Recently, in July 2021, destructive and deadly floods occurred in Western Europe. The estimated insured losses only in Germany could approach 5 billion Euros (AIR Worldwide, 2021). However, the total amount of the damage is currently not foreseeable due to the variety and complexity of the damage patterns and the unbelievable extent of the disaster. It seems the socio-economic losses will dramatically increase and break a new record in the insurance industry after evaluating the complete record of damages’ reports (see Figure 1). The death toll in Germany has risen to more than 135, which is irreparable (Bild 2021; Landesregierung Rheinland-Pfalz 2021).

Despite recent developments in flood management, humans cannot successfully control and predict this phenomenon precisely to prevent various subsequent damages. Furthermore, climate change has led to an enhancement in the numbers, frequency, and intensity of floods and heavy rainfalls (Seneviratne et al. 2012). After the recent floods, European Commission President Ursula von der Leyen, also mentioned: "The intensity and length of the extreme weather is a clear indication of the crisis" (Bloomberg Quicktake, 2021). This demands an emergency plan to employ new and innovative technologies.

Flooding in Nordrhein-Westfalen, Germany, July 2021. Photo: City Erftstadt, (Davies 2021)
Figure 1: Flooding in Nordrhein-Westfalen, Germany, July 2021. Photo: City Erftstadt, (Davies 2021)


Satellite-based monitoring of extreme events

One of the main problems related to floods and weather prediction is insufficient measurement gauges for hydrometeorological parameters like rainfall and discharge flow. The operation and maintenance of these observation stations are costly. Furthermore, it is impossible to cover all regions in a country due to different climates like deserts or locations with high elevations with no safe access roads. In addition, usually, cities have a priority to be protected against extreme events, and therefore, the density of observation networks is higher near the populated areas. Space-based sensors can be considered as a practical alternative to provide data and information. Nevertheless, radars and optical Remote Sensing (RS) products have their own limitation, especially during extreme events (Oberstadler, Hönsch, and Huth 1997; Rajabi, Nahavandchi, and Hoseini 2020).

The Global Navigation Satellite System (GNSS) is mainly designed for positioning, navigation, and timing. Having been reflected from the Earth's surface, GNSS signals can also provide information about the reflecting surface. Global Navigation Satellite System Reflectometry (GNSS-R) is an innovative, and cost-effective technique with a high spatiotemporal resolution aimed at deriving geophysical parameters by analyzing GNSS reflected signals from different Earth system components. This idea has been used in distinct research topics, such as flood management, soil moisture, and water cycle (Chew, Reager, and Small 2018; Wan et al. 2019; Rajabi, Nahavandchi, and Hoseini 2020). Numerous GNSS satellites, including Global Positioning System (GPS), Galileo, GLONASS, and BeiDou, are currently transmitting navigation signals based on spread-spectrum technology. A constellation of GNSS-R small satellites, like the recently launched Cyclone Global Navigation Satellite System (CYGNSS), has lower production and maintenance costs than earth observation satellites. Furthermore, the sensors in CYGNSS overcome limitations of optical sensors such as cloud cover and the lack of night vision. Therefore, they represent an efficient RS tool, which can provide much shorter revisit times (ESA 2020; 2021a). In contrast to most EO satellites, CYGNSS is orbiting in in the LEO and therefore has a shorter revisit time. The constellation consists of eight microsatellites that perfectly match each other to cover -35 to 35 degrees of the Earth (ESA 2020). CYGNSS has in average 4 hours revisit time. Each satellite has a mass of only 25kg, which makes it a microsatellite constellation (Ruf et al. 2016). Even though other satellite missions like Flock 1 are nanosatellites with shorter revisiting time (ESA 2021b), CYGNSS has a high temporal resolution using reflections of GNSS signals to collect meteorological and cyclonical parameters, a novel idea in space science. Two parameters, Integrated Water Vapor (IWV) and signal-to-noise ratio (SNR) can be obtained from GPS and CYGNSS and will be addressed below:

Integrated Water Vapour

IWV shows the maximum potential of liquid precipitation; for example, Figure 2 visualizes the worldwide column IWV for June to August from 1971-2001.

