciportal/) via “GOCI” or “Geostationary Ocean Color Imager,” resulting in 178 papers. The authors manually reviewed the titles and abstracts of the initially retrieved papers and excluded those where GOCI was only cited but not used. These papers were classified into four categories: SCI/SCIE, SCOPUS, KCI, and nonindexed. Thirteen papers published in nonindexed journals were excluded from the statistical analysis of this study. We further divided the remaining papers into international and domestic categories, with domestic papers limited to those indexed in the KCI.

3.1. Research Trends and International Collaboration

Fig. 2 summarizes the annual number of published papers, which are classified into domestic and international categories. Between 2005 and 2009, several papers on sensor development and preparatory research for analytical techniques were published prior to the launch of the GOCI; these papers have been combined and summarized together. In the case of domestic papers, up to 20 research papers were published in a year, which is a relatively high number. However, it is noteworthy that more than 10 papers were published even before the GOCI launch in 2010, which is a considerably higher number. This occurred because the development of the GOCI sensor began in 2003, and projects for developing analytical techniques for GOCI data processing started simultaneously. As a result, the outcomes of these research projects were published in the form of preliminary studies before the satellite launch, and a special issue on the GOCI was published in Korean Journal of Remote Sensing in 2010.

Figure 2. Number of published papers related to GOCI research conducted from 2005 to 2023. The blue bars represent papers published in international journals, whereas the red bars indicate papers published in domestic Korean journals. Papers from 2005 to 2009, the period before GOCI’s launch, are combined and presented together. The dotted line represents the trend in the number of GOCI-related published papers, which shows an increasing tendency.

Identifying a clear trend in the number of domestic papers after the launch of the GOCI was difficult. In 2013, 2018, and 2021, many papers—approximately 20—were published, whereas in 2011, 2020, and 2022, only a small number—approximately 4—were published. It is presumed that such fluctuations in domestic publication numbers are related to domestic events associated with the GOCI and the completion times of research projects.

The number of international papers was low, with only one paper published before 2010, three published in 2010, and approximately four published in 2011. However, after the publication of the OSJ special issue in 2012, which included 18 papers, the number significantly increased to 37 in 2016. The highest number of publications was in 2018, with 49 papers, and since then, approximately 40 papers have been published each year. The overall trend, which combines both domestic and international papers, shows a continuous increase in the total number of publications.

By analyzing the frequency of keywords and their correlations (Fig. 3), we found that the keywords “remote sensing” and “ocean color” centered on the GOCI were the most frequent, followed by “atmospheric correction” and “chlorophyll-a (Chl-a).” The frequencies of Moderate Resolution Imaging Spectroradiometer—a highly utilized polar-orbiting ocean color satellite—and the Geostationary Environmental Monitoring Sensor (GEMS), which is also mounted on Geo-KOMPSAT-2B—were also high. Research has concentrated on the East China Sea and Yellow Sea within the coverage of the GOCI.

Figure 3. Diagram showing the relationship using keywords presented in research papers. The larger the circle is, the more frequently the keyword is used. If keywords are used together in a paper, they are connected by lines, and the more frequently they are used together, the thicker the connecting lines are.

While traditional polar-orbiting satellites have focused extensively on global oceans, the GOCI has been the subject of many studies on Case-2 waters centered on turbid coastal areas. In addition to the ocean color field, many papers have focused on the aerosol optical depth (AOD) in the atmospheric field.

When words with the same or similar meanings are expressed in various ways, there are limitations in understanding relationships on the basis solely of keywords. For example, terms related to total suspended material (TSM), such as “suspended particle matter,” “suspended sediment concentration,” “suspended matter,” and “total suspended matter,” are used, making it difficult to quantitatively evaluate relationships and frequencies. In the case of chlorophyll, the authors also used various notations.

Fig. 4 presents statistics of leading research countries based on the affiliations of authors in international journals: (a) is based on corresponding authors, and (b) is based on coauthors, excluding corresponding authors. An analysis of the corresponding authors of GOCI research papers by country revealed that China accounted for 48% and Korea represented 32%, collectively comprising approximately 80%. The United States followed, with approximately 10%, and Japan had a relatively low proportion, at 1.8%. However, in terms of coauthors, the participation of the United States and Japan increased: China represented 36%; Korea, 29%; and the United States, Japan, the United Kingdom, and Germany accounted for approximately 18%, 3.2%, 1.7%, and 1%, respectively. The high degree of participation of European coauthors is thought to be due to multiple opportunities for joint research through three GOCI principal investigator workshops and significant interest in atmospheric correction and ocean color research for Case-2 waters in Europe. In the United States, many Chinese–American professors have participated, leading to numerous publications through collaborative research with their Chinese counterparts. An unexpected result was that few research papers were published from these three countries—Japan, Taiwan, and Russia—even though the GOCI coverage included large areas of Japanese waters, northern Taiwan, and some Russian waters (Chau et al., 2021).

