scholarly journals Precipitation isotope time series predictions from machine learning applied in Europe

2021 ◽  
Vol 118 (26) ◽  
pp. e2024107118
Author(s):  
Daniel B. Nelson ◽  
David Basler ◽  
Ansgar Kahmen
Keyword(s):  
Machine Learning ◽  
Time Series ◽  
Climate Data ◽  
Modeling Framework ◽  
History Of ◽  
Oxygen Isotopic ◽  
Input Variables ◽  
Hydrogen And Oxygen Isotope ◽  
High Level

Hydrogen and oxygen isotope values of precipitation are critically important quantities for applications in Earth, environmental, and biological sciences. However, direct measurements are not available at every location and time, and existing precipitation isotope models are often not sufficiently accurate for examining features such as long-term trends or interannual variability. This can limit applications that seek to use these values to identify the source history of water or to understand the hydrological or meteorological processes that determine these values. We developed a framework using machine learning to calculate isotope time series at monthly resolution using available climate and location data in order to improve precipitation isotope model predictions. Predictions from this model are currently available for any location in Europe for the past 70 y (1950–2019), which is the period for which all climate data used as predictor variables are available. This approach facilitates simple, user-friendly predictions of precipitation isotope time series that can be generated on demand and are accurate enough to be used for exploration of interannual and long-term variability in both hydrogen and oxygen isotopic systems. These predictions provide important isotope input variables for ecological and hydrological applications, as well as powerful targets for paleoclimate proxy calibration, and they can serve as resources for probing historic patterns in the isotopic composition of precipitation with a high level of meteorological accuracy. Predictions from our modeling framework, Piso.AI, are available at https://isotope.bot.unibas.ch/PisoAI/.

scholarly journals International agricultural trade forecasting using machine learning

Data & Policy ◽  
2021 ◽  
Vol 3 ◽  
Author(s):  
Munisamy Gopinath ◽  
Feras A. Batarseh ◽  
Jayson Beckman ◽  
Ajay Kulkarni ◽  
Sei Jeong
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Agricultural Trade ◽  
Supervised Machine Learning ◽  
Trade Patterns ◽  
Subjective Assessments ◽  
History Of ◽  
Network Approaches ◽  
Traditional Approaches

Abstract Focusing on seven major agricultural commodities with a long history of trade, this study employs data-driven analytics to decipher patterns of trade, namely using supervised machine learning (ML), as well as neural networks. The supervised ML and neural network techniques are trained on data until 2010 and 2014, respectively. Results show the high relevance of ML models to forecasting trade patterns in near- and long-term relative to traditional approaches, which are often subjective assessments or time-series projections. While supervised ML techniques quantified key economic factors underlying agricultural trade flows, neural network approaches provide better fits over the long term.

Long-term Cyclicities in Earthquake Energy Release and Major River Flow Volumes in Virginia and Missouri Seismic Zones

1988 ◽  
Vol 59 (4) ◽  
pp. 279-283 ◽  
Author(s):  
G. A. Bollinger ◽  
J. K. Costain
Keyword(s):  
Time Series ◽  
River Flow ◽  
Fluid Pressure ◽  
Fracture Permeability ◽  
Intraplate Earthquakes ◽  
Seismic Zones ◽  
Straight Line ◽  
History Of ◽  
Major River

Abstract We have investigated the time series for earthquake strain energy releases and flow volumes for the major rivers that bisect the regions of seismicity in Virginia (Giles County; central Virginia) and Missouri (New Madrid) seismic zones. Our procedure is to integrate with respect to time over data lengths up to 70 years duration and then to subtract a least squares straight-line fit. The resulting residual earthquake and flow volume time series and their spectral densities both exhibit dominant periods in the 20–30 year range. These common cyclities lend support for an important role of water in intraplate seismogenesis. The fracture permeability of crystalline rocks, caused by a long history of compressional and extensional tectonic episodes, together with the driving potential supplied by long-term cyclical variations in streamflow, can result in the diffusion of fluid pressure transients to focal depths as deep as 20 km. At those depths there is also present a quasi-static, hydrolytic weakening effect of water on asperities present in the fault zones. This combination of mechanical and chemical effects can cause intraplate earthquakes in highly-stressed crustal volumes.

