Eyes in the sky

EYES IN THE SKY:

06 SEPTEMBER 2019

Working with satellite imagery would be much easier if there were no clouds obstructing the view from satellites that orbit the earth far above the clouds.

Luckily, there are several clever ways to solve this problem. One way is to stitch satellite images of the same area together to create a composite.

Composites can be computed for any period of time, e.g. monthly, yearly, and for any area around the world.

Italy through the clouds



We have compiled a cloud-free composite of Italy, based on thousands of individual Sentinel 2 images from 2018, which can be accessed through our SATLAS site.

Image artefacts, such as orbit path and tile seamlines, cloud and cloud shadow interferences have been removed by our processing, creating a truly analysis-ready data layer, available at full Sentinel 2 resolution.

Use it for land cover classifications or for simple visualizations.

We have also previously produced monthly Sentinel 2 composites for crop type classification in Denmark and made seasonal composites for mangrove mapping in Myanmar.

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

Monitoring climate change from space

Monitoring climate change from space

09 AUGUST 2019

Yesterday saw the release of yet another disturbing report from the United Nations (UN) Intergovernmental Panel on Climate Change (IPPC), this time underpinning how critical human exploitation of land is a causal parameter of climate change.

The big takeaway is that while the UN has repeatably advocated for the reduction of the use of fossil fuels, this will not be enough to circumvent climate change and ensure a sustainable trajectory alone.

Unsustainable land use practices and a growing population has resulted in an ever-increasing emission of greenhouse gases into the atmosphere, exacerbating climate change, which consequently worsens the state of the planet – making it largely uninhabitable in many areas.

However, while the exploitation of land, and unsustainable land use practices is a big part of the problem, contributing almost 25% of global greenhouse emissions, it is also a major part of the solution.

In DHI GRAS we have a long history of almost 20 years, developing transformative applications based on satellite-data to provide critical information on the state of our planet, ecosystems and dynamics which can be used to mitigate the impacts of climate change and contribute to increase efficiency and effectiveness in our land use systems.

Using advanced machine learning algorithms and high-resolution data from the Sentinel satellites, we can measure and monitor a long range of parameters which can be used to optimize agricultural production systems and land use practices.

  • We can monitor and map land degradation and land degradation risk, which enables decision makers to take timely action to avoid, reduce and reverse land degradation which is critical to protect critical ecosystems which maintain biomass and mitigate climate change impacts.
  • We can monitor the productivity of production systems to provide key data and information which allows farmers and smallholders to optimize the use and application of fertilizers and pesticides to limit the release of CO2 and nitrous oxides.
  • Actual evapotranspiration can be measured at field level, by estimating transpiration from the vegetation and the evaporation from soil surfaces. As climate change is significantly reducing the quantity of renewable surface water and groundwater resources, this information is critical in order to optimize the amount of carbon assimilated as biomass in the crop, per water unit used in the production. We are currently applying this approach within an irrigation scheme in Uganda, under our AI4WATER project, aiming to reduce water consumption by 30%.


    Besides contributing with data and information relevant for the optimisation of land production systems, we deliver critical data which can be used to monitor the effects of climate change and mitigate impacts.

  • Based on historical and recent satellite imagery we can extract information about coastal dynamics, providing estimations of coastline changes (meters/year) caused by factors such as sand erosion/deposition, storm surges and sea level rise.
  • We can provide near real-time information on marine ecosystems including water quality, which provides key information allowing for proactive management of water incidents, such as algal blooms caused by release of excess nutrients from agriculture, red tides or sediment plumes.
  • We can proactively monitor flood risks and map the extent of flooding events, which is critical for efficient flood risk management and land use planning.
  • We can monitor temperature changes, i.e. to identify urban heat islands, and we can map impervious surfaces which can be used by city planners and public authorities to manage and plan green spaces and manage water runoff to prevent flooding, erosion and water pollution.

    So, while the IPCC concludes that the planet is on an unsustainable trajectory, we believe we are part of the solution – by bridging critical earth observation data and relevant stakeholders, we can enable decision makers to take confirmative actions to respond to the severity of climate change.