Column integrated water vapor (kg/m2) for June-August from 1971-2001 (ECMWF 2021a)
Figure 2: Column integrated water vapor (kg/m2) for June-August from 1971-2001 (ECMWF 2021a)


However, rain depends on more factors than water vapour. Hence, actual precipitation is not equal to the IWV. The total precipitation for the same time period as in Figure 2 is illustrated in Figure 3.

Total precipitation (mm/day) for June-August from 1971-2001 (ECMWF 2021b)
Figure 3: Total precipitation (mm/day) for June-August from 1971-2001 (ECMWF 2021b).


If a high IWV value correlates with high soil moisture, precipitation will directly contribute to runoff, because the soil does not have any capacity to keep the water. Tracking a fine temporal resolution of IWV (hourly and sub-hourly) and a spatial resolution of one square kilometre (WMO 2021) helps planners and authorities to warn vulnerable areas nearly real-time (NRT). However, results cannot be retrieved timely enough to make vital decisions such as evacuating cities. It is further possible to recognize a pattern of changing IWV in time and space for different seasons of the year to assist a short-term IWV prediction (Figure 4). So far, in synoptic meteorology the structure and behaviour of the real atmosphere have been investigated with different tools. The application of GNSS measurements can help provide better prediction results in combining current meteorological methods.

Integrated wateFigure 4: Integrated water vapor in 3D animation from 29.11.2012 to 07.12.2012 (NOAA 2021)r vapor in 3D animation from 29.11.2012 to 07.12.2012 (NOAA 2021)
Figure 4: Screenshot of a 3D animation of Integrated water vapor from 29.11.2012 to 07.12.2012 (NOAA 2021)


Soil moisture

The second parameter that can be obtained from CYGNSS satellites via GNSS reflectometry, is the SNR, a factor that is highly correlated with soil moisture (Morris et al. 2019). The signal reflected from the Earth's surface depends on a few parameters, most of them remaining constant. The only parameter that does show a variation is soil moisture. Because CYGNSS is a satellite constellation and the individual satellites have different radiation angles, the SNR needs calibration to be homogenized across all satellite data sources. CYGNSS works with the reflection of the four main GNSSs, including GPS, Galileo, GLONASS and BeiDou. That means, if we want to investigate any point within the coverage area, we will have at least one of these systems available.

A) Representation of the CYGNSS measurements along the satellite tracks, B) the interpolated data at 0.1* 0.1 grid points (Rajabi, Nahavandchi, and Hoseini 2020)
Figure 5: A) Representation of the CYGNSS measurements along the satellite tracks, B) the interpolated data at 0.1* 0.1 grid points (Rajabi, Nahavandchi, and Hoseini 2020)


Combining IWV and SNR parameters can help detect floods and to map them. Subsequently, it supports the creation of flood defence plans. Figure 5A illustrates satellite tracks over the study area. In Figure 5B, the red dots represent the areas with high soil moisture concentration, signifying a flood. One can hence conclude in which areas, the soil is completely saturated. Any further rainfall directly becomes runoff and triggers or intensifies the flood.

Flood mapping

It is possible to calculate the surface areas of inundated regions and estimate human lives, ecosystem and financial losses. This also allows recalculating the risk by changing the vulnerable areas (Modiri and Modiri 2016), which have been altered over the past decades as an effect of climate change. Moreover, these maps allow determining sub-catchments that spatially react simultaneously, which is a recently much-debated topic among scholars analysing floods. There are two forms currently classified as simultaneous reactions (Modiri and Bárdossy 2021):

  1. The sub-catchments react at the same time (simultaneously) in different not flow-connected areas. If floods occurred at the same time in the past, one could use the co-reaction to alarm related high-risk regions in the future. For example, in areas that are hydrologically far from each other (no river connection – two tributaries of a river). In the latest flood in Europe in 2021, flooding is recorded in Germany and Belgium in two different countries on a large scale. To improve flood defence, it is possible to determine some regions that usually reacted similarly.
  2. Floods happen subsequently with a time delay between usually less than one day. It is the form of joining floods at upstream sub-catchments to the main river and measuring extreme floods due to long term precipitation or a combination of snowmelt and intensive rainfall in downstream.

