the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Oct 2024
Brief communication: Evidence of the impacts of climate extremes on power system outages in India
Abstract. Electricity systems are prone to climate extremes such as heatwaves and cold spells, precipitation, high winds, and flood inundation. Yet, the impacts of these climate extremes on the provision of electricity is scarce, in particular in the Global South context. Here, we combine four years of daily electricity outages data from 370 locations across India with temperature, wind, precipitation and flood inundation data to provide empirical evidence of their impacts of the electricity system. We find that outages minutes can increase 20–70 % during days with high wind speed (>50 m/s), 80–220 % during days of intense precipitation (>40 mm/day), and around 15–60 % during heatwaves (>40 degrees Celsius). In terms of flooding, we find that severe flood inundation in urban areas can increase daily outage minutes with a factor 2.1 to 5.5. Our findings highlight how high frequency data can help empirically validate how climate extremes can affect essential services to customers, and how these impacts differ across types of locations. This information is key for those countries that aim to meet universal access to energy in the coming decades, yet at the same time will experience more frequent and intense climate extremes.
- Preprint
(1268 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2024-3176', Akhtar Alam, 05 Dec 2024
This article examines the impacts of weather extremes on power outages in India. The analysis uses data of 227,157 observation days for the period from 2015 to 2018, representing 370 locations across the country. The extremes that have been considered include maximum temperature, peak wind gust, total precipitation and flood inundation.
The correlation established in this study could serve as a crucial input for policies aimed at strengthening the resilience of power infrastructure.
However, I believe including northern stations would have provided valuable insights into an important aspect of extreme weather i.e., the form of precipitation (e.g., snowfall), which causes huge disruption to power supply and damage to infrastructure during the winter months in some northern parts of the country.
Moreover, placing greater emphasis on seasonal patterns of extreme weather and their impacts on system components such as power stations, transmission lines, and transformers could provide a more detailed understanding of the relationship between extreme weather and power outages in India.
Overall, I consider the article as a noteworthy contribution.
Citation: https://doi.org/10.5194/egusphere-2024-3176-CC1 -
RC1: 'Comment on egusphere-2024-3176', Anonymous Referee #1, 13 Dec 2024
This article examines the impacts of weather extremes on power outages in India. The analysis uses data of 227,157 observation days for the period from 2015 to 2018, representing 370 locations across the country. The extremes that have been considered include maximum temperature, peak wind gust, total precipitation and flood inundation.
The correlation established in this study could serve as a crucial input for policies aimed at strengthening the resilience of power infrastructure.
However, I believe including northern stations would have provided valuable insights into an important aspect of extreme weather i.e., the form of precipitation (e.g., snowfall), which causes huge disruption to power supply and damage to infrastructure during the winter months in some northern parts of the country.
Moreover, placing greater emphasis on seasonal patterns of extreme weather and their impacts on system components such as power stations, transmission lines, and transformers could provide a more detailed understanding of the relationship between extreme weather and power outages in India.
Overall, I consider the article as a noteworthy contribution.
Citation: https://doi.org/10.5194/egusphere-2024-3176-RC1 -
RC2: 'Comment on egusphere-2024-3176', Anonymous Referee #2, 21 Feb 2025
This manuscript examines the impact of climate extremes (heatwaves, high winds, heavy precipitation, and flooding) on electricity outages across India. Using four years (2015–2018) of high-frequency outage data from 370 locations, combined with climate data, the study employs fixed-effects panel regressions to quantify outage increases during extreme weather events. Findings indicate that outages rise significantly—by up to 220% during heavy rainfall and up to 5.5 times during severe floods. The study highlights vulnerabilities in power infrastructure, particularly in urban areas, and calls for improved data collection and climate-resilient energy investments in the Global South.
Although such a study is required for climate resilience planning, the current version has significant limitations. Fix the grammatical and sentence errors throughout the work. Write the whole manuscript more concisely. Below are some more specific comments for further improvement.
- In the abstract, this line ‘….80-220% during days of intense precipitation….’ is written (similar to the 3.1 section ‘…..increase to a factor 80-200% for precipitation of 40-60 mm per day…’), how did the author calculate and find 220%?
- There is redundancy in the introduction (paragraph 3) and methods (2.1 section) section. What are the current research gap(s) and novel contribution(s) of this work?
- In the abstract and introduction, the author mentioned about 370 power point stations. However, in section 2.1 the author mentioned both 370 and 500 stations. Observation days are mentioned in the introduction 227000; but in the 2.1 section 227157.
- The paper highlights the scarcity of empirical evidence, but does it truly provide novel insights for the global south, or does it mainly confirm what is already expected based on theory and prior studies?
- How likely is the accuracy of the ERA5 data compared to the station data?
- Given the interannual variability of extreme weather, a longer dataset (e.g., spanning multiple decades rather than only 4 years) yields more robust results.
- Write down the sources of data in a single table for better understanding.
- What type of cascading failures are you talking about?
- Why did you select fixed-effects (FE) panel regressions for this study over other statistical methods? Is it more advantageous than other methods?
- Is the use of negative binomial regression appropriate given the upper bound of daily outage minutes (1440 minutes)?
- The manuscript notes that geocoding errors could introduce spatial bias. Could this lead to misattribution of climate impacts, particularly for flood-related outages?
- The author should perform a statistical analysis (e.g., goodness of fit, sensitivity analysis) to evaluate the performance of their models.
- While the study finds strong statistical associations, does it adequately consider non-climatic factors (e.g., grid mismanagement, underinvestment, theft) that may be equally or more important in explaining electricity unreliability?
- Why do rural power systems found more resilient against high or low temperatures and wind than urban power systems?
- The work lacks convincing justification (discussion) for the results.
- The conclusion section is missing.
- Overall, this work relies solely on fixed-effects panel regressions without addressing potential endogeneity or alternative causal inference methods. The geocoding inaccuracies and limited four-year datasets could introduce spatial and temporal biases, reducing result reliability. Additionally, the study lacks a clear distinction between direct and indirect outage causes, potentially overstating climate impacts. The negative binomial regression's limitations (upper bound issue) and potential omitted variable bias further weaken causal claims. Finally, the findings are largely confirmatory rather than novel.
Citation: https://doi.org/10.5194/egusphere-2024-3176-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 231 | 148 | 35 | 414 | 17 | 19 |
- HTML: 231
- PDF: 148
- XML: 35
- Total: 414
- BibTeX: 17
- EndNote: 19
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1