Towards a long term global snow climate data record from satellite data generated within the Snow Climate Change Initiative

2020 ◽  
Author(s):  
Gabriele Schwaizer ◽  
Lars Keuris ◽  
Thomas Nagler ◽  
Chris Derksen ◽  
Kari Luojus ◽  
...  
Keyword(s):  
Time Series ◽  
Satellite Data ◽  
Snow Water Equivalent ◽  
Climate Model ◽  
Process Models ◽  
Climate Data ◽  
Seasonal Snow ◽  
Snow Water

<p>Seasonal snow is an important component of the global climate system. It is highly variable in space and time and sensitive to short term synoptic scale processes and long term climate-induced changes of temperature and precipitation. Current snow products derived from various satellite data applying different algorithms show significant discrepancies in extent and snow mass, a potential source for biases in climate monitoring and modelling. The recently launched ESA CCI+ Programme addresses seasonal snow as one of 9 Essential Climate Variables to be derived from satellite data.</p><p>In the snow_cci project, scheduled for 2018 to 2021 in its first phase, reliable fully validated processing lines are developed and implemented. These tools are used to generate homogeneous multi-sensor time series for the main parameters of global snow cover focusing on snow extent and snow water equivalent. Using GCOS guidelines, the requirements for these parameters are assessed and consolidated using the outcome of workshops and questionnaires addressing users dealing with different climate applications. Snow extent product generation applies algorithms accounting for fractional snow extent and cloud screening in order to generate consistent daily products for snow on the surface (viewable snow) and snow on the surface corrected for forest masking (snow on ground) with global coverage. Input data are medium resolution optical satellite images (AVHRR-2/3, AATSR, MODIS, VIIRS, SLSTR/OLCI) from 1981 to present. An iterative development cycle is applied including homogenisation of the snow extent products from different sensors by minimizing the bias. Independent validation of the snow products is performed for different seasons and climate zones around the globe from 1985 onwards, using as reference high resolution snow maps from Landsat and Sentinel- 2as well as in-situ snow data following standardized validation protocols.</p><p>Global time series of daily snow water equivalent (SWE) products are generated from passive microwave data from SMMR, SSM/I, and AMSR from 1978 onwards, combined with in-situ snow depth measurements. Long-term stability and quality of the product is assessed using independent snow survey data and by intercomparison with the snow information from global land process models.</p><p>The usability of the snow_cci products is ensured through the Climate Research Group, which performs case studies related to long term trends of seasonal snow, performs evaluations of CMIP-6 and other snow-focused climate model experiments, and applies the data for simulation of Arctic hydrological regimes.</p><p>In this presentation, we summarize the requirements and product specifications for the snow extent and SWE products, with a focus on climate applications. We present an overview of the algorithms and systems for generation of the time series. The 40 years (from 1980 onwards) time series of daily fractional snow extent products from AVHRR with 5 km pixel spacing, and the 20-year time series from MODIS (1 km pixel spacing) as well as the coarse resolution (25 km pixel spacing) of daily SWE products from 1978 onwards will be presented along with first results of the multi-sensor consistency checks and validation activities.</p>

A Long-Run Perspective on Comparative Advantage

2013 ◽  
Author(s):  
Giovanni Federico ◽  
Nikolaus Wolf
Keyword(s):  
Foreign Trade ◽  
Comparative Advantage ◽  
Commodity Markets ◽  
High Concentration ◽  
Long Run ◽  
Trading Partners ◽  
History Of ◽  
High Level ◽  
Global Commodity Markets

The history of Italy since its unification in 1861 was accompanied by a dramatic increase in the country's integration with European and global commodity markets: foreign trade in the long run grew on average faster than the overall economy. Italy's comparative advantage changed fundamentally, from a high concentration of a few trading partners and a handful of rather simple commodities, into a wide diversification of trading partners and more sophisticated commodities. The chapter uses a new long-term database on Italian foreign trade at a high level of disaggregation to document and analyze these changes. The chapter concludes with an assessment of Italy's prospects from a historical perspective.

Long-Term Radioactivity Release from Solidified High-Level Waste - Part II Parametric Study of Waste form Properties, Temperature and Time

1982 ◽  
Vol 15 ◽  
Author(s):  
Friedrich K. Altenhein ◽  
Werner Lutze ◽  
Rodney C. Ewing
Keyword(s):  
Parametric Study ◽  
Computer Code ◽  
Salt Dome ◽  
Waste Form ◽  
Glass Block ◽  
High Level Waste ◽  
Surface Layers ◽  
History Of ◽  
High Level

The computer code QTERM has been used to calculate the total released activity from a single glass block when in contact with brine in a salt dome repository as a function of: (1) waste form properties, (2) leaching mechanisms, (3) retention or precipitation of specific radionuclides in surface layers, (4) thermal history of the repository and (5) decreasing activity as a function of time.

scholarly journals Evaluation of Machine Learning Techniques for Daily Reference Evapotranspiration Estimation