Figure 4. Statistics of leading research countries based on authors’ affiliations compiled from papers published in international journals. The proportions of the total papers and the corresponding countries are indicated in pie charts. Research-leading countries based on (a) corresponding authors and (b) coauthors, excluding corresponding authors.

3.2. Classification and Analysis of GOCI Research

All the collected papers were broadly classified into data processing, ocean, atmosphere, and land categories, as shown in Table 1. A further subdivision of the papers was conducted through the analysis of titles, keywords, and abstracts. Papers related to the preprocessing of satellite images and the development of application algorithms have been broadly classified under data processing. The preprocessing steps were divided into four categories: geometric correction, radiometric correction, atmospheric correction, and other preprocessing steps. “Other preprocessing steps” include technologies related to Bidirectional Reflectance Distribution Function (BRDF), Infrared Drop Correction (IRDC), Modulation Transfer Function, band shifting, and sensor-related technologies. The studies were further divided into seven categories, including Level-2 product algorithm (GDPS) research, technology development for convergence research, and satellite operation and services.

Table 1 . Classification table used for categorizing research topics.

Major category	Subcategory
Data processing	Geometric correction Other preprocessing steps	Radiometric correction Level-2 algorithms Operation and service	Atmospheric correction Fusion techniques
Ocean	Chl-a Miscellaneous product Marine algae Integrated environment	TSM Ocean optics Red tide	CDOM Physics Cal/Val Other ocean application
Atmosphere	Application	Cloud/Fog	Disaster
Land	Chl-a Land optics	TSM Water quality Other land application	Organic carbon Land covers

Four main categories were used: data processing, ocean, atmosphere, and land. The subcategories were determined by analyzing the main research areas within each field..

The ocean category was subdivided into 11 products and applications: Chl-a; TSM; Colored Dissolved Organic Matter (CDOM); and other products, which are basic products of ocean color satellites; marine optics; ocean physical applications; red tides; marine biology, including green and brown tides; calibration/validation; marine convergence applications; and other marine applications (Li et al., 2021).

The atmosphere category was subdivided into three sections: atmospheric applications, clouds/sea fog, and atmospheric disasters. Atmospheric applications have included mainly AOD studies; many studies have focused on the effects of clouds or sea fog, and atmospheric disasters have included research on yellow dust and volcanic ash (Kim and Park, 2021).

The land category was subdivided into seven groups, mainly related to inland waters dealing with lakes, which were categorized into five types: Chl-a, TSM, organic carbon, water optics, and water quality(Portela et al., 2024). Additionally, land applications were divided into two types: land cover and vegetation.

As mentioned above, we broadly classified the GOCI-related research papers into four categories. An analysis of the number of research papers published over the past 18 years revealed that data processing accounted for approximately 26%, ocean studies accounted for 52%, atmosphere studies accounted for 13%, and land studies accounted for approximately 9% (Fig. 5). We further analyzed these results by dividing them into international and domestic journals. In international journals, data processing constituted 19%, ocean studies constituted 60%, atmosphere constituted 12%, and land constituted 9%. In domestic journals, data processing, ocean studies, atmosphere, and land accounted for 42%, 35%, 16%, and 7%, respectively. While there was no significant difference between international and domestic papers in terms of atmospheric and land applications, there was a substantial disparity in the ocean and processing fields. This may be attributed to the extensive independent research conducted in Korea for data processing, following the development of the world’s first geostationary ocean color satellite. However, we need to consider why the proportion of ocean applications is lower in domestic journals than in international journals, despite the greater focus on data processing.

Figure 5. Pie charts categorizing GOCI-related published papers from 2005 to 2023 based on four main classifications. The proportion of each field is shown in the figures. (a) Statistics including both international journals and Korean domestic journals, (b) statistics of papers published in international journals, and (c) statistics of papers published in Korean domestic journals. While research in the ocean field predominates, data processing research is more prevalent in Korean domestic journals.

