Reply on RC2

Question

1. What factors affect the runoff NSE in catchment areas?

2. What remote sensing products were used for water balance data?

3. What sensors and algorithms estimate global precipitation?

4. What are the five evapotranspiration products used in the study?

Accepted Answer

The study reveals that highly seasonal rainfall has the highest positive correlation with runoff NSE. Factors such as increasing snow cover, altitude, and latitude decrease the ability of RS products to close the water balance. The dominant climate zone of the catchment also impacts time series performance, with tropical areas showing the highest NSE values and arid areas the lowest. Interestingly, no correlation was found between catchment area and runoff NSE. These findings emphasize the need for further research on data product uncertainties and their interactions. Additionally, the study suggests that improving specific satellite products should be guided by these results, moving beyond simple water balance approaches. This comprehensive analysis provides valuable insights for researchers and practitioners in the field of hydrology and remote sensing.

Accepted Answer

Three precipitation products, five actual evapotranspiration products, and three total water storage change products were used. CHIRPS and TRMM do not cover areas north of 50 degrees N and south of 50 degrees S. The spatial extent of SSEBOP is limited to areas between 80 degrees N and 60 degrees S. SEBS has many missing pixels over large deserts and the Taiga. All products were re-sampled to a monthly time-scale and a spatial resolution of 0.05 degrees.

Accepted Answer

Different sensors and algorithms estimate global precipitation from remote sensing. Precipitation products combine measurements from multiple satellites to achieve higher temporal resolutions. In this study, three products were used: TRMM TMPA, CHIRPS version 2, and GPM IMERG. The datasets were resampled to 0.05deg spatial resolution. The GPM satellite launched in 2014, and the IMERG algorithm extended the time series back to 2000 using TRMM data. Post-2015, the TMPA algorithm was applied to GPM data to continue producing data.

Accepted Answer

The five evapotranspiration products used in the study are: 1. Operational Simplified Surface Energy Balance (SSEBop), 2. CSIRO MODIS Reflectance-based Evapotranspiration (CMRSET), 3. Global Land Evaporation Amsterdam Model (GLEAM), 4. Surface Energy Balance System (SEBS), and 5. MODIS Global Terrestrial Evapotranspiration Algorithm (MOD16). Each product utilizes different methods and data sources to estimate evapotranspiration rates. Detailed information about these products can be found in the respective publications listed for each product. These products were used to obtain monthly data at 0.05-degree spatial resolution by summing daily and dekadal fluxes, resampling, and filling missing data.

Accepted Answer

Total Water Storage Anomaly (TWSA) is obtained from the Gravity Recovery And Climate Experiment (GRACE) satellites, which map the Earth's gravity field approximately every 30 days. The TWSA represents the deviation in total water storage relative to the long-term mean. The GRACE data, available between January 2003 and July 2017, is used to calculate TWSA. The data is processed by three centers: the University of Texas -Center for Space Research (CSR), Geo Forschungs Zentrum (GFZ), and Jet Propulsion Laboratory (JPL). The data is interpolated to the 16th of each month using the central difference method. The final data is resampled to 0.05 degrees using the nearest neighbor method. This approach allows researchers to monitor changes in total water storage over time.

Accepted Answer

A total of 1,149 gauging stations were selected from the GRDC dataset. These stations were filtered to identify those with an upstream catchment larger than 10,000 km 2 and at least one record after January 1st, 2003. The majority of these stations are located in northern America, while the rest are spread across other continents. However, there are very few stations located in Northern Africa, Central Asia, and Southern Asia. Watershed boundaries were obtained from the GRDC, and the monthly mean GRDC data, given in m 3/s, was converted to mm/month for comparison with remote sensing data.

Accepted Answer

After considering data points and station coverage, 937 time-series were analyzed. This was achieved by omitting time-series with fewer than 30 matching data points, excluding stations outside of product coverage areas, and excluding months with more than 20% missing pixels. The total number of time-series was initially 51,705, but due to these factors, the final count was reduced to 937. This process ensures the reliability and accuracy of the runoff time-series analysis derived from remote sensing data.

Accepted Answer

11 RS derived catchment characteristics were selected based on the findings of earlier studies. These characteristics include catchment area, latitude of the outlet of the catchment, snow cover, altitude of the catchment outlet, dam storage capacity, standard deviation of monthly rainfall, ratio between runoff and precipitation, dominant climate class, and dominant land cover class. The Pearson coefficient was computed to assess the correlation between each characteristic and the NSE of the discharge time series. The significance of the correlations was tested using a two-sided student t-test.

Accepted Answer

The average NSE value per product combination varies. For the GPM rainfall product, an average of 925 NSE values per combination could be calculated. For the TRMM and CHIRPS products, an average of 599 NSE values per combination could be calculated due to their smaller spatial coverage. The median NSE values for all GRDC stations available for the 45 possible product combinations are presented in Appendix 1. On average, 43.4% of the generated discharge time series achieved a positive NSE value. The best performing combination was CHIRPS -MOD16 -JPL with 53.8% of positive NSE values, while the worst, GPM -SEBS -GFZ/JPL, yielded 34.7% of positive NSE values. Only an average of 3.2% of the discharge time series reached the threshold of 0.5, with the best combination (CHIRPS -CMRSET -GFZ) reaching this value for 5.9% of stations. When selecting only stations covered by all products, the combination with the highest number of positive NSE values was GPM -CMRSET -JPL with 56% of stations reaching positive median NSE values.

Accepted Answer

Remote sensing products showed varying performance in closing the water balance, with a positive Nash-Sutcliffe Efficiency (NSE) achieved for 72.5% of the stations. The performance was influenced by factors such as precipitation patterns, snow-cover, climate zones, and land cover classes. No single product consistently outperformed others. Improvements are needed in continental and arid climate zones, as well as for certain land cover types. Further analysis could consider additional characteristics like irrigation percentage and calibration areas. Collecting discharge data in underrepresented areas is recommended for future studies.

Reply on RC2

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What factors affect the runoff NSE in catchment areas?

2. What remote sensing products were used for water balance data?

3. What sensors and algorithms estimate global precipitation?

4. What are the five evapotranspiration products used in the study?

5. How is Total Water Storage Anomaly (TWSA) obtained?

6. How many gauging stations were selected from the GRDC dataset?

7. How many time-series were analyzed after considering data points and station coverage?

8. Which catchment characteristics were selected to investigate correlations with quality of RS estimates of discharge?

9. What is the average NSE value per product combination?

10. How do remote sensing products perform in closing water balance?

Related Papers (5)

Effect of precipitation bias correction on water budget calculation in Upper Yellow River, China

Modeling water balance with the ERHYM model on south Texas rangelands

An integrated assessment of runoff dynamics in the Amu Darya River Basin: Confronting climate change and multiple human activities, 1960–2017

Hydrological effects of land use change on small catchments at the Narayen research station, Queensland

Impact of climatic change on eco-hydrology: a case study for Poland