scholarly journals Neural network for seasonal climate precipitation prediction on the Brazil

2020 ◽  
Vol 42 ◽  
pp. e15
Author(s):  
Juliana Aparecida Anochi ◽  
Haroldo Fraga de Campos Velho
Keyword(s):  
Neural Network ◽  
Optimization Problem ◽  
Global Precipitation Climatology Project ◽  
Reanalysis Data ◽  
Particle Collision ◽  
Monthly Precipitation ◽  
Multi Layer Perceptron ◽  
Meteorological Field ◽  
Precipitation Prediction ◽  
Global Precipitation

Precipitation is the hardest meteorological field to be predicted. An approach based on and optimal neural network is applied for climate precipitation prediction for the Brazil. A self-configurated multi-layer perceptron neural network (MLP-NN) is used as a predictor tool. The MLP-NN topology is found by solving an optimization problem by the Multi-Particle Collision Algorithm (MPCA). Prediction for Summer and Winter seasons are shown. The neural forecasting is evaluated by using the reanalysis data from the NCEP/NCAR and data from satellite GPCP (Global Precipitation Climatology Project -- monthly precipitation dataset).

scholarly journals Comparison of the GPCP and CMAP Merged Gauge–Satellite Monthly Precipitation Products for the Period 1979–2001

2004 ◽  
Vol 5 (6) ◽  
pp. 1207-1222 ◽  
Author(s):  
Xungang Yin ◽  
Arnold Gruber ◽  
Phil Arkin
Keyword(s):  
Spatial Correlation ◽  
Input Data ◽  
Sampling Error ◽  
Global Precipitation Climatology Project ◽  
Monthly Precipitation ◽  
Tropical Ocean ◽  
Microwave Sounding Unit ◽  
Microwave Sounding ◽  
Microwave Imager ◽  
Global Precipitation

Abstract The two monthly precipitation products of the Global Precipitation Climatology Project (GPCP) and the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) are compared on a 23-yr period, January 1979–December 2001. For the long-term mean, major precipitation patterns are clearly demonstrated by both products, but there are differences in the pattern magnitudes. In the tropical ocean the CMAP is higher than the GPCP, but this is reversed in the high-latitude ocean. The GPCP–CMAP spatial correlation is generally higher over land than over the ocean. The correlation between the global mean oceanic GPCP and CMAP is significantly low. It is very likely because the input data of the two products have much less in common over the ocean; in particular, the use of atoll data by the CMAP is disputable. The decreasing trend in the CMAP oceanic precipitation is found to be an artifact of input data change and atoll sampling error. In general, overocean precipitation represented by the GPCP is more reasonable; over land the two products are close, but different merging algorithms between the GPCP and the CMAP can sometimes produce substantial discrepancy in sensitive areas such as equatorial West Africa. EOF analysis shows that the GPCP and the CMAP are similar in 6 out of the first 10 modes, and the first 2 leading modes (ENSO patterns) of the GPCP are nearly identical to their counterparts of the CMAP. Input data changes [e.g., January 1986 for Geostationary Operational Environmental Satellite (GOES) precipitation index (GPI), July 1987 for Special Sensor Microwave Imager (SSM/I), May 1994 for Microwave Sounding Unit (MSU), and January 1996 for atolls] have implications in the behavior of the two datasets. Several abrupt changes identified in the statistics of the two datasets including the changes in overocean precipitation, spatial correlation time series, and some of the EOF principal components, can be related to one or more input data changes.

scholarly journals Establishing the dominant source of uncertainty in drought indicators

2013 ◽  
Vol 10 (11) ◽  
pp. 13407-13440 ◽  
Author(s):  
G. Naumann ◽  
E. Dutra ◽  
P. Barbosa ◽  
F. Pappenberger ◽  
F. Wetterhall ◽  
...  
Keyword(s):  
Potential Evapotranspiration ◽  
Standardized Precipitation Index ◽  
Global Precipitation Climatology Project ◽  
Rain Gauge ◽  
Eastern Africa ◽  
Drought Monitoring ◽  
Monthly Precipitation ◽  
Precipitation Climatology ◽  
Drought Indicators ◽  
Global Precipitation

Abstract. Drought monitoring is a key component to mitigate impacts of droughts. Lack of reliable and up-to-date datasets is a common challenge across the Globe. This study investigates different datasets and drought indicators on their capability to improve drought monitoring in Africa. The study was performed for four river basins located in different climatic regions (the Oum er-Rbia in Morocco, the Blue Nile in Eastern Africa, the Upper Niger in Western Africa, and the Limpopo in South-Eastern Africa) as well as the Greater Horn of Africa. The five precipitation datasets compared are the ECMWF ERA – Interim reanalysis, the Tropical Rainfall Measuring Mission satellite monthly rainfall product 3B43, the Global Precipitation Climatology Centre gridded precipitation dataset, the Global Precipitation Climatology Project Global Monthly Merged Precipitation Analyses, and the Climate Prediction Center Merged Analysis of Precipitation. The set of drought indicators used includes the Standardized Precipitation Index, the Standardized Precipitation–Evaporation Index, Soil Moisture Anomalies and Potential Evapotranspiration. A comparison of the annual cycle and monthly precipitation time series shows a good agreement in the timing of the rainy seasons. The main differences between the datasets are in the ability to represent the magnitude of the wet seasons and extremes. Moreover, for the areas affected by drought, all the drought indicators agree on the time of drought onset and recovery although there is disagreement on the extent of the affected area. In regions with limited rain gauge data the estimation of the different drought indicators is characterised by a higher uncertainty. Further comparison suggests that the main source of error in the computation of the drought indicators is the uncertainty in the precipitation datasets rather than the estimation of the distribution parameters of the drought indicators.

scholarly journals Discretization and machine learning approximation of BSDEs with a constraint on the Gains-process

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Idris Kharroubi ◽  
Thomas Lim ◽  
Xavier Warin
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Differential Equations ◽  
Numerical Experiments ◽  
Optimization Problem ◽  
Learning Approach ◽  
The Neural Network ◽  
Machine Learning Approach ◽  
Mesh Grid

AbstractWe study the approximation of backward stochastic differential equations (BSDEs for short) with a constraint on the gains process. We first discretize the constraint by applying a so-called facelift operator at times of a grid. We show that this discretely constrained BSDE converges to the continuously constrained one as the mesh grid converges to zero. We then focus on the approximation of the discretely constrained BSDE. For that we adopt a machine learning approach. We show that the facelift can be approximated by an optimization problem over a class of neural networks under constraints on the neural network and its derivative. We then derive an algorithm converging to the discretely constrained BSDE as the number of neurons goes to infinity. We end by numerical experiments.

Sign in / Sign up

Export Citation Format

Share Document