In Fig. 6, to analyze annual research trends, we categorized the four main classifications into three groups—data processing, ocean, and atmosphere/land—and represented them as percentages. The consolidation of atmosphere and land applications into a single category, despite their relatively smaller proportion, is justified by their alignment with applications outside the original mission of GOCI, which was primarily intended for ocean color monitoring. This integrated grouping allows for a more meaningful interpretation of how GOCI has been utilized in non-marine applications. From before the launch until 2011, more than 60% of the papers were on data processing, but after 2012, this proportion decreased to less than 40%. Although GOCI-II was launched in 2020, this trend only slightly increased in 2021, one year after the launch. Moreover, the ocean research field tended to exceed 50% from 2012 onward, gradually maintaining a tendency greater than 50%. The atmosphere/land application fields also showed an increasing trend from approximately 20%after 2012, stabilizing at the 20% level from 2022. This trend can be interpreted as a result of increased user utilization following the stabilization of the GOCI Level-1B products.

Figure 6. To confirm the trend of changes in research fields by year, the papers classified into the main categories were represented as percentages. The atmosphere and land fields, which have relatively few studies overall, are combined and presented together. The processing, ocean, and other fields are represented in red, blue, and green, respectively. Over time, research in the processing field has decreased, whereas research in the ocean field has increased. Although the proportions of land and atmosphere fields are small, they continue to increase steadily.

Fig. 7(a) summarizes the number of papers according to the subcategories within the main category of data processing, with a total of 148 related domestic and international papers. We subdivided this category into seven groups. The contents related to the GOCI design, sensor development, reception, processing, and distribution were grouped under operation/service, which had the largest number of published papers. This is likely because, as the world’s first geostationary ocean color sensor observes from 50 times farther away compared to polar-orbiting satellites, a frame-capture-type sensor was used, leading to many studies on sensor calibration due to orbital differences (Asaoka et al., 2020).

Figure 7. Results of classifying the entire study into subcategories. Different colors represent the main categories of data processing: ocean, atmosphere, and land. (a) Data processing, (b) ocean, and (c) combined representation of atmosphere and land.

Next, nearly 30 studies were conducted on Level-1B and Level-2 product algorithms related to GDPS and G2GS software (Kim et al., 2016). In particular, several atmospheric calibration studies have been conducted, particularly those related to GDPS releases and version upgrades in 2011, 2013, 2014, and 2017 (Ahn et al., 2012; Ahn et al., 2015; Lee et al., 2013; Kim et al., 2016).

In addition, a number of papers on convergence research technologies using various satellites have been published in recent years regarding technologies for application analysis. Other processing topics included studies on BRDF and IRDC band shifts, followed by studies on radiometric and geometric correction. In international journals, papers on atmospheric correction and convergence technologies were the most common, whereas in domestic journals, operation/service papers were overwhelmingly more prevalent, and convergence technologies were fewer. This is thought to be because papers submitted by domestic authors to international journals are counted as international papers, and in domestic journals, many papers focus on a single technology rather than on convergence technologies across various fields.

As shown in Fig. 7(b), the basic products of ocean color satellites are the chlorophyll concentration, TSM, and dissolved organic matter concentration in seawater. By observing these parameters hourly, geostationary ocean color satellites can analyze changes in the marine environment in real-time (He et al., 2013; Kim et al., 2018; Brovchenko et al., 2022; Chen et al., 2023). This capability allows swift detection and response to marine productivity, pollution conditions, and the occurrence of red tides (Choi et al., 2014). A total of 303 domestic and international publications have been produced in the marine field, with the majority focused on the TSM. The TSM serves as a crucial indicator for tracking marine pollution by measuring the amount of fine particles and pollutants suspended in seawater. GOCI data have been widely utilized in research aimed at real-time monitoring of coastal pollution and tracking the dispersion of pollutants.

Leveraging the temporal resolution advantages of geostationary ocean color satellites, the most frequently submitted papers focused on diurnal variations in coastal areas. Since 2008, there has been a steady increase in studies on green and brown tides, which cause significant damage to fishermen and coastal regions (Hu et al., 2023). In contrast, the number of papers on red tides has decreased compared with that in the initial period. Instead, studies focusing on temporal changes in phytoplankton, which are based on the reliability of atmospheric correction and remote sensing reflectance (Rrs) values, have been published. Compared with the TSM, the detection of these more precise changes is thought to be closely related to the reliability of the GOCI data.

