首页|Researcher from Lawrence Berkeley National Laboratory Describes Findings in Mach ine Learning (Hybrid Machine Learning and Geostatistical Methods for Gap Filling and Predicting Solar-Induced Fluorescence Values)
Researcher from Lawrence Berkeley National Laboratory Describes Findings in Mach ine Learning (Hybrid Machine Learning and Geostatistical Methods for Gap Filling and Predicting Solar-Induced Fluorescence Values)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Researchers detail new data in artific ial intelligence. According to news reporting out of Berkeley, California, by Ne wsRx editors, research stated, “Sun-induced chlorophyll fluorescence (SIF) has p roven to be advantageous in estimating gross primary production, despite the lac k of a stable relationship. Satellite-based SIF measurements at Level 2 offer co mprehensive global coverage and are available in near real time.” Our news editors obtained a quote from the research from Lawrence Berkeley Natio nal Laboratory: “However, these measurements are often limited by spatial and te mporal sparsity, as well as discontinuities. These limitations primarily arise f rom incomplete satellite trajectories. Additionally, variability in cloud cover and periodic issues specific to the instruments can compromise data quality. Two families of methods have been developed to address data discontinuity: (1) mach ine learning-based gap-filling techniques and (2) geostatistical techniques (var ious forms of kriging). The former techniques utilize the relationships between ancillary data and SIF, while the latter usually rely on the available SIF data recordings and their covariance structure to provide estimates at unsampled loca tions. In this study, we create a synthetic approach for SIF gap filling by hybr idizing the two approaches under the umbrella of kriging with external drift. We performed leave-one-out cross-validation of the OCO-2 SIF retrieval aggregates for the entire year of 2019, comparing three methods: ordinary kriging, ML-based estimation using ancillary data, and kriging with external drift. The Mean Abso lute Error (MAE) for ML, ordinary kriging, and the hybrid approach was found to be 0.1399, 0.1318, and 0.1183 mW m2 sr-1 nm-1, respectively.”
Lawrence Berkeley National LaboratoryB erkeleyCaliforniaUnited StatesNorth and Central AmericaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
