the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Evaluating the effects of topography and land use change on hydrological signatures: a comparative study of two adjacent watersheds
Abstract. Watershed hydrological processes are significantly influenced by land use/land cover change (LULCC) and watershed characteristics such as topography. This study comparatively investigates the impacts of terrain slope and urbanization-driven LULCC on hydrological processes in two adjacent subtropical watersheds but with distinct terrain and land-cover conditions within the Greater Bay Area (GBA) of China. We developed an Integrated Surface-Subsurface Hydrological Model (ISSHM) using the Simulator for Hydrologic Unstructured Domains (SHUD), which was calibrated using data from river and groundwater flow monitoring stations in the watersheds. The calibrated model facilitated simulations to assess how terrain slope and LULCC affect surface runoff, subsurface flow, evapotranspiration (ET), and infiltration. Our results indicate that slope impacts hydrological processes differently in watersheds with varying characteristics. In mountainous areas, there are consistent high correlations between slope and annual surface runoff, infiltration, and subsurface flow across all watersheds. However, at lower elevations, the hydrological responses of steeper watersheds correlate weakly with local slope. Furthermore, urbanization (increase in impervious areas) has led to significant increases in annual surface runoff and significant decreases in annual infiltration and ET across all watersheds, especially in those with steeper slopes. On the other hand, in watersheds with gentler slopes, the annual increase in surface runoff is less than the percentage increase in impervious area, suggesting a buffering capacity of these flatter watersheds against urbanization. However, this buffering capacity is diminishing with increasing annual rainfall intensity.
- Preprint
(5762 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 23 Oct 2024)
-
RC1: 'Comment on hess-2024-136', Lele Shu, 28 Jul 2024
reply
This paper explores the impacts of land use change and basin topography on hydrological processes using the integrated surface-subsurface hydrological model (ISSHM), the SHUD model. The study conducted model simulations and analyses within the Shenzhen-Hong Kong basin. The experimental design is comprehensive, the arguments are credible, the charts are clear, and the writing is readable. However, I have some concerns that could make this study even more comprehensive and reliable.
-
The authors explored the relationship between slope and several hydrological processes. I did not see the process of slope calculation; therefore, I am uncertain if the slope here refers to the slope calculated from the DEM in different zones or the slope calculated based on the triangular mesh of the SHUD model. Since the resolution of the triangles is lower than that of the DEM, I believe that the slope based on triangles better reflects the impact of slope on hydrological processes. According to lines 187-189, the three zones are divided based on elevation differences, but the average slope attribute within the three zones is not displayed or discussed. Therefore, the article needs to clarify the slope calculation method.
-
Based on the SHUD model, the authors analyzed the impact of slope on hydrological processes (subsurface). Local subsurface flow is usually affected by soil characteristics, especially hydraulic conductivity, and is also influenced by slope runoff (slope + accumulating area). Therefore, the groundwater level is higher in the flat areas around the river channel, leading to a larger subsurface flow.
-
The SHUD model outputs subsurface flow, which may include the flow (Q1, Q2, Q3) in three directions of the triangle (positive outward) and the sum/net flow in the three directions (Qs = Q1 + Q2 + Q3, or eleqsub = eleqsub1 + eleqsub2 + eleqsub3). Then, the water balance of the saturated zone of this unit should be dS = Q1 + Q2 + Q3. In the long-term trend, dS should be close to zero. Here, dS is equivalent to Qs (eleqsub). Therefore, subsurface flow should be the sum of positive flows or the sum of negative flows. If the authors directly use new-flow (Qs/eleqsub) in the calculation, it no longer conforms to the meaning of subsurface flow. For example, the flow rates of the three sides of a triangle are eleqsub1 = 200, eleqsub2 = 100, and eleqsub3 = -310, then eleqsub = 200 + 100 + (-310) = -10. It indicates that the unit gains 10 units of water. In this case, the flow rate at this unit should be 300 (200 + 100) or -310, but definitely not -10. Similarly, the calculation of surface runoff may have the same concerns. I suggest that the authors clarify the reading and processing of variables in the appendix or supplementary materials.
-
Parameter calibration is a very challenging task for any ISSHMs. Although your study uses manual parameter calibration, I strongly recommend that you share more calibration details/experiences so that others can learn from your work. For example, how were sensitive parameters determined? How were parameters adjusted to gradually approach the observed results? How was it determined that the current parameter set reached "optimal" or "usable" levels?
-
Although the authors have clarified the data sources of this study, I strongly recommend that the authors share the research data of this article (including model input files, model source code used to implement this study, and calibrated parameters) in an open database (such as Zenodo). Of course, you can desensitize or not share meteorological and hydrological data that are restricted by copyright or confidentiality clauses. Sharing this data can help other researchers replicate your work and further expand the impact of your work. This is just a suggestion.
Â
Citation: https://doi.org/10.5194/hess-2024-136-RC1 -
AC1: 'Reply on RC1', Haifan Liu, 13 Aug 2024
reply
Dear Prof. Shu, please find attached our detailed responses to your comments. We appreciate the valuable feedback provided, which has been instrumental in refining and improving the manuscript.
-
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
332 | 117 | 20 | 469 | 10 | 15 |
- HTML: 332
- PDF: 117
- XML: 20
- Total: 469
- BibTeX: 10
- EndNote: 15
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1