IGP

https://youtu.be/uR93K-pSopo

https://github.com/CosiBerger/MapleHeat/blob/main/MapleHeatScript.js

Extreme heat events are increasing worldwide. Man-made climate change is a major contributor, posing a threat to humans and all other living things on Earth. Remote sensing data can help to detect heat events and especially forest fires at an early stage. By looking at the earth from a distance, it is easier to identify correlations and work out solutions to the problem. This will not prevent heat events, but it will help to identify them and warn the population in advance.

Model

To address this problem, the team developed a model that estimates heat risk based on satellite data. The model currently includes the Normalized Difference Moisture Index (NDMI) and the Normalized Difference Vegetation Index (NDVI). We also considered Land Surface Temperature (LST), but since we ultimately decided to implement the model based only on Sentinel-2 data we did not integrate LST. At this point, however, it should be noted that a model based on Landsat-8 data could use all three indices. However, since the spatial resolution of Landsat-8 is not as good as Sentinel-2 , we decided to use Sentinel. By combining the raster values of each index and setting appropriate thresholds, a heat risk map can be generated. Before calculating the indices, the clouds were masked from the satellite images because Sentinel-2 and Landsat-8 are optical satellites and the signals cannot penetrate the clouds.

The model was implemented using the Google Earth Engine. This has the advantage that satellite data can be easily obtained and analyzed. Especially with respect to raster data, Google Earth engine provides a number of valuable tools. A schematic overview of the proposed model is shown in Figure 1. The model is designed in such a way that it can be easily extended.

Fgure 1 - Model to calculate the risk of heat-up areas

Normalized Difference Vegetation Index (NDVI)

• (NIR - R) / (NIR + R), for Sentinel-2, NDVI = (Band 8 – Band 4) / (Band 8 + Band 4)
• "NDVI is a measure of the state of plant health based on how the plant reflects light at certain frequencies (some waves are absorbed and others are reflected). Chlorophyll (a health indicator) strongly absorbs visible light, and the cellular structure of the leaves strongly reflect near-infrared light. When the plant becomes dehydrated, sick, afflicted with disease, etc., the spongy layer deteriorates, and the plant absorbs more of the near-infrared light, rather than reflecting it. Thus, observing how NIR changes compared to red light provides an accurate indication of the presence of chlorophyll, which correlates with plant health." (Source: eos)

Normalized Difference Moisture Index (NDMI)

• (NIR - SWIR) / (NIR + SWIR), for Sentinel-2, NDMI = (Band 8– Band 11) / (Band 8 + Band 11)
• "The Normalized Difference Moisture (Water) Index (NDMI or NDWI) is a satellite-derived index from the Near-Infrared (NIR) and Short Wave Infrared (SWIR) channels. The SWIR reflectance reflects changes in both the vegetation water content and the spongy mesophyll structure in vegetation canopies, while the NIR reflectance is affected by leaf internal structure and leaf dry matter content but not by water content. The combination of the NIR with the SWIR removes variations induced by leaf internal structure and leaf dry matter content, improving the accuracy in retrieving the vegetation water content. The amount of water available in the internal leaf structure largely controls the spectral reflectance in the SWIR interval of the electromagnetic spectrum. SWIR reflectance is therefore negatively related to leaf water content. NDWI is computed using the near infrared (NIR) and the short wave infrared (SWIR) reflectance’s" (Source: LSRS)

Land Surface Temperature (LST)

• "Land Surface Temperature (LST) is the radiative skin temperature of the land derived from infrared radiation. In the SLSTR project, "skin" temperature refers to the temperature of the top surface when in bare soil conditions, and to the effective emitting temperature of vegetation "canopies" as determined from a view of the top of a canopy." (Source: ESA)

Study area

Canada was chosen as the study area for this project. This is due to the fact that it would be too computationally intensive to apply the model globally and also because Canada has been in the focus of the global community this year in connection with heat waves. For weeks, the heat wave had Canada firmly in its grip, and Lytton in British Columbia even recorded a new heat record of 49.6 °C on June 29, 2021 (see Figure 2). This heat event cannot be entirely attributed to global climate change, but it is clearly amplifying it. Not only do people suffer from the heat, but long-term damage can also be caused to animal and plant life. Among humans, the elderly are particularly affected by the heat, and for many it represents a life-threatening situation.

