An official website of the United States government
Abstract Observations of dissolved iron (dFe) in the subtropical North Atlantic revealed remarkable features: While the near‐surface dFe concentration is low despite receiving high dust deposition, the subsurface dFe concentration is high. We test several hypotheses that might explain this feature in an ocean biogeochemistry model with a refined Fe cycling scheme. These hypotheses invoke a stronger lithogenic scavenging rate, rapid biological uptake, and a weaker binding between Fe and a ubiquitous, refractory ligand. While the standard model overestimates the surface dFe concentration, a 10‐time stronger biological uptake run causes a slight reduction in the model surface dFe. A tenfold decrease in the binding strength of the refractory ligand, suggested by recent observations, starts reproducing the observed dFe pattern, with a potential impact for the global nutrient distribution. An extreme value for the lithogenic scavenging rate can also match the model dFe with observations, but this process is still poorly constrained.
more »
« less
Data-Driven Modeling of Dissolved Iron in the Global Ocean
The importance of dissolved Fe (dFe) in regulating ocean primary production and the carbon cycle is well established. However, the large-scale distribution and temporal dynamics of dFe remain poorly constrained in part due to incomplete observational coverage. In this study, we use a compilation of published dFe observations (n=32,344) with paired environmental predictors from contemporaneous satellite observations and reanalysis products to build a data-driven surface-to-seafloor dFe climatology with 1°×1° resolution using three machine-learning approaches (random forest, supper vector machine and artificial neural network). Among the three approaches, random forest achieves the highest accuracy with overall R 2 and root mean standard error of 0.8 and 0.3 nmol L -1 , respectively. Using this data-driven climatology, we explore the possible mechanisms governing the dFe distribution at various depth horizons using statistical metrics such as Pearson correlation coefficients and the rank of predictors importance in the model construction. Our results are consistent with the critical role of aeolian iron supply in enriching surface dFe in the low latitude regions and suggest a far-reaching impact of this source at depth. Away from the surface layer, the strong correlation between dFe and apparent oxygen utilization implies that a combination of regeneration, scavenging and large-scale ocean circulation are controlling the interior distribution of dFe, with hydrothermal inputs important in some regions. Finally, our data-driven dFe climatology can be used as an alternative reference to evaluate the performance of ocean biogeochemical models. Overall, the new global scale climatology of dFe achieved in our study is an important step toward improved representation of dFe in the contemporary ocean and may also be used to guide future sampling strategies.
more »
« less
- Award ID(s):
- 1840868
- PAR ID:
- 10443589
- Date Published:
- Journal Name:
- Frontiers in Marine Science
- Volume:
- 9
- ISSN:
- 2296-7745
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Heatwaves are extreme near-surface temperature events that can have substantial impacts on ecosystems and society. Early warning systems help to reduce these impacts by helping communities prepare for hazardous climate-related events. However, state-of-the-art prediction systems can often not make accurate forecasts of heatwaves more than two weeks in advance, which are required for advance warnings. We therefore investigate the potential of statistical and machine learning methods to understand and predict central European summer heatwaves on time scales of several weeks. As a first step, we identify the most important regional atmospheric and surface predictors based on previous studies and supported by a correlation analysis: 2-m air temperature, 500-hPa geopotential, precipitation, and soil moisture in central Europe, as well as Mediterranean and North Atlantic sea surface temperatures, and the North Atlantic jet stream. Based on these predictors, we apply machine learning methods to forecast two targets: summer temperature anomalies and the probability of heatwaves for 1–6 weeks lead time at weekly resolution. For each of these two target variables, we use both a linear and a random forest model. The performance of these statistical models decays with lead time, as expected, but outperforms persistence and climatology at all lead times. For lead times longer than two weeks, our machine learning models compete with the ensemble mean of the European Centre for Medium-Range Weather Forecast’s hindcast system. We thus show that machine learning can help improve subseasonal forecasts of summer temperature anomalies and heatwaves. Significance StatementHeatwaves (prolonged extremely warm temperatures) cause thousands of fatalities worldwide each year. These damaging events are becoming even more severe with climate change. This study aims to improve advance predictions of summer heatwaves in central Europe by using statistical and machine learning methods. Machine learning models are shown to compete with conventional physics-based models for forecasting heatwaves more than two weeks in advance. These early warnings can be used to activate effective and timely response plans targeting vulnerable communities and regions, thereby reducing the damage caused by heatwaves.more » « less
-