More information about climate related services at DHI GRAS

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

Preparing national paying agencies for CAP2020+

Preparing national paying agencies for CAP2020+:

02 July 2019

We are very pleased to announce that we will be supporting the Danish Agricultural Agency (DAA) in the coming years with a “first of its kind” national monitoring system under the CAP 2020+.

Through our past projects with the DAA, we have developed in-house data processing systems that are tailor-made for these two new project wins using satellite image analyses for agricultural monitoring, where DHI GRAS is the sole lead for the first project and will be joining forces with NEO BV Netherlands for the second one.

Our processing systems are well capable of handling the large amounts of satellite data and our experienced team is ready to deliver cutting-edge remote sensing services for a national-scale agricultural monitoring system.

DHI GRAS has since its beginning in 2000 kept at the forefront of Earth Observation (EO) technology, and continuously tested and developed new methods to enhance the uptake of EO-based services.

We are therefore excited to be part of shaping the new monitoring practices for the DAA and for other countries to tap into the experience and services developed during the projects.



More information about agricultural monitoring at DHI GRAS

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

Supporting the Copernicus Land Team at the European Environment Agency

Supporting the Copernicus Land Team at the European Environmental Agency:

28 June 2019

In continuation of our excellent collaboration on previous activities with the European Environment Agency (EEA), DHI GRAS is providing on-site consultancy services for the Copernicus Land Team at the centrally-located offices of the EEA in the heart of Copenhagen.

We are supporting the Copernicus Land Team with on-going Pan-European mapping activities and advising on technical specifications including recommendations on geospatial data and systems integration, such as the integration of Copernicus Land Monitoring Service (CLMS) data in DIAS (Data and Information Access Services) platforms.

We are also producing seamless high resolution image mosaics of 39 European countries (EEA39), which are both geometrically and radiometrically corrected.

Copernicus is the European Union's Earth Observation Programme, looking at our planet to provide users in the field of environment and other terrestrial applications with information based on space data in six thematic areas: land, marine, atmosphere, climate change, emergency management and security.



For more information about our previous activities with the EEA and Copernicus Land Monitoring Service

Saving water with satellites

Saving water with satellites:

17 June 2019

A new system is being considered for inclusion in the EU’s Earth Observation programme, Copernicus, to estimate water-use by plants evapotranspiration and to improve crop water-use productivity.

The Land Surface Temperature Monitoring (LSTM) mission will be able to identify the temperatures of individual fields and would be more than 400 times as detailed as what is delivered by the current Sentinel satellite land surface temperature sensors.

The increased resolution and precision of LSTM will enable specialists to calculate in real time how much water different plants require in different areas, and how often those plants need to be irrigated. It could also be used to help predict drought conditions.

The LSTM mission will complement the existing family of Copernicus satellites, sentinels, for observing the land and coastal areas, and will hopefully become an operational spacecraft within the next decade.

Our senior remote sensing specialist, Radoslaw Guzinski, is one of the authors behind the Mission Requirements Document, and we will follow the activities closely in the coming years.


In the meantime, we are working on innovative methods and tools to estimate field-scale plant water-use with the data coming from existing Copernicus satellites. See link below for project website.

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

National Geographic Explorer Festival

National Geographic Explorer Festival:

13 June 2019

As part of our AI4Earth grant funded by Microsoft and National Geographic Society, Torsten Bondo is in Washington DC for the annual National Geographic Explorer Festival.

It is an extremely exciting and inspiring venue featuring marine biologists, glaciologists, mountaineers, photographers and many other people with an interest in natural sciences.

Torsten Bondo is there to discuss our Uganda project using AI and satellites to assist agricultural irrigation schemes and safeguard the scarce water resources. A method that has the potential to work on a global scale.

Berlingske Media has published an article about the project [in Danish].

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

World Oceans Day 2019

World Oceans Day:

08 June 2019

DID YOU KNOW

You can get bathymetry data from satellites?

Bathymetry is a key element in Hydrodynamic Modelling and Marine Spatial Planning activities.

Accurate and detailed bathymetry data is therefore an important factor for a successful implementation of a sustainable Blue Economy.

We celebrate World Oceans Day with an example of a bathymetry map from an area with intense focus on Offshore Wind Energy.

