Data assimilation of physical and chlorophyll a observations in the California Current System using two biogeochemical models

Mattern, Jann Paul; Song, Hajoon; Edwards, Christopher A.; Moore, Andrew M.; Fiechter, Jérôme

doi:10.1016/j.ocemod.2016.12.002

Cited by 39 publications

(43 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because the two solutions are so similar, we switch to a longer DA simulation to examine the cumulative effect of employing FDA. For this purpose, we start two longer simulations at 1 January 2013 and perform 30 consecutive 4‐day cycles each using the same 4D‐Var setup as in Mattern et al (, ; for configuration details, see long DA simulation in Table ). Because the results of the reference 4D‐Var setup have been reported thoroughly in the aforementioned studies, our description here focuses solely on the difference between the reference and FDA‐based simulations.…”

Section: Resultsmentioning

confidence: 99%

“…Tangent linear and adjoint code for the source‐minus‐sinks terms of NEMURO exists as well but is currently not part of the ROMS codebase. This code was created by hand, has been used in several studies (see Mattern et al, , ), and is used here as a reference for comparison with FDA.…”

Section: Methodsmentioning

confidence: 99%

“…The FDA of the biological source‐minus‐sink terms presented here is general, however, and suitable for use in other variational DA configurations. More details about the coupled physical‐BGC model and the DA system and results obtained from the system can be found in Mattern et al (, ); both studies use the same DA setup employed here.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

A Simple Finite Difference‐Based Approximation for Biogeochemical Tangent Linear and Adjoint Models

Mattern

Edwards

2019

JGR Oceans

Self Cite

View full text Add to dashboard Cite

We present a technique that accurately approximates tangent linear and adjoint models for data assimilation applications using only evaluations of the nonlinear model. The approximation offers a simple way to create tangent linear and adjoint model codes that are easily maintainable, as only major changes to the nonlinear model formulation necessitate modifications of the tangent linear or adjoint model code. The approach is particularly well suited to marine biogeochemical models and takes advantage of typical features of these types of models to be computationally viable. We illustrate the approximation in a realistic application, using a three-dimensional coupled physical-biogeochemical 4D-Var data assimilation system, set in the California Current system, in which the approximation is only applied to the 11 state variable biogeochemical model. In this application, the approximation-based model solution tracks the reference solution accurately over thirty 4-day assimilation cycles but leads to a ∼10% increase in the computational cost compared to the hand-coded reference.Plain Language Summary Data assimilation refers to the process by which computer simulations of the Earth's atmosphere or ocean are constrained by measured data to be closer to the natural system. For example, in weather forecasting, measurements of the atmosphere are used to improve the computer model-generated forecast. Many data assimilation applications require extra computer code to calculate the appropriate adjustments shifting the model toward observations. This data assimilation code is time consuming to build; it must be updated whenever changes are made to the model code. We describe a relatively easy way to automatically generate the data assimilation code using only the underlying model code. Using this method, changes to the model code are automatically propagated to the data assimilation code. Although this approach is simple to construct and maintain, it comes at a cost of increased computer execution time. It is therefore not sensible for all computer models. We show that our approach is useful for marine biological models that simulate biological and chemical interactions in the ocean. For our biological model of the U.S. west coast, the automatically generated data assimilation code takes 10% longer to run than the assimilation code created by hand and results in functionally equivalent results. Key Points:• A technique that accurately approximates tangent linear and adjoint models using only evaluations of the nonlinear model • Simple way to create tangent linear and adjoint code suitable for biogeochemical ocean models nonlinear model. FDA is easy to implement and takes advantage of two characteristics of typical BGC models that makes its use especially, but perhaps exclusively, appealing for these types of models. In section 1.1, we briefly examine the basics of variational DA, introduce our mathematical notation, and review prior work approximating tangent linear and adjoint models. Section 1.2 introduces BGC models and t...

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Simple Finite Difference‐Based Approximation for Biogeochemical Tangent Linear and Adjoint Models

Mattern

Edwards

2019

JGR Oceans

Self Cite

View full text Add to dashboard Cite

show abstract

“…Continued provision of satellite data as well as increased availability of subsurface measurements (e.g., from profiling floats and underwater gliders) are both crucial moving forward. Similarly, the emerging field of biophysical data assimilation (e.g., Mattern et al, 2017) currently uses primarily satellite-derived estimates of surface chlorophyll concentration. Although assimilation of these surface fields in a coupled physical-biogeochemical ocean model has shown some success in improving the chlorophyll distribution over the entire water column, as well as the surface carbon dioxide fluxes (Ford and Barciela, 2017), the availability of vertical profiles of variables including chlorophyll concentration and irradiance appears to be extremely important to achieve an accurate estimation of primary production in the upper ocean.…”

