**4. Discussion**

In this paper, by retrieving the relevant literature on remote sensing monitoring protected areas, we revealed hidden knowledge underlying this significant body of research. The number of publications shows a trend of continuous growth, demonstrating that more and more scholars have paid attention to this research field. From the perspective of subject categories, environmental sciences ranked first, followed by remote sensing and ecology, which shows that the remote sensing monitoring of PAs is a field closely related to environmental science, remote sensing, and ecology. The top three journals are all well-known journals in the field of remote sensing, include Remote Sensing of Environment, the International Journal of Remote Sensing, and the ISPRS Journal of Photogrammetry and Remote Sensing.

For country of origin, the USA is in the leading position. Moreover, the top 20 countries are mostly European countries. When considering institutions, the Chinese Academy of Sciences published the largest number of papers. The United States has the largest number of research institutions in the top 15, accounting for more than half of them. Through the co-authorship analysis of countries and institutions, this study determined that the USA was at the center of international cooperation, and the cooperation among national research institutions was relatively close, while its international cooperation was relatively less prevalent, which is not conducive to the long-term development of remote sensing monitoring for PAs. Countries and institutions should strengthen their knowledge exchanges and cooperation to more e ffectively discuss research trends and research status in the research field by holding relevant academic forums and conferences.

The analysis results showed that studies have mainly concentrated on terrestrial PAs, while literature on MPA monitoring is relatively less common. Future research should make full use of new monitoring technology and methods to establish a long-term, scientific, and systematic monitoring system and thereby provide a data-based foundation for evaluating the e ffectiveness of MPAs. Based on the changes in keywords, it can be seen that remote sensing monitoring of PAs mainly focused on vegetation classification, landscape pattern analysis, biodiversity protection, and the monitoring of changes in PAs. Future research trends will focus on the impact of climate change on PAs.

Considering temporal variation, the use of Landsat, MODIS, and LiDAR show a clear fluctuating and increasing trend. LiDAR and SAR have been increasingly used to monitor and evaluate the landscape in recent years. Di fferent satellites and sensors are now applied in di fferent fields and at di fferent scales of PAs. For example, Landsat products include the Thematic Mapper (TM), Enhanced Thematic Mapper Plus (ETM+), and Operational Land Imager (OLI), which can be used to monitor vegetation dynamics and assess land-cover/land-use change. However, MODIS sensors are more appropriate for vegetation phenology and forest fire monitoring and can provide high temporal resolution time series data at the landscape, regional, and global spatial scales. LiDAR makes it possible to estimate tree height, biomass, and leaf area index in large areas of the world [69,70]. SAR facilitates the estimation of forest biomass and tree height at small and medium scale [71]. Furthermore, SPOT or QuickBird may be used for species or specific vegetation change monitoring [72–74]. AVHRR sensors are mainly used to analyze the impact of climate change on vegetation coverage in PAs. The high spatial resolution of ASTER can also be used to study land- cover/land-use change in PAs [75]. Other high-resolution satellites, such as IKONOS and WorldView, focus on mapping vegetation types or habitat associated with endangered fauna [76,77].

In recent years, with the rapid development of satellites, sensors, and techniques, the applications of remote sensing have been broadly employed in the monitoring and managemen<sup>t</sup> of PAs. The relevant research results for improving the level of monitoring in PAs, formulating di fferentiated regional protection policies, and guiding sustainable development play an important role. According to this bibliometric analysis, research on the remote sensing monitoring of PAs has mainly focused on the inventory and classification of vegetation, change detection, habitat degradation, the impact of climate change, and biodiversity conservation. Among the various methods, classification, time series analysis, and model methods were the most frequently used types for PA monitoring. In the foreseeable future, there will be more new methods to monitor PAs. For example, big data approaches are being adopted to process large amounts of remote sensing data [78–80].

There are still some limitations to this study. Firstly, the single database that we used does not index all scientific journals and theme books, which could exclude some relevant articles. For example, some gray literature on this topic from governmen<sup>t</sup> agencies, nature conservancies and other non-profits might have been excluded. Expanding the search across multiple databases, such as Scopus and Google Scholar, will help reduce omissions in the analysis. Secondly, setting 1991 as the starting time may omit some earlier studies. However, most articles relevant to remote sensing applications in PAs were published in professional journals from 1991 onwards. Therefore, we believe that using 1991 as the starting point is still representative and appropriate. Thirdly, the VOSviewer software has some functional restrictions. Another consideration is that other bibliometric analysis tools, such as Citespace, could be applied in combination with VOSviewer in the future to more extensively cover the published research on the remote sensing of PAs. In the meantime, we acknowledge that it is almost impossible to include all remote sensing applications for PAs by limiting the search to include "remote sensing" alone. Other terms and descriptions, such as "land-cover monitoring", "landscape configuration and

composition", "habitat analysis", "biodiversity conservation", and "bathymetry assessment", among other examples, could be very relevant to studies in protected areas with remote sensing applications, but could be missed in the analysis. This is particularly true for monitoring changing terrestrial and marine environments under impacts from the natural and anthropogenic disturbances of protected areas. This challenge might be resolved when searches for bibliometric analysis are able to include into the full contents of published articles through the use of improved technologies, such as big data and artificial intelligence, instead of using limited and selected combination of keywords.
