Analysis of Regional Distribution of Tree Species Using Multi-Seasonal Sentinel-1&2 Imagery within Google Earth Engine

Google Earth Engine and Artificial Intelligence (AI): A Comprehensive Review

Remote sensing (RS) plays an important role gathering data in many critical domains (e.g., global climate change, risk assessment and vulnerability reduction of natural hazards, resilience of ecosystems, and urban planning). Retrieving, managing, and analyzing large amounts of RS imagery poses substantial challenges. Google Earth Engine (GEE) provides a scalable, cloud-based, geospatial retrieval and processing platform. GEE also provides access to the vast majority of freely available, public, multi-temporal RS data and offers free cloud-based computational power for geospatial data analysis. Artificial intelligence (AI) methods are a critical enabling technology to automating the interpretation of RS imagery, particularly on object-based domains, so the integration of AI methods into GEE represents a promising path towards operationalizing automated RS-based monitoring programs. In this article, we provide a systematic review of relevant literature to identify recent research that incorporates AI methods in GEE. We then discuss some of the major challenges of integrating GEE and AI and identify several priorities for future research. We developed an interactive web application designed to allow readers to intuitively and dynamically review the publications included in this literature review.

Combination of Sentinel-1 and Sentinel-2 Data for Tree Species Classification in a Central European Biosphere Reserve

Microwave and optical imaging methods react differently to different land surface parameters and, thus, provide highly complementary information. However, the contribution of individual features from these two domains of the electromagnetic spectrum for tree species classification is still unclear. For large-scale forest assessments, it is moreover important to better understand the domain-specific limitations of the two sensor families, such as the impact of cloudiness and low signal-to-noise-ratio, respectively. In this study, seven deciduous and five coniferous tree species of the Austrian Biosphere Reserve Wienerwald (105,000 ha) were classified using Breiman’s random forest classifier, labeled with help of forest enterprise data. In nine test cases, variations of Sentinel-1 and Sentinel-2 imagery were passed to the classifier to evaluate their respective contributions. By solely using a high number of Sentinel-2 scenes well spread over the growing season, an overall accuracy of 83.2% was achieved. With ample Sentinel-2 scenes available, the additional use of Sentinel-1 data improved the results by 0.5 percentage points. This changed when only a single Sentinel-2 scene was supposedly available. In this case, the full set of Sentinel-1-derived features increased the overall accuracy on average by 4.7 percentage points. The same level of accuracy could be obtained using three Sentinel-2 scenes spread over the vegetation period. On the other hand, the sole use of Sentinel-1 including phenological indicators and additional features derived from the time series did not yield satisfactory overall classification accuracies (55.7%), as only coniferous species were well separated.

Improved Forest Canopy Closure Estimation Using Multispectral Satellite Imagery within Google Earth Engine

The large area estimation of forest canopy closure (FCC) using remotely sensed data is of high interest in monitoring forest changes and forest health, as well as in assessing forest ecological services. The accurate estimation of FCC over the regional or global scale is challenging due to the difficulty of sample acquisition and the slow processing efficiency of large amounts of remote sensing data. To address this issue, we developed a novel bounding envelope methodology based on vegetation indices (BEVIs) for determining vegetation and bare soil endmembers using the normalized differences vegetation index (NDVI), modified bare soil index (MBSI), and bare soil index (BSI) derived from Landsat 8 OLI and Sentinel-2 image within the Google Earth Engine (GEE) platform, then combined the NDVI with the dimidiate pixel model (DPM), one of the most commonly used spectral-based unmixing methods, to map the FCC distribution over an area of more than 90,000 km2. The key processing was the determination of the threshold parameter in BEVIs that characterizes the spectral boundary of vegetation and soil endmembers. The results demonstrated that when the threshold equals 0.1, the extraction accuracy of vegetation and bare soil endmembers is the highest with the threshold range given as (0, 0.3), and the estimated spatial distribution of FCC using both Landsat 8 and Sentinel-2 images were consistent, that is, the area with high canopy density was mainly distributed in the western mountainous region of Chifeng city. The verification was carried out using independent field plots. The proposed approach yielded reliable results when the Landsat 8 data were used (R2 = 0.6, RMSE = 0.13, and 1-rRMSE = 80%), and the accuracy was further improved using Sentinel-2 images with higher spatial resolution (R2 = 0.81, RMSE = 0.09, and 1-rRMSE = 86%). The findings demonstrate that the proposed method is portable among sensors with similar spectral wavebands, and can assist in mapping FCC at a regional scale while using multispectral satellite imagery.