Section: Ocean Data Assimilationmentioning

confidence: 99%

Observational Needs Supporting Marine Ecosystems Modeling and Forecasting: From the Global Ocean to Regional and Coastal Systems

et al. 2019

View full text Add to dashboard Cite

Many coastal areas host rich marine ecosystems and are also centers of economic activities, including fishing, shipping and recreation. Due to the socioeconomic and ecological importance of these areas, predicting relevant indicators of the ecosystem state on sub-seasonal to interannual timescales is gaining increasing attention. Depending on the application, forecasts may be sought for variables and indicators spanning physics (e.g., sea level, temperature, currents), chemistry (e.g., nutrients, oxygen, pH), and biology (from viruses to top predators). Many components of the marine ecosystem are known to be influenced by leading modes of climate variability, which provide a physical basis for predictability. However, prediction capabilities remain limited by the lack of a clear understanding of the physical and biological processes Frontiers in Marine Science | www.frontiersin.org 1 October 2019 | Volume 6 | Article 623 Capotondi et al. Marine Ecosystems Modeling and Forecastinginvolved, as well as by insufficient observations for forecast initialization and verification. The situation is further complicated by the influence of climate change on ocean conditions along coastal areas, including sea level rise, increased stratification, and shoaling of oxygen minimum zones. Observations are thus vital to all aspects of marine forecasting: statistical and/or dynamical model development, forecast initialization, and forecast validation, each of which has different observational requirements, which may be also specific to the study region. Here, we use examples from United States (U.S.) coastal applications to identify and describe the key requirements for an observational network that is needed to facilitate improved process understanding, as well as for sustaining operational ecosystem forecasting. We also describe new holistic observational approaches, e.g., approaches based on acoustics, inspired by Tara Oceans or by landscape ecology, which have the potential to support and expand ecosystem modeling and forecasting activities by bridging global and local observations.

show abstract

“…Identification of the best parameters to use in the model usually requires a lengthy process of fine-tuning often carried out manually and occasionally in an automated fashion (e.g., Mattern et al, 2016). Both approaches attempt to reproduce 5 available observations of state variables; due to its continuous acquisition and spatial coverage, remotely sensed chlorophyll is the most abundant BGC data set for marine ecosystem model evaluation.…”

Section: Introductionmentioning

confidence: 99%

A High-resolution Biogeochemical Model (ROMS 3.4 + bio_Fennel) of the East Australian Current System

Rocha¹,

Edwards²,

Roughan³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Understanding phytoplankton dynamics is critical across a range of topics, spanning from fisheries management to climate change mitigation. It is particularly interesting in the East Australian Current (EAC) System, as the region's eddy 10 field strongly conditions nutrient availability and, therefore, phytoplankton growth. Numerical models provide unparalleled insight into these biogeochemical dynamics. Yet, to date, modelling efforts off southeastern Australia have either targeted case studies (small spatial and temporal scales) or encompassed the whole EAC System but focused on climate change effects at the mesoscale (with a spatial resolution of 1/10º). Here we couple a model of the pelagic nitrogen cycle (bio_Fennel) to a 10-year high-resolution (2.5 -5 km horizontal) three-dimensional ocean model (ROMS) to resolve both 15 regional and finer scale biogeochemical processes occurring in the EAC System. We use several statistical metrics to compare the simulated surface chlorophyll to an ocean colour dataset (Copernicus-GlobColour) for the 2003-2011 period and show that the model can reproduce the observed phytoplankton surface patterns with a domain-wide rmse of approximately 0.2 mg chla m -3 and a correlation coefficient of 0.76. This coupled configuration will provide a much-needed framework to examine phytoplankton variability in the EAC System providing insight into important ecosystem dynamics 20 such as regional nutrient supply mechanisms and biogeochemical cycling occurring in EAC eddies.Geosci. Model Dev. Discuss., https://doi.

show abstract

Data assimilation of physical and chlorophyll a observations in the California Current System using two biogeochemical models

Cited by 39 publications

References 40 publications

A Simple Finite Difference‐Based Approximation for Biogeochemical Tangent Linear and Adjoint Models

A Simple Finite Difference‐Based Approximation for Biogeochemical Tangent Linear and Adjoint Models

Observational Needs Supporting Marine Ecosystems Modeling and Forecasting: From the Global Ocean to Regional and Coastal Systems

A High-resolution Biogeochemical Model (ROMS 3.4 + bio_Fennel) of the East Australian Current System

Contact Info

Product

Resources

About