Land Cover Change in the Lower Yenisei River Using Dense Stacking of Landsat Imagery in Google Earth Engine

Abstract

Climate warming is occurring at an unprecedented rate in the Arctic due to regional amplification, potentially accelerating land cover change. Measuring and monitoring land cover change utilizing optical remote sensing in the Arctic has been challenging due to persistent cloud and snow cover issues and the spectrally similar land cover types. Google Earth Engine (GEE) represents a powerful tool to efficiently investigate these changes using a large repository of available optical imagery. This work examines land cover change in the Lower Yenisei River region of arctic central Siberia and exemplifies the application of GEE using the random forest classification algorithm for Landsat dense stacks spanning the 32-year period from 1985 to 2017, referencing 1641 images in total. The semiautomated methodology presented here classifies the study area on a per-pixel basis utilizing the complete Landsat record available for the region by only drawing from minimally cloudand snow-affected pixels. Climatic changes observed within the study area's natural environments show a statistically significant steady greening (~21,000 km 2 transition from tundra to taiga) and a slight decrease (~700 km 2 ) in the abundance of large lakes, indicative of substantial permafrost degradation. The results of this work provide an effective semiautomated classification strategy for remote sensing in permafrost regions and map products that can be applied to future regional environmental modeling of the Lower Yenisei River region.

Determination of Vegetation Thresholds for Assessing Land Use and Land Use Changes in Cambodia using the Google Earth Engine Cloud-Computing Platform

“…Landsat TOA reflectance-based EVI values were generally higher than surface reflectance (SR) based values [78] and therefore TOA data showed greater potential to provide accurate phenology-based classification [53,79,84,85]. Several studies have successfully applied TOA data for mapping the paddy rice [63,78,79,86,87] and mapping the vegetation distribution [17,22,53,84,88] using phenology-based methods. These studies have suggested that TOA data can be used for mapping with higher accuracy.…”

“…In recent years, GEE has been widely used for global-scale applications such as characterizing global forest cover change; forest expansion, loss, and gain from 2000 using large collections of Landsat scenes [5]; and crop yield estimation [21,22]. Other studies have also confirmed the ease of integrating various sources of temporal satellite imagery data and automating image classification routines for vegetation and land cover mapping using the GEE [17,[21][22][23][24][25][26].…”

“…Google Earth Engine is a web-based opensource platform for geospatial analysis, and it is currently free of charge for research and education purpose [14,[17][18][19]. Using millions of servers and the start-of-the-art cloud-computing and storage capability, the GEE has archived a large catalog of remote sensing data, which enables researchers and the scientific community to work on trillions of images in an intrinsically-parallel processing way at scale at no cost [17,19]. The GEE public data archive includes more than 40 years of historical imagery and scientific datasets, which are updated and expanded daily.…”

“…Several studies have attempted to classify the land use categories using the threshold-based approach by applying phenology methods with the aid of remote sensing data [6,22,27,[31][32][33][34][35][36]. The combined remote sensing and vegetation phenology approaches can measure heterogeneous seasonality in vegetation, because remote sensing can provide time series data at any scale; therefore, these approaches are essential for understanding the role of vegetation in climate change [17,[37][38][39][40][41][42].…”

Integration of Machine Learning and Open Access Geospatial Data for Land Cover Mapping

“…Inglada et al, provided the land cover of France using Sentinel-2 products for 17 land cover classes [7]. Also, a cloud-based platform and dense stack satellite time series were utilized to provide artic land cover [8]. Belgiu et al, investigated the ability of a Machine Learning (ML) methodology in land cover mapping in different agro-ecological regions of the planet [9].…”