Dissolved iron (dFe) is an essential micronutrient for phytoplankton, with vanishingly low oceanic dissolved concentrations (pico- to nanomoles per kg) known to limit growth—and thus influence primary productivity and carbon cycling—over much of the surface ocean. However, because of the considerable challenges associated with contamination-free sample collection and accurate analysis of such low dFe concentrations, the first reliable dFe measurements came only in the 1980s. Further, by 2003, despite several decades of research, there were only ~25 full-depth oceanic dFe profiles worldwide, with dust considered to be the main oceanic dFe source. Since 2008, facilitated by the extensive field campaign and rigorous intercalibration of the international GEOTRACES program, there has been an “explosion” in the availability of oceanic dFe data, with hundreds of profiles now available. Concurrently, there has been a paradigm shift to a view of the marine Fe cycle where multiple sources contribute, and some forms of dFe can be transported great distances through the intermediate and deep ocean. Here, we showcase the GEOTRACES dFe datasets across the different ocean basins, synthesize our current multi-source view of the oceanic Fe cycle, spotlight sediments as an important dFe source, and look to future directions for constraining oceanic dFe boundary exchange.more » « less
-
Thenkabail, Prasad S. (Ed.)Physically based hydrologic models require significant effort and extensive information for development, calibration, and validation. The study explored the use of the random forest regression (RFR), a supervised machine learning (ML) model, as an alternative to the physically based Soil and Water Assessment Tool (SWAT) for predicting streamflow in the Rio Grande Headwaters near Del Norte, a snowmelt-dominated mountainous watershed of the Upper Rio Grande Basin. Remotely sensed data were used for the random forest machine learning analysis (RFML) and RStudio for data processing and synthesizing. The RFML model outperformed the SWAT model in accuracy and demonstrated its capability in predicting streamflow in this region. We implemented a customized approach to the RFR model to assess the model’s performance for three training periods, across 1991–2010, 1996–2010, and 2001–2010; the results indicated that the model’s accuracy improved with longer training periods, implying that the model trained on a more extended period is better able to capture the parameters’ variability and reproduce streamflow data more accurately. The variable importance (i.e., IncNodePurity) measure of the RFML model revealed that the snow depth and the minimum temperature were consistently the top two predictors across all training periods. The paper also evaluated how well the SWAT model performs in reproducing streamflow data of the watershed with a conventional approach. The SWAT model needed more time and data to set up and calibrate, delivering acceptable performance in annual mean streamflow simulation, with satisfactory index of agreement (d), coefficient of determination (R2), and percent bias (PBIAS) values, but monthly simulation warrants further exploration and model adjustments. The study recommends exploring snowmelt runoff hydrologic processes, dust-driven sublimation effects, and more detailed topographic input parameters to update the SWAT snowmelt routine for better monthly flow estimation. The results provide a critical analysis for enhancing streamflow prediction, which is valuable for further research and water resource management, including snowmelt-driven semi-arid regions.more » « less
-
null (Ed.)Coastal mangrove forests provide important ecosystem goods and services, including carbon sequestration, biodiversity conservation, and hazard mitigation. However, they are being destroyed at an alarming rate by human activities. To characterize mangrove forest changes, evaluate their impacts, and support relevant protection and restoration decision making, accurate and up-to-date mangrove extent mapping at large spatial scales is essential. Available large-scale mangrove extent data products use a single machine learning method commonly with 30 m Landsat imagery, and significant inconsistencies remain among these data products. With huge amounts of satellite data involved and the heterogeneity of land surface characteristics across large geographic areas, finding the most suitable method for large-scale high-resolution mangrove mapping is a challenge. The objective of this study is to evaluate the performance of a machine learning ensemble for mangrove forest mapping at 20 m spatial resolution across West Africa using Sentinel-2 (optical) and Sentinel-1 (radar) imagery. The machine learning ensemble integrates three commonly used machine learning methods in land cover and land use mapping, including Random Forest (RF), Gradient Boosting Machine (GBM), and Neural Network (NN). The cloud-based big geospatial data processing platform Google Earth Engine (GEE) was used for pre-processing Sentinel-2 and Sentinel-1 data. Extensive validation has demonstrated that the machine learning ensemble can generate mangrove extent maps at high accuracies for all study regions in West Africa (92%–99% Producer’s Accuracy, 98%–100% User’s Accuracy, 95%–99% Overall Accuracy). This is the first-time that mangrove extent has been mapped at a 20 m spatial resolution across West Africa. The machine learning ensemble has the potential to be applied to other regions of the world and is therefore capable of producing high-resolution mangrove extent maps at global scales periodically.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3389/fmars.2022.837183
