Linking Hydro-Physical Variables and Landscape Metrics using Advanced Data Mining for Stream-Flow Prediction
(2022) In Water Resources Management 36(11). p.4255-4273- Abstract
In Streamflow prediction the most important triggering/controlling variables are related to climate, physiography, and landscape patterns. This study investigated the effect of different landscape metrics to relate spatial patterns to surface runoff processes and predict monthly streamflow using climatic and physiographic variables for the 42 sub-basins of the Urmia Lake Basin in Iran. We developed an innovative data-driven framework and considered two different modelling approaches i.e., modelling in homogenous clusters (local approach) and modelling in the entire area as an entity (global approach). The results of basin LULC monitoring from the 20-year experimental period display drastic changes in the land use of the basin such as... (More)
In Streamflow prediction the most important triggering/controlling variables are related to climate, physiography, and landscape patterns. This study investigated the effect of different landscape metrics to relate spatial patterns to surface runoff processes and predict monthly streamflow using climatic and physiographic variables for the 42 sub-basins of the Urmia Lake Basin in Iran. We developed an innovative data-driven framework and considered two different modelling approaches i.e., modelling in homogenous clusters (local approach) and modelling in the entire area as an entity (global approach). The results of basin LULC monitoring from the 20-year experimental period display drastic changes in the land use of the basin such as reduction in lake area (48.3%) due to increasing irrigated areas (22.5%), increasing residential areas (14.2%), and decrease in rangeland (6.0%). Streamflow prediction results in the global experiment showed Group Method of Data Handling (GMDH) and Random Forest (RF) with NSE of 0.76 and NRMSE of 6.44% have similar results and outperformed Partial Least Squares regression (PLS), but in clustering experiment GMDH with NSE of 0.88 and NRMSE of 5% shows the highest accuracy and outperformed both RF and PLS. The results confirmed that modelling in homogenous clusters (local prediction) significantly enhanced the performance of prediction.
(Less)
- author
- Moosavi, Vahid ; Karami, Ayoob ; Behnia, Negin ; Berndtsson, Ronny LU and Massari, Christian
- organization
- publishing date
- 2022-09
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Ecological landscape metrics, Machine learning, Runoff prediction, Urmia Lake
- in
- Water Resources Management
- volume
- 36
- issue
- 11
- pages
- 19 pages
- publisher
- Springer
- external identifiers
-
- scopus:85134365387
- ISSN
- 0920-4741
- DOI
- 10.1007/s11269-022-03251-9
- language
- English
- LU publication?
- yes
- id
- 0c9a95e5-1e34-4498-bca8-76c8ddd98fbc
- date added to LUP
- 2022-09-13 14:21:54
- date last changed
- 2023-10-09 09:20:24
@article{0c9a95e5-1e34-4498-bca8-76c8ddd98fbc, abstract = {{<p>In Streamflow prediction the most important triggering/controlling variables are related to climate, physiography, and landscape patterns. This study investigated the effect of different landscape metrics to relate spatial patterns to surface runoff processes and predict monthly streamflow using climatic and physiographic variables for the 42 sub-basins of the Urmia Lake Basin in Iran. We developed an innovative data-driven framework and considered two different modelling approaches i.e., modelling in homogenous clusters (local approach) and modelling in the entire area as an entity (global approach). The results of basin LULC monitoring from the 20-year experimental period display drastic changes in the land use of the basin such as reduction in lake area (48.3%) due to increasing irrigated areas (22.5%), increasing residential areas (14.2%), and decrease in rangeland (6.0%). Streamflow prediction results in the global experiment showed Group Method of Data Handling (GMDH) and Random Forest (RF) with NSE of 0.76 and NRMSE of 6.44% have similar results and outperformed Partial Least Squares regression (PLS), but in clustering experiment GMDH with NSE of 0.88 and NRMSE of 5% shows the highest accuracy and outperformed both RF and PLS. The results confirmed that modelling in homogenous clusters (local prediction) significantly enhanced the performance of prediction.</p>}}, author = {{Moosavi, Vahid and Karami, Ayoob and Behnia, Negin and Berndtsson, Ronny and Massari, Christian}}, issn = {{0920-4741}}, keywords = {{Ecological landscape metrics; Machine learning; Runoff prediction; Urmia Lake}}, language = {{eng}}, number = {{11}}, pages = {{4255--4273}}, publisher = {{Springer}}, series = {{Water Resources Management}}, title = {{Linking Hydro-Physical Variables and Landscape Metrics using Advanced Data Mining for Stream-Flow Prediction}}, url = {{http://dx.doi.org/10.1007/s11269-022-03251-9}}, doi = {{10.1007/s11269-022-03251-9}}, volume = {{36}}, year = {{2022}}, }