**6. Future Research Directions**

Remote sensing technology has been widely applied in the research of UHI and UHIRIP. The most important advantage of using remote sensing thermal data is the wallto-wall coverage of UHI patterns that can meet the needs of spatial and temporal analyses. Remotely sensed data can be used to investigate the surface temperatures of cities and urban agglomerations for various ecosystems with different climate conditions, for example tropical and sub-tropical, temperate and cold temperate, coastal and inland, and arid and semi-arid land at regional scales. These studies are needed to describe surface temperature characteristics in these specific environments and how climate change may be modulating UHI patterns. UHIRIP produces an aggregate impact on weather conditions, land use, human health, biodiversity, ecosystem security, economics, and urban planning [16,244].

Land surface temperature and emissivity retrieval (separation) has always been challenging. Generally, the LSE values needed to apply the method have been estimated from a procedure that uses the visible and near-infrared bands. The algorithm was created using the brightness temperature of the thermal and emissivity of different land cover types, derived from visible and near-infrared bands of various sensors. Compared with field-based observation, remote sensing offers the advantages of a harmonized, long-term, and spatially extensive record to observe LST change. The retrieved LSTs are verified using the near surface temperature of weather station datasets, which will help to improve the accuracy of LST derived from thermal bands. The difference between retrieved LST and Automatic Weather Station (AWS) data indicates that the technique works by giving an error of ±3 ◦C [245]. These differences can be because of the difference between the resolutions of thermal and visible bands, and a comparison was made between the point measurement (AWS data) 2 m above the surface and surface temperature (retrieved LST). Communicating the results of time-series LST studies that are based on both field weather station observations and remote-sensing time-series data to urban planners, policymakers, and the general public could help inform urban design and decision making.

Using temporally dense time series of remotely sensed data at a high spatial resolution is a growing trend in UHI and UHIRIP research, facilitated by increasing computer capabilities to handle big datasets, machine leaning, deep learning, and Google Earth Engine applications. Landsat ARD, in particular, has grea<sup>t</sup> potential to derive LST. Models used to fuse data from across multiple sensors will be developed to increase data temporal density and spatial resolution. Moreover, future sensor improvement on Landsat and aircraft thermal data are possible options. On the other hand, in order to determine the temporal variation of LST using satellite data with restricted overpass times, it appears necessary to use long-time weather station observations to investigate diurnal UHI in various ecosystems, although some new sensors (e.g., ECOSTRESS) can provide this information. Future research is anticipated to improve on methods to simultaneously derive LST and land surface emission (LSE) from hyperspectral TIR, multi spectral-temporal, and TIR-microwave data; additionally, future methods will consider aerosol and cirrus effects [18]. Another viable angle of potential future studies is urban development strategies for mitigating UHI, such as increasing vegetation and water surfaces in urban development.

Climate models are the only tools that account for the complex set of processes that will determine future climate change at both a global and regional level, and assessing regional impacts of climate change begins with the development of climate projections at relevant temporal and spatial scales [246]. The most current existing climate change modeling covers large geographic areas at regional and global levels with relatively low spatial resolutions (>10 km). In the future, LST that is derived from remotely sensed data will support climate change modeling (regional climate models and statistical downscaling models) in UHI and UHIRIP analyses in urban and surrounding areas.

Our analysis indicated that determination is still a central topic of UHI research. Modeling will continue to provide vital and useful results on the spatiotemporal assessment of UHI, especially when models more effectively combine thermal data from multiple sensors. ML (DL) and AI are continuing to grow in popularity in UHI and UHIRIP research. For time series analyses with remote sensing data, a cloud computing platform such as GEE could bring about a substantial change in UHI and UHIRIP analyses, as they have the capability to process big remote sensing datasets and assess the spatiotemporal dynamics of the area quickly. A better integration of remote sensing and station measurements into models is expected. This study also suggests that direct and indirect UHIRIP, especially human health issues, heat wave impacts, air pollution, and ecological security, will receive increasing scientific attention in the future. Research on controlling and adapting to UHI impacts may warrant special attention. The interaction of UHI and UHIRIP, and their changes to LULC based on urban planning, are actively being studied.

**Author Contributions:** Original draft preparation, H.S.; review and editing with ideas and inputs, G.X., R.A., K.G. and Q.Z. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We are grateful to the USGS National Land Imaging Program for supporting this research. We would like to thank Carol Deering for assistance in collecting and preparing the literary material. This manuscript was improved thanks to comments and suggestions from Norman Bliss, Matthew Rigge, and Thomas Adamson. Hua Shi and Qiang Zhou's work was performed under USGS contracts G13PC00028 and 140G0119C0001. For Kevin Gallo, the scientific results and conclusions, as well as any views or opinions expressed herein, are those of the author and do not necessarily reflect those of NOAA or the Department of Commerce. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

**Conflicts of Interest:** The authors declare no conflict of interest. The funding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; and in the decision to publish the results.