Areas that react simultaneously demand to have similar action plans to manage floods. If a severe flood happens in one of these sub-catchments, the chance of flood in others will increase, and the risk of vulnerability will enhance. Therefore, monitoring corrected SNR may help provide maps with the capability to update resulting information production shortly before, during, and after floods. Alongside, long-time series of IWV can support flood monitoring.

A) Flooding areas highlighted in blue colour in Germany in 2021 (Copernicus EMS 2021), B) Simultaneous flood areas in the Neckar catchment in Germany (Modiri and Bárdossy 2021). Dark blue: A first cluster in the west part of the upper Neckar catchment with high elevation; Light green: The small sub-catchments in the east of upper Neckar with different geological features; Yellow: The low land areas.
Figure 6: A) Flooding areas highlighted in blue colour in Germany in 2021 (Copernicus EMS 2021) 6B) Simultaneous flood areas in the Neckar catchment in Germany (Modiri and Bárdossy 2021). Dark blue: A first cluster in the west part of the upper Neckar catchment with high elevation; Light green: The small sub-catchments in the east of upper Neckar with different geological features; Yellow: The low land areas.


Figure 6A shows three distinct regions in Germany, which faced severe floods in July 2021. According to the latest scientific findings, it is advisable to revise flood management regulations to prepare for possible extreme weather events in the coming years. The applied clustering approach in Figure 6B illustrates regions where floods have similar patterns. Here, flood behaviour mainly follows the topography of the studied area. In contrast, the main factors used in larger scale regions include both, the topography plus meteorological circulation patterns. Moreover, recent research shows that floods' seasonality has a relationship with simultaneous floods and corresponding clusters.

Another aspect worth considering in the monitoring of floods is the possible relationship between floods and observed heatwaves. An increase in the Earth's surface’s temperature causes more water vapor in the atmosphere. As a consequence, intense and highly frequent precipitation can be expected. The possible interaction is recommended to be studied in further depth.

To conclude, it needs to be highlighted that accurate flood monitoring and management demand emergency action for scientists and responsible organizations. Determining a high risk of flooding for different catchments and forecasting the magnitude of floods is a scientific task for scholars, which can inform the work of national and international institutions that implement research results in emergency management systems.

Two GNSS-R parameters can play a vital role in flood detection, supporting management schemes and in overcoming restrictions caused by clouds, ungauged basins, missing values, and lack of light. Missing values can occur for several reasons such as: The measuring point on the river or bridge was submerged in water, devices did not work properly, or the person responsible for measuring in non-automatic stations could not measure. Therefore, GNSS-R plays an important role in filling the gaps in recording data. A short revisiting time of CYGNSS tracks provides fine temporal data. However, the footprint covers only tropical and subtropical areas from 35° N to 35° S. This problem could be addressed by future satellite missions. Furthermore, the spatial resolution of CYGNSS, which is around 25*25 km, could be refined. It is sufficient for some hydrological evaluations in large scale extreme events (Huang, Bárdossy, and Zhang 2019). However, it is better to have a finer spatial resolution. The simulated streamflow improves as the spatial resolution is increased (Etchevers et al. 2001). In general, rainfall is a local phenomenon and may happen in small scall areas. For example, if we want to investigate a flood in a village or a small sub-catchment, it is demanded to have the highest resolution to run a hydrological model or to calculate possible losses in these areas.


AIRWorldwide. 2021. “Verisk’s AIR Worldwide Estimates Insured Losses for July Floods in Germany Could Approach EUR 5 Billion.” 2021.….

Bild. 2021. “Storm News in the Live Ticker.” Bild, 2021.….

Chew, Clara, John T Reager, and Eric Small. 2018. “CYGNSS Data Map Flood Inundation during the 2017 Atlantic Hurricane Season.” Scientific Reports 8 (1): 9336.