2019 ◽  
Author(s):  
Ali Rashid Niaghi ◽  
Oveis Hassanijalilian ◽  
Jalal Shiri
Keyword(s):  
Machine Learning ◽  
Reference Evapotranspiration ◽  
Meteorological Data ◽  
Gene Expression Programming ◽  
Machine Learning Techniques ◽  
Limiting Factors ◽  
Support Vector ◽  
Daily Maximum ◽  
Climate Data ◽  
Input Variables

The ASCE-EWRI reference evapotranspiration (ETo) equation is recommended as a standardized method for reference crop ETo estimation. However, various climate data as input variables to the standardized ETo method are considered limiting factors in most cases and restrict the ETo estimation. This paper assessed the potential of different machine learning (ML) models for ETo estimation using limited meteorological data. The ML models used to estimate daily ETo included Gene Expression Programming (GEP), Support Vector Machine (SVM), Multiple Linear Regression (LR), and Random Forest (RF). Three input combinations of daily maximum and minimum temperature (Tmax and Tmin), wind speed (W) with Tmax and Tmin, and solar radiation (Rs) with Tmax and Tmin were considered using meteorological data during 2003–2016 from six weather stations in the Red River Valley. To understand the performance of the applied models with the various combinations, station, and yearly based tests were assessed with local and spatial approaches. Considering the local and spatial approaches analysis, the LR and RF models illustrated the lowest rate of improvement compared to GEP and SVM. The spatial RF and SVM approaches showed the lowest and highest values of the scatter index as 0.333 and 0.457, respectively. As a result, the radiation-based combination and the RF model showed the best performance with higher accuracy for all stations either locally or spatially, and the spatial SVM and GEP illustrated the lowest performance among models and approaches.

Pesticides and Human Health

2016 ◽  
Author(s):  
Pierluigi Cocco
Keyword(s):  
Agricultural Land ◽  
Allergic Diseases ◽  
Complex Pattern ◽  
History Of ◽  
Exposure Patterns ◽  
Multiple Treatments ◽  
High Level ◽  
Changes Over Time ◽  
Developed World

The fight against agricultural and household pests accompanies the history of humanity, and a total ban on the use of pesticides seems unlikely to happen in the foreseeable future. Currently, about 100,000 different chemicals, inorganic and organic, are currently in the market, grouped according to their function as insecticides, herbicides, fungicides, fumigants, rodenticides, fertilizers, growth regulators, etc. against specific pests, such as snails or human parasites, or their chemical structure—organochlorines, organophosphates, pyrethroids, carbamates, dithiocarbamates, organotin compounds, phthalimides, phenoxy acids, heterocyclic azole compounds, coumarins, etc. Runoff from agricultural land and rain precipitation and dry deposition from the atmosphere can extend exposure to the general environment through the transport of pesticides to streams and ground-water. Also, the prolonged bio-persistence of organochlorines generates their accumulation in the food chain, and their atmospheric drift toward remote geographical areas is mentioned as the cause of elevated fat contents in Arctic mammals. Current regulation in the developed world and the phasing out of more toxic pesticides have greatly reduced the frequency of acute intoxications, although less stringent regulations in the developing world contribute to a complex pattern of exposure circumstances worldwide. Nonetheless, evidence is growing about long-term health effects following high-level, long-lasting exposure to specific pesticides, including asthma and other allergic diseases, immunotoxicity, endocrine disruption, cancer, and central and peripheral nervous system effects. Major reasons for uncertainty in interpreting epidemiological findings of pesticide effects include the complex pattern of overlapping exposure due to multiple treatments applied to different crops and their frequent changes over time to overcome pest resistance. Further research will have to address specific agrochemicals with well-characterized exposure patterns.

Applications of Artificial Intelligence in the Realm of Business Intelligence

2021 ◽  
pp. 358-386
Author(s):  
Prakhar Mehrotra
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Time Series ◽  
Natural Language Processing ◽  
Language Processing ◽  
Business Intelligence ◽  
Machine Learning Techniques ◽  
Current State ◽  
Learning Techniques ◽  
High Level

The objective of this chapter is to discuss the integration of advancements made in the field of artificial intelligence into the existing business intelligence tools. Specifically, it discusses how the business intelligence tool can integrate time series analysis, supervised and unsupervised machine learning techniques and natural language processing in it and unlock deeper insights, make predictions, and execute strategic business action from within the tool itself. This chapter also provides a high-level overview of current state of the art AI techniques and provides examples in the realm of business intelligence. The eventual goal of this chapter is to leave readers thinking about what the future of business intelligence would look like and how enterprise can benefit by integrating AI in it.

Sign in / Sign up

Export Citation Format

Share Document