Modeling daily soil temperature using data-driven models and spatial distribution

Soil temperature and moisture have a close relationship, the accurate controlling of which is important for crop growth. Mechanistic models built by previous studies need exhaustive parameters and seldom consider time stochasticity and lagging effect. To circumvent these problems, this study designed a data-driven stochastic model analyzing soil moisture-heat coupling. Firstly, three vector autoregression models are built using hourly data on soil moisture and temperature at the depth of 10, 30, and 90 cm. Secondly, from impulse response functions, the time lag and intensity of two variables’ response to one unit of positive shock can be obtained, which describe the time length and strength at which temperature and moisture affect each other, indicating the degree of coupling. Thirdly, Granger causality tests unfold whether one variable’s past value helps predict the other’s future value. Analyzing data obtained from Shangqiu Experiment Station in Central China, we obtained three conclusions. Firstly, moisture’s response time lag is 25, 50, and 120 h, while temperature’s response time lag is 50, 120, and 120 h at 10, 30, and 90 cm. Secondly, temperature’s response intensity is 0.2004, 0.0163, and 0.0035 °C for 1% variation in moisture, and moisture’s response intensity is 0.0638%, 0.0163%, and 0.0050% for 1 °C variation in temperature at 10, 30, and 90 cm. Thirdly, the past value of soil moisture helps predict soil temperature at 10, 30, and 90 cm. Besides, the past value of soil temperature helps predict soil moisture at 10 and 30 cm, but not at 90 cm. We verified this model by using data from a different year and linking it to soil plant atmospheric continuum model.

Download Full-text

A Novel Stacked Long Short-Term Memory Approach of Deep Learning for Streamflow Simulation

Sustainability ◽

10.3390/su132313384 ◽

2021 ◽

Vol 13 (23) ◽

pp. 13384

Author(s):

Majid Mirzaei ◽

Haoxuan Yu ◽

Adnan Dehghani ◽

Hadi Galavi ◽

Vahid Shokri ◽

...

Keyword(s):

Spatial Distribution ◽

Short Term Memory ◽

Data Driven ◽

Rainfall Runoff ◽

Short Term ◽

Runoff Simulation ◽

Physically Based ◽

Long Short Term Memory ◽

Using Data ◽

Modelling Approach

Rainfall-Runoff simulation is the backbone of all hydrological and climate change studies. This study proposes a novel stochastic model for daily rainfall-runoff simulation called Stacked Long Short-Term Memory (SLSTM) relying on machine learning technology. The SLSTM model utilizes only the rainfall-runoff data in its modelling approach and the hydrology system is deemed a blackbox. Conversely, the distributed and physically-based hydrological models, e.g., SWAT (Soil and Water Assessment Tool) preserve the physical aspect of hydrological variables and their inter-relations while taking a wide range of data. The two model types provide specific applications that interest modelers, who can apply them according to their project specification and objectives. However, sparse distribution of point-data may hinder physical models’ performance, which may not be the case in data-driven models. This study proposes a specific SLSTM model and investigates the SLSTM and SWAT models’ data dependency in terms of their spatial distribution. The study was conducted in the two distinct river basins of Samarahan and Trusan, Malaysia, with over 20 years of hydro-climate data. The Trusan basin’s rain gauges are scattered downstream of the basin outlet and Samarahan’s are located around the basin, with one station within each basin’s limits. The SWAT was developed and calibrated following its general modelling approach, however, the SLSTM performance was also tested using data preprocessing with principal component analysis (PCA). Results showed that the SWAT performance for daily streamflow simulation at Samarahan has been superior to that of Trusan. Both the SLSTM and PCA-SLSTM models, however, showed better performance at Trusan with PCA-SLSTM outperforming the SLSTM. This demonstrates that the SWAT model is greatly affected by the spatial distribution of its input data, while data-driven models, irrespective of the spatial distribution of their entry data, can perform well if the data adequacy condition is met. However, considering the structural difference between the two models, each has its specific application in a water resources context. The study of catchments’ response to changes in the hydrology cycle requires a physically-based model like SWAT with proper spatial and temporal distribution of its entry data. However, the study of a specific phenomenon without considering the underlying processes can be done using data-driven models like SLSTM, where improper spatial distribution of data cannot be a restricting factor.

Download Full-text