Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Areal Precipitation Coverage Ratio for Enhanced AI Modelling of Monthly Runoff: A New Satellite Data-Driven Scheme for Semi-Arid Mountainous Climate

Hosseini, Seyyed Hasan LU ; Hashemi, Hossein LU orcid ; Fakheri Fard, Ahmad and Berndtsson, Ronny LU orcid (2022) In Remote Sensing 14(2).
Abstract
Satellite remote sensing provides useful gridded data for the conceptual modelling of hydrological processes such as precipitation–runoff relationship. Structurally flexible and computationally advanced AI-assisted data-driven (DD) models foster these applications. However, without linking concepts between variables from many grids, the DD models can be too large to be calibrated efficiently. Therefore, effectively formulized, collective input variables and robust verification of the calibrated models are desired to leverage satellite data for the strategic DD modelling of catchment runoff. This study formulates new satellite-based input variables, namely, catchment- and event-specific areal precipitation coverage ratios (CCOVs and ECOVs,... (More)
Satellite remote sensing provides useful gridded data for the conceptual modelling of hydrological processes such as precipitation–runoff relationship. Structurally flexible and computationally advanced AI-assisted data-driven (DD) models foster these applications. However, without linking concepts between variables from many grids, the DD models can be too large to be calibrated efficiently. Therefore, effectively formulized, collective input variables and robust verification of the calibrated models are desired to leverage satellite data for the strategic DD modelling of catchment runoff. This study formulates new satellite-based input variables, namely, catchment- and event-specific areal precipitation coverage ratios (CCOVs and ECOVs, respectively) from the Global Precipitation Mission (GPM) and evaluates their usefulness for monthly runoff modelling from five mountainous Karkheh sub-catchments of 5,000–43,000 km2 size in west Iran. Accordingly, 12 different input combinations from GPM and MODIS products were introduced to a generalized deep learning scheme using artificial neural networks (ANNs). Using an adjusted five-fold cross-validation process, 420 different ANN configurations per fold choice and 10 different random initial parameterizations per configuration were tested. Runoff estimates from five hybrid models, each an average of six top-ranked ANNs based on six statistical criteria in calibration, indicated obvious improvements for all sub-catchments using the new variables. Particularly, ECOVs were most efficient for the most challenging sub-catchment, Kashkan, having the highest spacetime precipitation variability. However, better performance criteria were found for sub-catchments with lower precipitation variability. The modelling performance for Kashkan indicated a higher dependency on data partitioning, suggesting that long-term data representativity is important for modelling reliability. (Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Artificial intelligence, Data scarcity, GPM-IMERG, k-fold data partitioning, MODIS Terra, NDVI, Rainfall, Semiarid, Soil moisture, Streamflow
in
Remote Sensing
volume
14
issue
2
article number
270
pages
26 pages
publisher
MDPI AG
external identifiers
  • scopus:85122319266
ISSN
2072-4292
DOI
10.3390/rs14020270
language
English
LU publication?
yes
id
a68b3b81-e147-4b09-9231-142f7bc36733
date added to LUP
2022-01-10 14:18:51
date last changed
2023-10-04 01:57:03
@article{a68b3b81-e147-4b09-9231-142f7bc36733,
  abstract     = {{Satellite remote sensing provides useful gridded data for the conceptual modelling of hydrological processes such as precipitation–runoff relationship. Structurally flexible and computationally advanced AI-assisted data-driven (DD) models foster these applications. However, without linking concepts between variables from many grids, the DD models can be too large to be calibrated efficiently. Therefore, effectively formulized, collective input variables and robust verification of the calibrated models are desired to leverage satellite data for the strategic DD modelling of catchment runoff. This study formulates new satellite-based input variables, namely, catchment- and event-specific areal precipitation coverage ratios (CCOVs and ECOVs, respectively) from the Global Precipitation Mission (GPM) and evaluates their usefulness for monthly runoff modelling from five mountainous Karkheh sub-catchments of 5,000–43,000 km2 size in west Iran. Accordingly, 12 different input combinations from GPM and MODIS products were introduced to a generalized deep learning scheme using artificial neural networks (ANNs). Using an adjusted five-fold cross-validation process, 420 different ANN configurations per fold choice and 10 different random initial parameterizations per configuration were tested. Runoff estimates from five hybrid models, each an average of six top-ranked ANNs based on six statistical criteria in calibration, indicated obvious improvements for all sub-catchments using the new variables. Particularly, ECOVs were most efficient for the most challenging sub-catchment, Kashkan, having the highest spacetime precipitation variability. However, better performance criteria were found for sub-catchments with lower precipitation variability. The modelling performance for Kashkan indicated a higher dependency on data partitioning, suggesting that long-term data representativity is important for modelling reliability.}},
  author       = {{Hosseini, Seyyed Hasan and Hashemi, Hossein and Fakheri Fard, Ahmad and Berndtsson, Ronny}},
  issn         = {{2072-4292}},
  keywords     = {{Artificial intelligence; Data scarcity; GPM-IMERG; k-fold data partitioning; MODIS Terra; NDVI; Rainfall; Semiarid; Soil moisture; Streamflow}},
  language     = {{eng}},
  month        = {{01}},
  number       = {{2}},
  publisher    = {{MDPI AG}},
  series       = {{Remote Sensing}},
  title        = {{Areal Precipitation Coverage Ratio for Enhanced AI Modelling of Monthly Runoff: A New Satellite Data-Driven Scheme for Semi-Arid Mountainous Climate}},
  url          = {{http://dx.doi.org/10.3390/rs14020270}},
  doi          = {{10.3390/rs14020270}},
  volume       = {{14}},
  year         = {{2022}},
}