Here the Satellite-Derived Bathymetry has been merged with other existing data sets to create a seamless, fully updated and detailed bathymetry layer.

In recent years, advances in the satellite technology and physical modelling has allowed for vast improvements in the accuracy and availability of Satellite-Derived Bathymetry data.

It is also much more cost-efficient than traditional survey methods and mitigates the environmental impact of onsite surveys.

To learn more, visit:

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

Working with Forests of the World in Honduras

Working with Forests of the World in Honduras:

06 June 2019

The past two weeks Silvia Huber and Daniel Druce have been in Honduras, where they have held a hands-on training session on using specialized remote sensing tools for forest monitoring with our partner, Forests of the World (Verdens Skove) and assessed the results in the field.

The aim of the training was to teach local key individuals from Forests of the World how to use Google Earth Engine and other tools for monitoring pine forests affected by the destructive pine bark beetle, including the methodology of a fire risk analysis.

We also evaluated the effectiveness of the results, we have created for pine bark beetle monitoring. Few areas are currently under attack from pine bark beetles, which limits the validation results, but attacks are expected in the near future, which is why it is important to develop monitoring systems now so they are in place when the attacks happen again.

Burnt areas were easily detected, which is good news as both threats cause the pine trees to turn red. Therefore, it is likely, that it will also prove a good indicator of pine bark beetle attacks but further validation is necessary.

Our training was well received, and we look forward to continuing our activities in Honduras with Forests of the World.

Forest of the World has written a blog post about our joint ventures in Honduras [in Danish], follow the link.

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

World Environment Day 2019

World Environment Day 2019:

05 June 2019

Today is World Environment Day, and we want to highlight the importance of wetlands.

Wetlands are vital ecosystems for countless of species and are among the world’s most productive environments. The properties of wetlands serve as Ecosystem services ranging from freshwater supply to climate change mitigation.

Wetlands are sensitive to atmospheric disturbances, and the ecosystem services that many livelihoods depend on are threatened.

90% of all UN member states have ratified the Ramsar Convention, which provides the framework for the conservation and wise-use of wetlands and their resources.

We are part of a project led by the Ramsar Center for Eastern Africa (RAMCEA), setting out to explore the potential of Earth Observation satellite data for taking stock of and monitoring wetlands.

The project has demonstrated how Earth Observation can support wetland monitoring in Uganda and with continued training sessions can become a valuable tool for wetland managers and practitioners, not only in Uganda, but potentially also paving the way for other East African countries to achieve a complete wetland inventory in the future.

To learn more, please visit:

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871

Deep Learning

Why is it important?

Information from space is becoming much more accessible with the exponential growth in number of satellites being launched together with advancements in Big Data and Cloud Computing.

This allows for enormous quantities of data to be stored and analysed at a much faster pace and can lead to solving new challenges and optimise traditional ways of monitoring and mapping.

How does it help?

The common approach has been to train classical machine learning algorithms (e.g. Support Vector Machine or Random Forest) to label each pixel in a satellite image.

However, one of the shortcomings of this approach is that each pixel in the image is classified independently of its neighbouring pixels – meaning the spatial context in imagery is not considered.

The main advantage of Deep Learning algorithms is that they can recognize patterns, shapes, and context in imagery, and use this to better map different objects of interest.

By doing so, they use not only the spectral information from pixels, but also the surrounding spatial information that is associated with objects.

This extra information has provided a huge leap forward in computer vision capabilities and is used to more accurately identify specific objects and land cover classes of interest.

What you get!

Significant improvements in classification accuracy using the latest Deep Learning algorithms for image segmentation

Successful object detection and recognition in satellite and aerial images even with limited training data available

Significant improvements in mapping features in open landscapes using Deep Neural Networks

Typical customers?

Maritime industry for detection of reefs, hazardous objects, ships and illegal vessels

Construction industry for detection of roads, built-up areas and general activity in construction zones

Municipalities for detection of traffic patterns, car parks and illegal constructions

Specifications

We develop tailored solutions based on your specific needs.

Price list

Contact us for a quote.

Earth Observation Centre of Excellence part of the DHI GROUP

info@dhigroup.com
+45 4516 9200

Agern Alle 5,
2970 Hørsholm,
Denmark

CVR: 36466871