Numerous papers have been published on mesoscale eddies and upwelling related to red tides in the field of physical oceanography, followed by application studies on other products (Shin et al., 2018). There are relatively few papers specifically on ocean optics (Kd, Particulate Organic Carbon; POC, and Photosynthetically Available Radiation; PAR), calibration/validation, and CDOM. CDOM is not commonly used independently but has been employed in several studies as a parameter for estimating sea salinity (Son and Choi 2022). Notably, research on the tidal flats of the Chinese coast in the Yellow Sea and the western coast of the Korean Peninsula has been published using high-temporal-resolution GOCI data (Lee et al., 2021a). Research on fisheries applications has been very limited, and studies on the full-disk observations of GOCI-II are expected to increase in number in the future. The international trends were similar to the overall trend; however, domestically, most papers were published on green and brown tides, with relatively few studies on the TSM (Son et al., 2015).

The GOCI is utilized not only in marine applications but also for tracking changes in the concentrations of atmospheric particles such as aerosols. This capability enables the monitoring of atmospheric pollution issues, such as yellow dust and industrial emissions, as well as the study of correlations with climate change. In atmosphere/land fields, atmospheric applications related to the AOD are overwhelmingly prevalent (Lee et al., 2010). Many studies have focused solely on AOD using GOCI data, as well as combined analyses using the GOCI and Meteorological Imager (MI). For GOCI-II, there have been numerous fusion studies with the GEMS, which is also mounted on the Geo-Kompsat-2B satellite. GEMS, the world’s first geostationary environmental sensor, has garnered significant attention from atmospheric and meteorological researchers, leading to a substantial number of studies published shortly after its launch (Kim et al., 2020, Lee et al., 2021c, Choi et al., 2023; Lee et al., 2023).

Convergence research utilizing sensors that simultaneously observe marine and atmospheric environments on a single platform is expected to constitute a global milestone in the field of geostationary satellite applications. As indicated by the keyword analysis in Fig. 3, the high temporal resolution of geostationary ocean color satellites is considered highly beneficial for atmospheric and meteorological research. From the perspective of spatial resolution, land applications have led to several studies on vegetation changes and land cover variations, utilizing the high temporal resolution of GOCI despite its lower spatial resolution than high-resolution land satellites. Several papers on terrestrial lakes have also been submitted, primarily focusing on China, which has large lakes suitable for the spatial scale of the GOCI. In contrast, Korea, which lacks lakes of a scale compatible with the spatial resolution of the GOCI, does not have any published studies on lakes.

As the world’s first geostationary ocean color satellite, GOCI has revolutionized short-term, regional, and operational monitoring by providing hourly observations—capabilities that were impossible with polar-orbiting ocean color satellites due to their limited temporal resolution. This enhanced temporal resolution has enabled the observation of various diurnal variations in coastal and marine environments (Lou and Hu, 2014; Huang et al., 2015), leading to the publication of numerous research papers. Fig. 8 presents an analysis of the number of studies utilizing such diurnal information. Among all the studies, 73% did not take advantage of the high temporal resolution—these are thought to include most of the papers on data processing. In contrast, 27%of the papers utilized diurnal information, with GOCI having the greatest advantage in this regard. Among these papers, 74 were in the marine field, 16 in the land field, and relatively few were in the atmosphere and data processing fields.

Figure 8. Analysis of the use of diurnal information—the GOCI’s advantage of acquiring images every hour during the daytime—in papers published in international journals. Studies utilizing hourly acquired data are further categorized according to the main classifications. The numbers inside the inner circle represent the percentage of the total research, and the numbers outside indicate the number of papers.

Even among the papers addressing diurnal information, most have focused on TSM, with relatively few studies concentrating on biological variability such as red tides or Chl-a, which require the detection of more sensitive changes (Feng et al., 2021; Li et al., 2022; Cui et al., 2023; Xu and Chen, 2023). Detecting such subtle biological variability necessitates a clear standard for analyzing the uncertainty and accuracy of GOCI products. Therefore, as the atmospheric correction and data processing performance of GOCI data improves, studies on biological variability are expected to increase.

The development of the GDPS has been pivotal in enhancing the accuracy of GOCI ocean color algorithms. From its initial version 1.0 in 2011 to version 2.0 in 2017, each iteration introduced significant improvements in atmospheric correction algorithms. Notably, the implementation of vicarious calibrations and the incorporation of near-real-time meteorological data have substantially increased the precision of Rrs and other ocean color products. Through continuous calibration and validation activities, the accuracy of GOCI ocean color algorithms has steadily improved, with these enhancements reflected in successive updates to the GDPS. During the approximately seven-year calibration and validation period following GOCI’s launch, the error rate of Rrs—the atmospheric correction output and the most crucial data for marine environmental analysis—was reduced by approximately half. Consequently, the accuracies of all ocean color products that use Rrs as inputs, such as chlorophyll and suspended matter concentrations, also significantly improved. This advancement is likely related to the sharp increase in the number of research papers since 2016, as illustrated in Fig. 2, highlighting the necessity for ongoing efforts to enhance the quality of GOCI data.