Figure 2 - Excerpt from a BBC report from July 2, 2021 (Source: BBC News)

Heat warning application

It is therefore important that the population can find out whether they live in an area that is potentially affected by heat. The easiest way to do this is with an app that can be accessed directly on a smartphone or tablet. However, the application is not only intended to inform the user about the general situation, but also to provide further information and thus raise awareness of the issue. In addition, the user receives tips on how to behave in the event of a heat wave and how he or she can generally help to prevent man-made climate change from intensifying.

As part of the project, a web application was developed that presents the results of the model described above in an appealing and simplified way for the general public. This web application was also created with the Google Earth Engine, which limits the design elements of the web application, but the results of the analysis of the satellite images can be easily integrated. Therefore, the focus of this first draft of the web application is clearly on the presentation of the evaluated data and less on the external appearance.

Figure 3 - Screenshot of the web application

Crowdsourcing

In addition to the satellite imagery used, other data can add value to the model and the application. For example, crowdsourced data can be included in the analysis as an additional source of information. In our case, this means that data collected by third parties is included in the analysis. Here, it makes sense to tap into existing data sources that are maintained and kept up to date by a large number of users. Therefore, witter could be a good choice.

The advantage of Twitter data is that it can be obtained and analyzed in real time via the Twitter API. For example, we could search for current trends in tweets that are related to heat and spatially located within the area of interest (#heatwave). In a first step, the data could be added as poups to the map in the web application, so that the user can see all relevant tweets in his area and thus get an idea of the current situation. In the second step, the found data could be integrated into the model described above. Since the information can be obtained from Twitter in real time, it is a good complement to the satellite data and thus provides added value to the model.

The Twitter data could also be used to warn the public of rapidly developing events. These include, in particular, forest fires, which can spread very quickly and thus pose a serious threat to the population. The fact that other people report on a fire on Twitter means that other residents in the neighborhood can be warned at an early stage. Currently, Twitter data is not yet integrated into our project, but as we just explained, it can be a useful extension to our model.

Maple Heat

For our project, we used Sentinel-2 data from the EUROPEAN SPACE AGENCY (ESA) to calculate the above indices. In addition, we used Landsat-8 data in the development process of the model. With the Landsat-8 data, we also managed to include the LST in the model.

For the hackathon, the three of us joined forces as work colleagues. Each of us brings different skills to the team, which makes working together not only productive but also fun. In our daily work life, our topics don't always overlap, so we rarely work together directly. The hackathon weekend therefore gave us the chance to work together creatively and freely on a topic that we all find exciting.

We chose the challenge "Warning: Things are heating up" because it is a very current topic/problem that we can all identify with. Since we already have a background in image processing and software development, the challenge seemed to fit us perfectly. Already early in the weekend we decided to use the Google Earth Engin as a tool. This simplifies the retrieval of satellite data and its analysis. Creating the web application using the Google Earth Engin editor, on the other hand, was often a challenge. Since the functions are very limited, it is difficult to develop a web application with the same standards as one is used to.

• Google Earth Engine
• Sentinel-2 (and Landsat -8) data
• Administrative boundaries of the Canadian provinces
• US-Canada heatwave: Visual guide to the causes
• 9 things you can do about climate change
• Normalized Difference Vegetation Index (NDVI)
• Normalized Difference Moisture Index (NDMI)
• Land Surface Temperature

#satelliteImages #heatwave #GoogleEarthEngine #sentinel #twitterAPI #crowdsourcing #canada #forestFire

This project has been submitted for consideration during the Judging process.