‘Baseflow one of most important factors driving river flooding in Peninsular India’

Krishnarajasagar dam on the Cauvery river, Karnataka. The researchers studied Cauvery basin, among others. Photo: iStock

A recent study looking into heavy rainfall, baseflow and river floods in Peninsular India has provided new insights into what factors drive flooding in the regions, using actual river flow data to understand flood dynamics better.

Delhi-based think tank Centre for Science and Environment’s Mehak Puri talks to study author Shailza Sharma, researcher on hydroclimatic extremes at Indian Institute of Science, Bangalore, to better understand river flooding in the region.

The study, Baseflow significantly contributes to river floods in Peninsular India, was published January 13, 2024 in journal Scientific Reports of Nature portfolio. 

Mehak Puri: Your study explores the correlation between baseflow and river floods in Peninsular India. Which river basins did you examine?

 

Shailza Sharma: Water storage in the upper layer of soil (the unsaturated zone) is referred to as soil moisture, while water storage in the saturated zone is known as groundwater. Below the water table, groundwater can flow into a river, which is generally referred to as baseflow. 

This study examines six major river basins — Narmada, Tapi, Mahanadi, Godavari, Krishna and Cauvery — to identify the dominant mechanisms that generate floods.

Most studies on floods in India have focused on rainfall data because discharge data has only recently been made publicly available through the India-Water Resource Information System website. This is the first study to use observed discharge data from Indian river basins to enhance our understanding of the process-based factors influencing river floods.

MP: Rainfall and soil moisture have always been recognised as fundamental drivers of flood events. However, your study also acknowledges the role of groundwater in this context. Could you explain this correlation?

SS: Extreme events for all four variables are extracted using the annual maxima approach and the correlation between trends in floods and the flood drivers (rainfall, soil moisture, and baseflow) is examined using the Kendall rank correlation. This correlation coefficient is particularly suitable for studying correlations in extreme events. 

A very high value of Kendall’s correlation coefficient was observed between trends in floods and trends in baseflow in Peninsular India. Trends in the annual mean baseflow showed a pattern similar to trends in flood magnitudes, indicating a relatively strong correlation.

In contrast, rainfall and soil moisture trends were less consistent with flood trends, resulting in a weaker correlation. These findings suggest that multi-decadal trends in baseflow have had a significant impact on trends in flood hazards in Peninsular India.

We examined the relationship between annual floods and their drivers for antecedent periods ranging from 1 to 14 days. Baseflow is the dominant factor for shorter time lags, while rainfall plays a more significant role at longer antecedent periods. A catchment with higher baseflow indicates wetter conditions, leading to an increased likelihood of rapid runoff with incoming rainfall events. 

On the other hand, the correlation between rainfall and floods is relatively high at longer time lags. It’s important to remember that rainfall not only drives the flood peak but also contributes to soil moisture and groundwater levels. Accumulated rainfall over a longer period can eventually increase baseflows, resulting in more water contributing to river flows.

We examined the correlation between floods and baseflow on the flooding day and the days leading up to it to further validate our findings. Across Peninsular catchments, a high positive correlation was observed five days before the flood event, reinforcing the strong influence of baseflow on floods.

To understand what happens to this association on the flooding day, we investigated the relative contribution of baseflow to peak flows using the Baseflow Index (BFI). 

A negative correlation was observed between flood magnitudes and BFI, indicating that while baseflow contributes more to river flows, its share in the event flow magnitude decreases as surface runoff accounts for a larger proportion of flood discharge on the day of the flood. Moreover, a flood event cannot occur without substantial rainfall, even if the landscape has high baseflow.

MP: Your study indicates that the construction of reservoirs has impacted the hydrological system by altering the natural flow. This type of construction is occurring on many major rivers in India. How does this affect flood events?

SS:  The construction of reservoirs changes the downstream flow by storing and releasing water according to specific reservoir operation rules designed to meet human needs, such as irrigation, water supply, hydropower, flood control and recreational use.

The catchments in Peninsular India are heavily regulated, so we examined how flow regulations affect flood characteristics like peak, volume, and duration. 

Sudden releases of water from a dam can raise river levels during extreme rainfall events, especially when the dam is nearly full. The impact of sudden floods is more destructive because people living in nearby areas are often unprepared. Without a dam, floodwaters rise slowly over time, allowing people to take preventive measures.

However, sudden floods caused by dam releases offer little or no time to prepare, increasing the risks associated with dam-induced flood hazards.

Furthermore, the construction of dams alters the natural regime of rivers, leading to changes in sediment transport and erosion patterns. These changes can heighten the risk of flooding downstream from a dam.

MP: According to your study, the duration of floods has increased following the construction of dams in Peninsular India. Could you provide some examples to illustrate this?

SS: Indeed, the duration of floods has increased following the construction of dams in Peninsular India. Reservoir regulation has extended flood duration by approximately 65 per cent. Floods in regulated catchments tend to last longer but are generally less severe, with smaller peaks and lower volumes, compared to floods in natural catchments. The longer flood duration is a result of flood control measures, where discharge is distributed over a longer period.

It’s important to note that releasing excess water quickly can cause severe flash floods in downstream areas. To control floods and reduce the risk of flash floods, excess water is released over extended periods. This approach serves as a flood attenuation strategy to protect downstream areas from sudden and severe flooding.

MP: In your opinion, what should be the key solutions for managing floods in India, considering all the triggering factors in your study?

SS: I believe that hydrologic processes need more attention, alongside atmospheric processes, for a better understanding and management of floods.

As I mentioned earlier, most of the focus has been on extreme rainfall (pluvial floods), largely because rainfall data for the country is readily available on a daily time scale from India Meteorological Department (IMD). However, river floods have not been extensively studied, as the data was not freely available.

Efforts are required from both the scientific community and government policymakers for improved flood management in India. Hydrologists should concentrate more on understanding the hydrological processes responsible for transforming rainfall into runoff in catchments.

The government should invest more in monitoring different components of the hydrological cycle, rather than focusing solely on rainfall and temperature. Without a thorough understanding of fundamental estimates of runoff, baseflow, and evaporation in our catchments, it’s impossible to fully grasp the evolution of fluvial floods.

Structural measures for flood control are designed for events of normal coincidence, but there is always some uncertainty about future floods. Therefore, it’s essential to accept that only a certain degree of protection can be achieved with passive measures. Moreover, maintaining a high level of protection through flood control structures requires substantial investment in both maintenance and operation.

Integrating flood forecasting with dam releases could be beneficial in controlling fluvial floods. Generally, these floods occur during the monsoon season, when dams are nearly full and have limited capacity to accommodate inflow during extreme rainfall events.

As a result, dam engineers are often forced to execute sudden releases of water from dams. Effective flood forecasting, with a sufficient lead time, would help dam engineers manage dam water levels and maintain sufficient flood storage capacity.

Subscribe to Daily Newsletter :

SUPPORT US

We are a voice to you; you have been a support to us. Together we build journalism that is independent, credible and fearless. You can further help us by making a donation. This will mean a lot for our ability to bring you news, perspectives and analysis from the ground so that we can make change together.