Assessing trends and uncertainties in satellite‐era ocean chlorophyll using space‐time modeling

Abstract

The presence, magnitude, and even direction of long-term trends in phytoplankton abundance over the past few decades are still debated in the literature, primarily due to differences in the data sets and methodologies used. Recent work has suggested that the satellite chlorophyll record is not yet long enough to distinguish climate change trends from natural variability, despite the high density of coverage in both space and time. Previous work has typically focused on using linear models to determine the presence of trends, where each grid cell is considered independently from its neighbors. However, trends can be more thoroughly evaluated using a spatially resolved approach. Here a Bayesian hierarchical spatiotemporal model is fitted to quantify trends in ocean chlorophyll from September 1997 to December 2013. The approach used in this study explicitly accounts for the dependence between neighboring grid cells, which allows us to estimate trend by "borrowing strength" from the spatial correlation. By way of comparison, a model without spatial correlation is also fitted. This results in a notable loss of accuracy in model fit. Additionally, we find an order of magnitude smaller global trend, and larger uncertainty, when using the spatiotemporal model: À0.023 ± 0.12% yr À1 as opposed to À0.38 ± 0.045% yr À1 when the spatial correlation is not taken into account. The improvement in accuracy of trend estimates and the more complete account of their uncertainty emphasize the solution that space-time modeling offers for studying global long-term change.

Assessing the Presence of Discontinuities in the Ocean Color Satellite Record and Their Effects on Chlorophyll Trends and Their Uncertainties

“…In Hammond et al (2017), trends were detected in 17 of the 23 regions, as opposed to 19 such regions in the present study. The large-scale latitudinal pattern (whereby higher latitudes tend to have more positive trends) is also similar in both studies; 16 of the 23 regions in Hammond et al (2017) have the same trend directions as the N scenario. Although there are differences between the two studies which may be partly attributable to MODIS aging effects, these are nevertheless minor and do not affect our conclusions.…”

“…The MODIS‐Aqua sensor is known to be affected by sensor aging, particularly toward the end of the study period; thus, caution is advised for temporal analysis including the post‐2012 period (Mélin et al, ). To assess whether a drift in the MODIS‐Aqua sensor may affect our results, we compare the trends detected over the period 1997–2016 in the present study to the trends detected over 1997–2013 in Hammond et al (), which uses the ESA OC‐CCI v2.0 data set with the R2013.0.1 reprocessing MODIS‐Aqua data. In Hammond et al (), trends were detected in 17 of the 23 regions, as opposed to 19 such regions in the present study.…”

“…The spTimer package in R is used to estimate the model fit (Bakar & Sahu, 2015). See the supporting information and Hammond et al (2017) for additional details on the model setup.…”

“…To assess whether a drift in the MODIS-Aqua sensor may affect our results, we compare the trends detected over the period 1997-2016 in the present study to the trends detected over 1997-2013 in Hammond et al (2017), which uses the ESA OC-CCI v2.0 data set with the R2013.0.1 reprocessing MODIS-Aqua data. In Hammond et al (2017), trends were detected in 17 of the 23 regions, as opposed to 19 such regions in the present study. The large-scale latitudinal pattern (whereby higher latitudes tend to have more positive trends) is also similar in both studies; 16 of the 23 regions in Hammond et al (2017) have the same trend directions as the N scenario.…”

Variability and 20-Year Trends in Satellite-Derived Surface Chlorophyll Concentrations in Large Marine Ecosystems around South and Western Central America

“…However, there is general consensus in that there is high global variability, with regions of increasing and decreasing trends. For example, Hammond et al (2017) detected a small global decline in chlorophyll between 1997 and 2013, but observed regional variability with both positive and negative trends present across the globe ranging from ∼−2 to 1% year −1 . Our results indicate that within the LMEs analyzed the greatest changes have occurred at the Patagonia (PLME) and Humboldt (HLME) Large Marine Ecosystems, where the largest and almost exclusively increasing trends were recorded and the largest percent areas showed significant changes.…”