Green LAI Mapping and Cloud Gap-Filling Using Gaussian Process Regression in Google Earth Engine.

Luca Pipia,Matías Salinero-Delgado,Jochem Verrelst,Eatidal Amin,Santiago Belda

doi:10.3390/rs13030403

Abstract

For the last decade, Gaussian process regression (GPR) proved to be a competitive machine learning regression algorithm for Earth observation applications, with attractive unique properties such as band relevance ranking and uncertainty estimates. More recently, GPR also proved to be a proficient time series processor to fill up gaps in optical imagery, typically due to cloud cover. This makes GPR perfectly suited for large-scale spatiotemporal processing of satellite imageries into cloud-free products of biophysical variables. With the advent of the Google Earth Engine (GEE) cloud platform, new opportunities emerged to process local-to-planetary scale satellite data using advanced machine learning techniques and convert them into gap-filled vegetation properties products. However, GPR is not yet part of the GEE ecosystem. To circumvent this limitation, this work proposes a general adaptation of GPR formulation to parallel processing framework and its integration into GEE. To demonstrate the functioning and utility of the developed workflow, a GPR model predicting green leaf area index (LAIG) from Sentinel-2 imagery was imported. Although by running this GPR model into GEE any corner of the world can be mapped into LAIG at a resolution of 20 m, here we show some demonstration cases over western Europe with zoom-ins over Spain. Thanks to the computational power of GEE, the mapping takes place on-the-fly. Additionally, a GPR-based gap filling strategy based on pre-optimized kernel hyperparameters is also put forward for the generation of multi-orbit cloud-free LAIG maps with an unprecedented level of detail, and the extraction of regularly-sampled LAIG time series at a pixel level. The ability to plugin a locally-trained GPR model into the GEE framework and its instant processing opens up a new paradigm of remote sensing image processing.

Highlights

The estimation of quantitative vegetation variables is fundamental to assess the dynamic response of vegetation to changing environmental conditions [1]
This paper presents a workflow for the implementation of Gaussian process regression (GPR) models into the Google Earth Engine (GEE) cloud platform
While GPR has emerged as a powerful machine learning regression algorithm for processing optical satellite data into biophysical variables, it is not yet part of the GEE ecosystem

Summary

Introduction

The estimation of quantitative vegetation variables is fundamental to assess the dynamic response of vegetation to changing environmental conditions [1]. Earth observation sensors in the optical domain enable the spatiotemporally-explicit retrieval of plant biophysical parameters [2]. Since the advent of optical remote sensing science, a variety of retrieval methods for vegetation attribute extraction emerged. Quantification of surface biophysical variables from spectral data always relies on a model, enabling the interpretation of spectral observations and their translation into a surface biophysical variable. These retrieval models can be classified into the following four categories: (1) parametric regression, e.g., spectral indices combined with a fitting function, (2) non-parametric regression, e.g., machine learning regression algorithms (MLRAs), (3) physically-based, i.e., inverting radiative transfer models (RTMs), and (4) hybrid methods. See [3,4] for a comprehensive review of these methods and mapping applications

Methods

Findings

Discussion

Conclusion