Webinar Questions: How to conduct rapid eligibility assessments for forest carbon projects using geospatial data
Modified on Fri, 3 Mar, 2023 at 2:26 PM
TABLE OF CONTENTS
- Classification
- Datasets
- Biomass
- Earth Blox Features and Capabilities
- Integrations and APIs
- Resolution
- Risk Assessments
Classification
| Can Earth Blox make a distinction between native forests and exotic plantations? | Historical time series analysis and the combination of different sensors help to do that. There are some open shapefiles for some countries that define plantations and you can upload those to Earth Blox. Earth Blox project workflows can visualise timelines for forests/vegetation from different data sources including:
|
| Can you make a distinction between dry forests and dry bushes in areas where rain is scarce? | Indeed, distinguishing between trees and bushes can be tricky with medium-to-high-resolution imagery. The Copernicus 100m land classification has some separation of forest classes. Very high-resolution imagery could help here, as by visual inspection longer shadows, or the lack of them, can suggest the presence of trees or bushes, respectively. Try looking at the NICFI 5m Planet base map in the map layer. If you have a licence to a commercial data set this can be made available within Earth Blox (for example, we already have clients using their licence to high-resolution Maxar data). You can find a workflow for the visualisation of optical data in the template project:
|
| What is the difference between a stressed forest and a destroyed forest? | Stressed forests are hotspots for forest decline. Phenology and effects on forests by extreme weather events can be monitored by satellite data with a short revisit period. This can help in the early identification of harmful conditions for forests and in taking actions towards restoration. In Earth Blox, you can find template workflows which return forests coverage, biomass, fire risks and damages, vegetation indices, weather condition, and urban expansion, amongst others. These data support decision-makers in taking the proper actions to avoid forest decline and death. A destroyed forest is an area where the forest is actually lost, due to illegal logging, fires, or insect infestations, for example. These areas will suddenly appear as tree-free areas in a time series with a sufficient revisit time. You can find examples of workflow visualising a timeline for forest and vegetation under the following templates:
|
| Is it possible to classify land cover? | Yes, we have a range of options for classifying land cover. Either by using off-the-shelf products like the Dynamic World workflow (which gives you weekly classification) or one of the annual classification maps. Alternatively, you can perform your own classification from a range of options for both supervised, and unsupervised classification. A tutorial to apply image classification (supervised or unsupervised) can be found in Earth Blox Knowledge Hub. |
Datasets
| Different data sources have different granularities and methodologies. How have you standardised the data sets, what is the resolution you provide, and what are the data sets you are using? | We use data sets that vary in spatial resolution from 5m to 10's km. The resolution of the output is set by the input. The processing is able to combine images of different resolutions. There are now over 30 datasets available in Earth Blox and it is growing every week. |
| Is Hansen data only applicable to deforestation? Do you have other ways to assess degradation? | There are various published attempts at mapping degradation. If there is a methodology you trust, then we can implement it in Earth Blox. Hansen is well known for forest loss and gain, but as a global dataset, you should always check its quality at the project level. |
| What version of Sentinel data do you have and is it possible to see February 20223? | All of Sentinel 1 (IW mode) and Sentinel 2 data are available in Earth Blox. S1 is GRD data that has been terrain corrected but not terrain flattened. |
| How is the GEDI data available and what time frame does it cover? | GEDI is not yet included as a dataset on Earth Blox because it occurs as strips of samples, not wall-to-wall images. We do include derivative datasets from the GEDI platform such as the Earth Blox global biomass map. Data from GEDI spans 2019-22. |
| Are you only dealing with land cover types or also with quantification, e.g., water flows, and forest biomass? | There are many different data layers available in Earth Blox. Some are the original satellite data in an analysis-ready form. Some are thematic layers such as land classification maps. Others are derived product layers such as forest biomass or surface water history, to name just two. Earth Blox also allows you to derive your own land classification or derived product. For example, you can apply your own image classification (supervised or unsupervised). Quantification of a specific variable can be performed by using the Calculator Block. See in the Knowledge Hub. Examples of template projects in Earth Blox include:
|
| Does Earthblox use only Google Earth Engine datasets? | Earth Blox uses data from a number of sources, including Earth Engine. For some of our customers, we have integrated their commercial data. |
| Are there these types of layers and data available for Europe? | Yes, all the datasets you have seen today are available globally. The one exception is the NICFI 5m base map which only covers the tropical region. |
| Where do we find the source of the layer? | The dataset finder in Earth Blox provides a range of external sources for finding out more information about the datasets. This ensures you have full sight of the source of the day and how it has been processed. |
