Machine learning modeling of plant phenology based on coupling satellite and gridded meteorological dataset

Abstract

Changes in the timing of plant phenological phases are important proxies in contemporary climate research. However, most of the commonly used traditional phenological observations do not give any coherent spatial information. While consistent spatial data can be obtained from airborne sensors and preprocessed gridded meteorological data, not many studies robustly benefit from these data sources. Therefore, the main aim of this study is to create and evaluate different statistical models for reconstructing, predicting, and improving quality of phenological phases monitoring with the use of satellite and meteorological products. A quality-controlled dataset of the 13 BBCH plant phenophases in Poland was collected for the period 2007–2014. For each phenophase, statistical models were built using the most commonly applied regression-based machine learning techniques, such as multiple linear regression, lasso, principal component regression, generalized boosted models, and random forest. The quality of the models was estimated using a k-fold cross-validation. The obtained results showed varying potential for coupling meteorological derived indices with remote sensing products in terms of phenological modeling; however, application of both data sources improves models’ accuracy from 0.6 to 4.6 day in terms of obtained RMSE. It is shown that a robust prediction of early phenological phases is mostly related to meteorological indices, whereas for autumn phenophases, there is a stronger information signal provided by satellite-derived vegetation metrics. Choosing a specific set of predictors and applying a robust preprocessing procedures is more important for final results than the selection of a particular statistical model. The average RMSE for the best models of all phenophases is 6.3, while the individual RMSE vary seasonally from 3.5 to 10 days. Models give reliable proxy for ground observations with RMSE below 5 days for early spring and late spring phenophases. For other phenophases, RMSE are higher and rise up to 9–10 days in the case of the earliest spring phenophases.

Bulbous perennials precisely detect the length of winter and adjust flowering dates

“…However, the developing concept of machine learning (e.g. Li et al ., 2015; Czernecki et al ., 2018) provides an appropriate tool to tackle such a redundant task by unambiguously identifying the most relevant explanatory variables.…”

Comparative Quality and Trend of Remotely Sensed Phenology and Productivity Metrics across the Western United States

“…The growing popularity of platforms able to process large data quickly (such as Google Earth Engine) may render daily, global datasets derived from the DLC method readily available in the future. In terms of improving LSP estimates, the use of mechanistic models to predict key metrics [82][83][84] may help to address data quality issues and discrepancies across land cover types and ecoregions. These models couple remote sensing data with local observations or other models of elevation, temperature, precipitation, and plant phenology to improve phenology and productivity metrics.…”

Text Mining in Remotely Sensed Phenology Studies: A Review on Research Development, Main Topics, and Emerging Issues

“…The start of the season (SOS) represents the metric most investigated; its study has increased through time to such an extent that it became a major, common topic and, hence, lost its relevance. In the last decade, a large number of papers were published about spring phenology and its response to climate change [51,[71][72][73]; on the contrary, end of season (EOS) and length of season (LOS) received far less consideration. As clearly stated by [50], autumn remains a relatively disregarded season in climate change research both in temperate and arctic ecosystems, notwithstanding the role of autumn in determining the length of the growing season [74,75] and in controlling influence on the carbon-uptake period [76].…”