Sensor networks and the future of forecasting

Data and low-cost sensor networks can spot extreme weather before it hits.

We can’t control the climate, but are there ways to mitigate and avoid the negative effects extreme weather brings? I believe the starting point for potential solutions lies in data acquisition and environmental sensor networks (ESN).

Current technologies and sensors, ranging from cell phones to satellites, allow a “global environmental cyberinfrastructure” to be more than a topic for discussion at academic conferences. Researchers have studied connections and system interactions for some time, but now a broader segment of society is becoming aware of the precarious relationship between weather, climate and humanity. This awareness is sometimes motivated by the need to help. Other times there’s a profit incentive. The reason doesn’t matter if the result is a better low-cost global sensor network that can be tapped by anyone with a signal.

A systems approach to identifying natural hazards, coupled with a communications framework that can easily make data available to the public, is the crucial cornerstone of a functional environmental sensor network. The global monitoring of short- and long-range weather patterns and the linking of sensor-network data could allow forecasters to identify potential problems before they manifest.

The weather link between Russia and Pakistan

Weather has caused great disruption to many lives in both Russia and Pakistan in recent months. While these are separate circumstances, they share common physical factors. The following is a look at how events in one part of the world influence weather elsewhere.

The Russian heat wave

Global wheat prices spiked in early August. Much of that activity stemmed from potential crop losses in Russia, and it was helped by ubiquitous stories of parched fields and decimated crops. To be fair, part of the price spike came on the heels of a Russian export ban. Nonetheless, this story’s origin is tied to weather.

The two maps below show the monthly year-over-year (2010 vs. 2009) changes in maximum temperature and precipitation for July. This weather was known to people in the agricultural and weather communities. The stage was set long ago for potential problems in western Russia.


(Click to enlarge)


(Click to enlarge)

A persistent high-pressure system centered over eastern Europe has combined with changes in the jet stream’s pattern to keep temperatures high. When this high is as entrenched as it has been, it serves to do two things:

  1. It diverts a jet stream that would normally steer cooler air into parts of central Russia and northeastern Africa.
  2. It blocks moist air from the southeast, which exacerbates the dryness. This is a reason why parts of central Africa are seeing better moisture in recent months.
  3. There’s an active west-to-east jet stream that travels above western Russia. This stream typically exhibits a seasonal shift to the east, and in the process allows moister air from the west to migrate into the region. The jet did not shift in July, and the result was a prolonged period of moisture-free air. When this combines with a strong high, the region experiences weather like it’s seen over the last month. And when there is a high-pressure system in one region, there is often a corresponding low elsewhere.

    Flooding in Pakistan

    The low in this case has been over the mountainous region of northern Pakistan. This cold low has been the catalyst for a good portion of the excess rains. So, while located in distinct climate zones, the heat in Russia has a connection to the floods in Pakistan.

    But there’s more to this puzzle. Every year, the annual Indian monsoon is anticipated throughout India and Pakistan, as much of the commercial activity that takes place in both countries is agrarian in nature. The monsoon was deficient in 2009, leading to short crops in many sectors. The arrival of the rainy season this year carried a heightened importance.

    The onset of the 2010 monsoon was healthy and most regions have been receiving beneficial moisture totals. But the placement of another area of high pressure over northeast India has, thus far, kept India’s northern states dry. In the process, this high has been diverting even more moisture, which flows from southeast to northwest into central/western India and along into Pakistan.

    The first map below from NOAA’s Earth System Research Laboratory depicts the storm tracks from the last week (as viewed via anomalies in outgoing longwave radiation), where the excess moisture is visible directly over Pakistan. The second map below, from the NOAA Climate Prediction Center, highlights wind anomalies over the last week. We see from this map that the stronger winds originating from the southeast were actively driving the moisture into areas that needed it the least.


    (Click to enlarge)


    (Click to enlarge)

    Pakistan’s drainage infrastructure, which is silt-laden, has made the rainfall situation worse. When excess rains fall, it takes much longer to drain than necessary.

    Unfortunately, any excess rains in the coming weeks will likely be met with more problems for civilians. As of mid-August, estimates put flood-related deaths above 1,300, and more than 15 million people have been affected by the floods. Both of these figures are expected to rise.

    Satellites and risk management

    Could these separate-but-related crises have been foreseen? If so, what measures could have been initiated to mitigate some of the fallout?

    The map below is an indicator of vegetation health as derived via satellite for the wheat regions to the north of the Caspian and Black Seas. This particular graphic depicts the Normalized Difference Vegetation Index (NDVI) for the region, shown as an anomaly vs. the five-year average (for mid-to-late July). The index assigns a value for crop health, and based on the color scale shown in the legend, it is clear that the region has been exhibiting severe vegetation stress.

    The important thing to note here is that wheat prices started to increase in July, then exhibited a violent spike in early August. As this map is from mid-to-late July, we can see that by using tools such as satellite indices in conjunction with a long-range weather forecast, the current impact in wheat prices and ensuing financial turmoil could have been anticipated. To a certain degree, it could have been mitigated through a proactive physically-based risk management strategy. In addition, the same-satellite derived images that were capturing the movement of monsoon rains across India’s agricultural regions could have been used to view the excess moisture in regions where the Pakistan floods originated.

    We can, of course, explore the questions surrounding how much lead time is necessary to avoid a crisis (remember Katrina). Nonetheless, it is clear that many were not aware of these disastrous systems until it was too late.


    (Click to enlarge)

    More coming soon

    This column is a starting point for discussions that examine climate, weather, sensors, networks, and their influence on society. The last couple of years have seen a heightened interest in this area from the research side, evidenced by new topics and sessions presented at the annual conferences of the American Geophysical Union, the American Meteorological Society, and others. I hope to bring some of these discussions to a broader audience while also helping readers understand how closely related their lives and decisions are to weather and climate.

    Related:

tags: , , ,

Get the O’Reilly Data Newsletter

Stay informed. Receive weekly insight from industry insiders.

Comments are closed.