| Can we expect to use PRODES and Mapbiomas Alert specific datasets ingested? (Brazilian very common data sources). | Our product team are looking at how we can bring these data sources into Earth Blox for you to use. We frequently get requests for new data layers and we prioritise them based on user needs. |
| Does Earthblox have InSAR data from Sentinel-1? | Currently, InSAR data is not available from Sentinel-1. We are hoping to be able to provide this through our data suppliers soon. |
| Is it possible to import one's own data for analysis? | Yes, of course. To import your own shape file see the article Uploading your own area of interest in the Knowledge Hub. If you have external data, including commercial data with a user licence, please contact the Earth Blox team so that we can look at the best way to make that available for use in the platform. |
| If we bring in our own commercial satellite data, can we use these images to create our own LULC or biomass maps in the platform? | Yes. If you have external data, including commercial data with a user licence, please contact the Earth Blox team so that we can look at the best way to make that available for use in the platform. You then have a range of options for undertaking your own LULC analysis. Including supervised and unsupervised options. See the article: Appling image classification (supervised or unsupervised) in the Knowledge Hub. Template projects in Earth Blox:
|
| Could I upload my own thematic temporal series? | If you have external data, including commercial data with a user licence, please contact the Earth Blox team so that we can look at the best way to make that available for use in the platform. |
| If we have our own dataset, is it also integrated into the dashboard as well for the zonal statistics thing? | If you have external data, including commercial data with a user licence, please contact the Earth Blox team so that we can look at the best way to make that available for use in the platform. |
Biomass
| How do you estimate biomass? | The calculation of biomass shown in the webinar is based on satellite radar data and follows the peer-reviewed article: Avtar et al 2013, PLOS ONE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism" Note that the biomass equation used here is typical for the geographic area. Earth Blox will be providing some alternative equations, or you can find coefficients for other areas of the world from the peer-reviewed literature. Let us know if you are having difficulty and we will help you find the right one. The workflow shown in the webinar can be found under the template projects:
|
| Can you discuss the ability to discern forest biomass by specific types of trees and growth patterns over time? Is this mostly a responsibility on the user end in customizing the data, algorithmic, or both? | The webinar example was a quick demonstration for early appraisal of forest projects. The same methods can be used to look at regrowth as well as forest loss. The flexibility of Earth Blox allows the user to bring their own insights to the problem to improve the results. This may well include the use of forest classification to stratify the types of trees. If you need advice on how best to do this, contact the Earth Blox team. |
| Can you share the article name used to calculate the biomass? | Avtar et al 2013, PLOS ONE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism" |
| What is the accuracy of the biomass estimation? How close are the biomass calculation from Earth Blox in comparison to the existing projects, where field data exists? | Generally, the accuracy of biomass estimation varies from project to project, depending on the accuracy of the in-situ measurements, the quality of satellite or airborne data, and the sampling correlation biases, among others. The accuracy of the algorithm used to estimate the biomass in the webinar can be found in the published paper. Avtar et al 2013, PLOS OJE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism" |
| Is it possible to specify by specific methodology number by different standards? This really matters for soil carbon pools for biomass inclusion. | The calculation of biomass shown in the webinar is based on satellite radar data and follows the peer-reviewed article: Avtar et al 2013, PLOS OJE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism" Although the biomass equation used is typical for the geographic area, the right coefficients for other areas of the world are covered in peer-reviewed literature. The calculator block allows you to add your own algorithms, so yes, it is fully customisable. See how to use the Calculator Block in the Knowledge Hub: 2.0 Applying your own algorithms: The Calculator Block Soil carbon is difficult to measure by satellite, but Earth Blox give you access to a belowground database, "Belowground living biomass carbon stock density" See for example the template project:
|
| Does the carbon content data include soil carbon? Can it be assessed as well? | No. It is difficult to measure soil carbon from space. Models that link surface carbon to soil carbon exist. The belowground biomass carbon stock density for 2010 is provided in the dataset: "Belowground living biomass carbon stock density" in the Dataset finder, and selecting the belowground biomass band. Note that this is model-based, not sample-based. On how to use the input blocks, please refer to the following article in the Knowledge Hub: 2.0 Explanation of the INPUT blocks in the Toolbox See for example the template project:
|
