Spatiotemporal forecasting of water change trends in Urmia Lake through to 2030, using STC-based models

Abstract

The purpose of this paper is to forecast the spatiotemporal water change trends in Urmia Lake through 2030. Three space-time cube-based models were applied. The Forest-based forecast model with a mean of 0.14 forecast RMSE and 0.39 validation RMSE, had a better performance. According to the model’s results, the peripheral parts of the lake will mainly stay completely arid with a 42.5% extension in the entire period, 21.7% in spring, and a 27% reduction in winter. In the middle parts, the aridity will increase by 121% in summer, 93% in fall, and 38% in spring. Eventually, the centric areas will not be completely arid, however, the water patterns extent will decrease by 53.46% in fall, 34.6% in spring, and 28.5% in summer. Additionally, southern, eastern, western, and northern areas will experience worse conditions, respectively. The findings can be used for water resource management and water restoration plans for the lake.

Keywords:

• water body change
• spatiotemporal forecasting
• trend analysis
• space-time cube
• Urmia Lake

Disclaimer

As a service to authors and researchers we are providing this version of an accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proofs will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to these versions also.

Funding

This research did not receive any specific grant from funding agencies.

Disclosure statement

The authors report there are no competing interests to declare.

Data availability

Datasets related to this article can be downloaded at https://prod-dcd-datasets-cache-zipfiles.s3.eu-west-1.amazonaws.com/9m6thb35p9-3.zip, an open-source online data repository hosted at Mendeley Data.

Notes

1. Constricting the Shahid-Kalantari bridge (in use since 2007), divided the lake into two northern and southern lakes (Fig. 2).

2. The water bodies have lower values than the determined threshold.

3. Emerging hot spot analysis uses cubes to construct various patterns in 2D space to evaluate trends. Consequently, it is not feasible to investigate the trends in each bin of the cubes. In this regard, a .sxd file was created, which consists of 3D interactive files that make it possible to investigate the hot and cold spots in each bin at all times and areas of the lake. The file can be accessed using the URL mentioned in the section on data availability.

4. Will never be covered with water

5. Will rarely be covered with water