Dataset Information

Precipitation isotope time series predictions from machine learning applied in Europe.

ABSTRACT: 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/.

SUBMITTER: Nelson DB

PROVIDER: S-EPMC8256050 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ImportanceForecasting the volume of hospital discharges has important implications for resource allocation and represents an opportunity to improve patient safety at periods of elevated risk.ObjectiveTo determine the performance of a new time-series machine learning method for forecasting hospital discharge volume compared with simpler methods.DesignA retrospective cohort study of daily hospital discharge volumes at 2 large, New England academic medical centers between January 1, 2005, and December 31, 2014 (hospital 1), or January 1, 2005, and December 31, 2010 (hospital 2), comparing time-series forecasting methods for prediction was performed. Data analysis was conducted from February 28, 2017, to August 30, 2018. Group-level data for all discharges from inpatient units were included. In addition to conventional methods, a technique originally developed for allocating data center resources, and comparison strategies for incorporating prior data and frequency of model updates, was conducted to identify the model application that optimized forecast accuracy.Main outcomes and measuresModel calibration as measured by R2 and, secondarily, number of days with errors greater than 1 SD of daily volume.ResultsDuring the forecasted year, hospital 1 had 54?411 discharges (daily mean, 149) and hospital 2 had 47?456 discharges (daily mean, 130). The machine learning method was well calibrated at both sites (R2, 0.843 and 0.726, respectively) and made errors greater than 1 SD of daily volume on only 13 and 22 days, respectively, of the forecast year at the 2 sites. Last-value-carried-forward models performed somewhat less well (calibration R2, 0.781 and 0.596, respectively) with 13 and 46 errors of 1 SD or greater, respectively. More frequent retraining and training sets of longer than 1 year had minimal effects on the machine learning method's performance.Conclusions and relevanceVolume of hospital discharges can perhaps be reliably forecasted using simple carry-forward models as well as methods drawn from machine learning. The benefit of the latter does not appear to be dependent on extensive training data and may enable forecasts up to 1 year in advance with superior absolute accuracy to carry-forward models.

Dataset Information

Precipitation isotope time series predictions from machine learning applied in Europe.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets