MEMO

What do you see in the picture?

This is a visualization to demonstrate the concept of monitoring rainfall using microwave links in telecommunication networks.

The coloured areas on the map show ongoing or past rainfall in the selected location. The colour represents estimated rainfall intensity at each time step.

This is a prototype and as such may be missing data or show erroneous information at times. For example, coloured pixels may appear even if it did not rain due to non-rain signal distortions. Coloured areas that do not move is an indication of such noise. In case no colour is visible there is either no rainfall recorded at the moment, or the service is unavailable. Then the Example event can be useful to gain insight about the technique during an earlier event.

Using microwave links to monitor rainfall

Microwave links constitute an important part of the backhaul network in telecommunication networks. The commercial microwave links (CML) essentially transmit a microwave signal between two telecommunication tower locations through the lower atmosphere. Rainfall attenuates the signal, and the attenuation is proportional to rainfall intensity. This provides an opportunity to monitor rainfall – and potentially also other variables – using such networks.

Monitoring rainfall with microwave links can have several advantages. The large number of links provide opportunity for monitoring at many more locations, particularly in cities and in areas with limited radar coverage. The operational nature of the networks enables monitoring at very high temporal resolution and in near real time. From this, more detailed rainfall maps can be created to depict ongoing rainfall, as visualized above.

MEMO Microwave-based Environmental Monitoring at SMHI

The potential to monitor rain and other environmental variables prompted SMHI to initiate a joint pilot project in Gothenburg in 2015 together with Ericsson and Hi3G Sweden. This was the starting point for the development of SMHI’s Microwave-based Environmental Monitoring algorithm (MEMO ®), on which the visualization above is based.

The Hi3G microwave link network in and around Gothenburg, Sweden has been monitored ever since and used to derive rainfall intensity at 1-minute resolution based on raw signal strength data sampled at 10-second resolution from more than 350 bi-directional links. Based on spatial interpolation of the data from each link, rainfall intensity maps are constructed for central Gothenburg at 500×500m resolution. Subsequent initiatives have expanded the work to other areas, e.g. Stockholm and Rwanda.

What are the results?

The results of the Gothenburg pilot project indicate that the microwave links provide several advantages compared to conventional rainfall monitoring systems (e.g. weather radar and gauges), including:

• more complete spatial coverage (350 links vs. 11 operational gauges in Gothenburg),
• higher temporal resolution (1-min for links vs. 15-min for SMHI’s operational gauges and radar),
• higher spatial resolution (primarily in urban areas, 0.25km2 for links vs. 4km2 for SMHI’s operational radar),
• near-ground sampling (weather radars have a risk of missing precipitation from low clouds or measure precipitation that evaporates before reaching the ground),
• high-frequency sampling (every 10s, avoiding that a quick event is missed which can happen for e.g. 15-min sampling by radar),
• and more accurate representation of temporal dynamics and peak rainfall intensities.