A Prediction Method for Soil Moisture Time Series

One of the limitations of presently available Synthetic Aperture Radar (SAR) surface soil moisture (SSM) products is their moderated temporal resolution (e.g., 3-4 days) that is non optimal for several applications, as most user requirements point to a temporal resolution of 1-2 days or less. A possible path to tackle this issue is to coordinate multi-mission SAR acquisitions with a view to the future Copernicus Sentinel-1 (C&D and Next Generation) and L-band Radar Observation System for Europe (ROSE-L).In this respect, the recent agreement between the Japanese (JAXA) and European (ESA) Space Agencies on the use of SAR Satellites in Earth Science and Applications provides a framework to develop and validate multi-frequency and multi-platform SAR SSM products. In 2019 and 2020, to support insights on the interoperability between C- and L-band SAR observations for SSM retrieval, Sentinel-1 and ALOS-2 systematic acquisitions over the TERENO (Terrestrial Environmental Observatories) Selhausen (Germany) and Apulian Tavoliere (Italy) cal/val sites were gathered. Both sites are well documented and equipped with hydrologic networks.The objective of this study is to investigate the integration of multi-frequency SAR measurements for a consistent and harmonized SSM retrieval throughout the error characterization of a combined C- and L-band SSM product. To this scope, time series of Sentinel-1 IW and ALOS-2 FBD data acquired over the two sites will be analysed. The short time change detection (STCD) algorithm, developed, implemented and recently assessed on Sentinel-1 data [e.g., Balenzano et al., 2020; Mattia et al., 2020], will be tailored to the ALOS-2 data. Then, the time series of SAR SSM maps from each SAR system will be derived separately and aggregated in an interleaved SSM product. Furthermore, it will be compared against in situ SSM data systematically acquired by the ground stations deployed at both sites. The study will assess the interleaved SSM product and evaluate the homogeneous quality of C- and L-band SAR SSM maps.&#160;&#160;ReferencesBalenzano. A., et al., &#8220;Sentinel-1 soil moisture at 1km resolution: a validation study&#8221;, submitted to Remote Sensing of Environment (2020).Mattia, F., A. Balenzano, G. Satalino, F. Lovergine, A. Loew, et al., &#8220;ESA SEOM Land project on Exploitation of Sentinel-1 for Surface Soil Moisture Retrieval at High Resolution,&#8221; final report, contract number 4000118762/16/I-NB, 2020.

Download Full-text

A Novel Multivariable Nonlinear Time Series Prediction Method for APSO-Elman Network

2019 IEEE 8th Data Driven Control and Learning Systems Conference (DDCLS) ◽

10.1109/ddcls.2019.8909014 ◽

2019 ◽

Author(s):

Ke Ren ◽

Yongjian Wang ◽

Bo Yang ◽

Hongguang Li

Keyword(s):

Time Series ◽

Time Series Prediction ◽

Prediction Method ◽

Nonlinear Time Series ◽

Elman Network

Download Full-text

A geographical location prediction method based on continuous time series Markov model

PLoS ONE ◽

10.1371/journal.pone.0207063 ◽

2018 ◽

Vol 13 (11) ◽

pp. e0207063 ◽

Cited By ~ 5

Author(s):

Yongping Du ◽

Chencheng Wang ◽

Yanlei Qiao ◽

Dongyue Zhao ◽

Wenyang Guo

Keyword(s):

Time Series ◽

Markov Model ◽

Continuous Time ◽

Geographical Location ◽

Prediction Method ◽

Location Prediction

Download Full-text

A single-station prediction model as a contribution to instantaneous mapping

Annals of Geophysics ◽

10.4401/ag-4222 ◽

1994 ◽

Vol 37 (2) ◽

Author(s):

I. Stanislawska

Keyword(s):

Time Series ◽

Prediction Model ◽

Prediction Method ◽

Physical Processes ◽

The Past ◽

Ionospheric Station ◽

Single Station ◽

Daily Forecast

The paper presents two opposite approaches for single-station prediction and forecast. Both methods are based on different assumptions of physical processes in the ionosphere and need the different set of incoming data. Different heliogeophysical data, mainly f0F2 parameters from the past were analyzed for f0F2 obtaining for the requested period ahead. In the first method - the autocovariance prediction method - the time series of f0F2 from one station are used for daily forecast at that point. The second method may be used for obtaining f0F2 not only at the particular ionospheric station, but also at any point within the considered area.

Download Full-text

Financial Time Series Image Algorithm Based on Wavelet Analysis and Data Fusion

Journal of Sensors ◽

10.1155/2021/5577852 ◽

2021 ◽

Vol 2021 ◽

pp. 1-11

Author(s):

Wuwei Liu ◽

Jingdong Yan

Keyword(s):

Time Series ◽

Wavelet Transform ◽

Mean Square Error ◽

Financial Time Series ◽

Prediction Method ◽

Mean Square ◽

Rbf Network ◽

Network Prediction ◽

The Mean ◽

Hidden Layer

In recent years, people are more and more interested in time series modeling and its application in prediction. This paper mainly discusses a financial time series image algorithm based on wavelet analysis and data fusion. In this research, we conducted an in-depth study on the scale decomposition sequence and wavelet transform sequence in different scale domains of wavelet transform according to the scale change rule based on wavelet transform. We use wavelet neural network with different input neurons and hidden neurons to predict, respectively. Finally, the prediction results are integrated into the final prediction results based on the original time series by using wavelet reconstruction technology. Using RBF algorithm in neural network and SPSS Clementine, the wavelet transform sequences on five scales are modeled. Each network model has three layers: one input layer, one hidden layer, and one output layer, and each output layer has only one output element. In order to compare the prediction effect of the model proposed in this study, the ordinary RBF network is used to model and predict the log yield itself. When the input sample is 5, the minimum mean square error is obtained when the hidden layer is 6, and the mean square error is 1.6349. The mean square error of the training phase is 0.0209, and the validation error is 1.6141. The results show that the prediction results of the wavelet prediction method combined with the RBF network prediction method are better than those of wavelet prediction or RBF network prediction.

Download Full-text