While GOCI has facilitated numerous studies focused on short-term monitoring—a characteristic advantage of geostationary ocean color satellites—the availability of 14 years of continuous data from the GOCI series opens new opportunities. Future research is expected to expand to include long-term studies related to climate change and its impact on the marine environment (Park et al., 2021; Park et al., 2022).

On a parallel track, the application of AI in GOCI-related research has seen a noticeable increase in recent years as shown in Fig. 9(a) (Kim et al., 2014; Jang et al., 2016; Fan et al., 2021; Fan et al., 2023; Shin et al., 2024). Since 2016, more than 20% of the research has incorporated AI techniques, and this trend has continued to grow, with over 50% of studies in 2022 employing AI methods. This rise can be attributed to the increasing need to process large volumes of data generated by geostationary satellites like GOCI, which provide frequent, high-resolution observations. Reflecting broader trends in satellite data analysis, AI has proven effective in enhancing the detection of environmental changes and improving predictive modeling (Song et al., 2023). AI applications are distributed across various fields—including marine, atmospheric, and terrestrial research—indicating comprehensive utilization across GOCI research domains (Fig. 9b). As GOCI-II continues to provide high-quality data, the role of AI is expected to expand further, particularly in fields requiring rapid and accurate interpretation of complex environmental data.

Figure 9. Statistical analysis of papers applying machine learning in GOCI-related research. (a) Number of studies applying machine learning by year (blue bars) and the proportion relative to all research (red line), showing an increasing trend since 2014. (b) Classification of studies utilizing machine learning according to the main categories.

Through advancements in data processing systems, international collaborative efforts, and the integration of AI technologies, GOCI’s impact on ocean color research continues to grow. Ongoing enhancements in data quality and processing capabilities will not only facilitate more studies on sensitive biological variability but also enable long-term environmental monitoring crucial for understanding the impacts of climate change.

This study analyzed a total of 578 research papers related to the GOCI and GOCI-II from 2005 to 2023. The number of papers increased to more than 40 per year after the launch of the GOCI, with a peak of 49 papers in 2018. Among domestic papers, 42%focused on data processing, and 35% focused on ocean research, whereas 60% of international papers focused on ocean research and 19% focused on data processing, indicating differences between research areas. The main keywords used were “remote sensing,” “ocean color,” “atmospheric correction,” and “Chl-a,” and the studies focused mainly on the turbid coastal waters of the Yellow Sea and the East China Sea. In terms of author nationality, China accounted for 46% and Korea 35% of the lead authors, making up more than 80% of the research, while coauthors from the U.S., Japan, the UK, and Germany actively participated. Ocean research accounted for 52% of the total studies, followed by atmosphere (13%), land (9%), and data processing (26%). The use of temporal resolution in studies was found in 27% of the total, with a notable focus on TSM research, and more than 50% of the papers in 2022 utilized artificial intelligence.

The results of this study clearly indicate the necessity of developing a follow-up satellite, GOCI-III, on the basis of the successful operation of the GOCI and GOCI-II. GOCI-III is expected to overcome the limitations of current satellites by incorporating new technologies, such as blue carbon-related data products, hyperspectral and polarimetric observations for precise disaster monitoring, and long-term predictions of marine ecosystem changes. This would enable real-time monitoring of various disasters, such as harmful algal blooms, suspended matter, and coastal water pollution, contributing significantly to national marine policy. Moreover, the development of GOCI-III could serve as a valuable reference for similar geostationary ocean color satellite projects planned by National Aeronautics and Space Administration, European Space Agency, and Scripps Institution of Oceanography. It is anticipated that GOCI-III will not only advance Korea’s marine environmental monitoring capabilities but also play a crucial role in strengthening Korea’s leadership in the international marine observation community.

This research was funded by the Korea Institute of Marine Science & Technology Promotion (KIMST) funded by the Ministry of Oceans and Fisheries (20220546) and KIMST funded by the Ministry of Oceans and Fisheries (20220407) and Korea-China Joint Ocean Research Center. We would like to express our sincere gratitude to Euihyun Kim and Seonju Lee for their invaluable assistance and support in the preparation of this paper. Their contributions have greatly enhanced the quality of this work.

No potential conflict of interest relevant to this article was reported.