Copernicus EMS. 2021. “The Copernicus Emergency Management Service Forecasts, Notifies, and Monitors Devastating Floods in Germany, Netherlands, Belgium and Switzerland | COPERNICUS EMERGENCY MANAGEMENT SERVICE.” Copernicus . July 16, 2021.….

Davies, Richard. 2021. “Germany – More Than 100 Fatalities Reported After Floods in Western States.” FloodList. July 16, 2021.

ECMWF, ERA-40 Atlas. 2021a. “Column Integrated Water Vapour.” European Centre for Medium-Range Weather Forecasts (ECMWF). 2021.

———. 2021b. “Total Precipitation.” European Centre for Medium-Range Weather Forecasts (ECMWF): 2021.

ESA. 2020. “Newcomers Earth Observation Guide | ESA Business Applications.” ESA. 2020.

———. 2021a. “CYGNSS | Satellite Missions | Directory.” ESA, Eo Portal Directory. 2021.

———. 2021b. “Flock 1 Constellation - Satellite Missions - EoPortal Directory.” EO Portal. 2021.

Etchevers, Pierre, Yves Durand, Florence Habets, Eric Martin, and Joël Noilhan. 2001. “Impact of Spatial Resolution on the Hydrological Simulation of the Durance High-Alpine Catchment, France.” Annals of Glaciology 32 (September): 87–92.

Huang, Yingchun, András Bárdossy, and Ke Zhang. 2019. “Sensitivity of Hydrological Models to Temporal and Spatial Resolutions of Rainfall Data.” Hydrology and Earth System Sciences 23 (6): 2647–63.

Landesregierung Rheinland-Pfalz. 2021. “Zahlen Und Fakten Hochwasser Ahr.” 2021.

Modiri, Ehsan, and András Bárdossy. 2021. “Clustering Simultaneous Occurrences of the Extreme Floods in the Neckar Catchment.” Water 13 (4).

Modiri, Ehsan, and Sadegh Modiri. 2016. “Zonation of Vulnerability of Precipitation Changes in the Great Khorasan.” Arabian Journal of Geosciences 9 (5): 337.

Morris, Mary, Clara Chew, John T Reager, Rashmi Shah, and Cinzia Zuffada. 2019. “A Novel Approach to Monitoring Wetland Dynamics Using CYGNSS: Everglades Case Study.” Remote Sensing of Environment 233: 111417.

NOAA. 2021. “NOAA PSL Climate Data Repository - 3D Integrated Water Vapor.” NOAA PSL Climate Data Repository. 2021.….

Oberstadler, R, H Hönsch, and D Huth. 1997. “Assessment of the Mapping Capabilities of ERS-1 SAR Data for Flood Mapping: A Case Study in Germany.” Hydrological Processes 11 (10): 1415–25.

Rajabi, Mahmoud, Hossein Nahavandchi, and Mostafa Hoseini. 2020. “Evaluation of CYGNSS Observations for Flood Detection and Mapping during Sistan and Baluchestan Torrential Rain in 2020.” Water 12 (7).

Ruf, C, M P Clarizia, S Gleason, Z Jelenak, J Murray, M Morris, S Musko, et al. 2016. CYGNSS Handbook Cyclone Global Navigation Satellite System. Ann Arbor, MI, USA: Michigan Publishing. National Aeronautics and Space Administration.….

Seneviratne, Sonia I., Neville Nicholls, David Easterling, Clare M. Goodess, Shinjiro Kanae, James Kossin, Yali Luo, et al. 2012. “Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation.” In Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the Intergovernmental Panel on Climate Change, 9781107025:109–230.

Wan, Wei, Baojian Liu, Ziyue Zeng, Xi Chen, Guiping Wu, Liwen Xu, Xiuwan Chen, and Yang Hong. 2019. “Using CYGNSS Data to Monitor China’s Flood Inundation during Typhoon and Extreme Precipitation Events in 2017.” Remote Sensing .

WMO. 2021. “Details for Variable: Integrated Water Vapour (IWV).” WMO OSCAR. 2021.