| Is this only for tree biomass? What about grassland and soil carbon? | Although it is difficult to assess soil carbon from space (see above), a combination of different satellite data sources and expert modelling could help in evaluating grassland carbon. Talk to the Earth Blox team to develop a suitable approach. Moreover, belowground living biomass carbon stock density is provided in the dataset: "Belowground living biomass carbon stock density" in the Dataset finder. |
| I wonder why ESA launches a Biomass mission if you already have high-quality data now from a Sentinel 2 mission? | The ESA BIOMASS mission is a P-band radar which is designed to deliver crucial information about high-biomass forests. P-band is particularly good at this. All current methods struggle at high biomass (>300 t/ha) mapping. BIOMASS will also look at the three-dimensional structure of these forests. This mission will also be time-limited, and so, like GEDI, will provide a unique snapshot of forest biomass, but won't provide continuous mapping like the Sentinel missions. |
| What was the name of the organization whose guidelines were mentioned as setting the standard for geospatial estimates of biomass? | The IPCC was mentioned as setting criteria for carbon estimates based on land use classes. The IPCC Good Practice Guidance for Land Use, Land-Use Change and Forestry (GPG-LULUCF) can be found here. |
| Does Earth Blox currently have a dataset/template for estimating forest carbon sequestration rates for North America? | The webinar demonstrates an algorithm specifically for Cambodia, but there are other algorithms available for other locations in peer-reviewed literature. Carbon sequestration is much harder to do than standing biomass. There are some products now becoming available that estimate primary productivity. But mapping carbon change is difficult. For long-term change, there are a number of carbon datasets in Earth Blox. Earth Blox provide access to off-the-shelf biomass products:
|
| How many satellite data points per year are used for a particular area? | This really depends on the specific data source. Each has its own repeat period. The dataset finder on Earth Blox provides this information on a per-dataset basis. |
| What climate data are you using to generate these maps? | In the webinar, we used the dataset ERA5-Land Monthly Averaged Reanalysis. Climate datasets accessible with Earth Blox also include ERA5-Land Hourly - ECMWF Climate Reanalysis An example of how to use climate data ERA5 is shown in the template project:
|
| Are there algorithms available for urban forest data? | Vegetation cover can be assessed also in cities if you use medium to high-resolution data. Data available in Earth Blox include optical, such as Landsat and Sentinel-2, and radar, such as PALSAR and Sentinel-1. A tutorial to apply image classification (supervised or unsupervised) can be found in Earth Blox Knowledge Hub at: https://support.earthblox.io/support/solutions/articles/101000456243-2-0-apply-image-classification-supervised-or-unsupervised-how-to-guide An example in the template projects of the use of supervised classification which uses both Sentinel-1 and Sentinel-2 data is:
|
| What can GEDI data do for the carbon project? | GEDI provides information about the height of the canopy, which has been calibrated using field data to produce carbon and biomass estimates. The Earth Blox team have incorporated GEDI data into their own global biomass product for 2020. We use this to validate/calibrate against other methods, and you are welcome to do the same. |
| What is ESA's Biomass *major* added value as you already deliver biomass data? | There are many forest biomass products available today. Some are free and open and global, but may not work so well in certain locations. Others are project-based and commercial, and very expensive. Accuracy varies across forest types and methods used. The other challenge is repeatability for long-term change mapping. With Earth Blox, you have access to both snap-shot global derived products and the original data (going back to 1982 for Landsat) so that you can get time series of change. If you have bespoke biomass maps for your area of interest and want to include them in Earth Blox, then get in touch and we can include them (only for your use). |
| Is Earth Blox involved in Verra's initiative to establish a new digital measurement, reporting, and verification (DMRV) working group? | At the moment we choose not to work with any one agency, but aim to make Earth Blox applicable to all reporting systems that require input from remote sensing data. |
| With your tool would I be able to count the number of trees lost/cut in a given area for a given time period? | The default data sets are not high enough resolution to see individual trees. Some of our customers have their own licences for very high-resolution base maps that can discern individual tree crowns. If you have access to data that you want to include in Earth Blox, please get in touch. |
| What are your forest biomass uncertainty bands? | The biomass example shown in the webinar comes from a paper published in the scientific literature. They have information on uncertainty in the paper. Avtar et al 2013, PLOS ONE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism." |
| What are your experience in mangrove forests and their biomass? | Our users have access to the Global Mangrove Watch data, which contains an annual time series of global mangrove extent. These datasets are provided in the Dataset finder. The proper algorithm to evaluate the mangrove biomass can be inserted in the Calculator Block. On how to use the input blocks, please refer to the following article in the Knowledge Hub: 2.0 Explanation of the INPUT blocks in the Toolbox Moreover, if there is a methodology you trust, then we can implement it in Earth Blox. See how to use the Calculator Block: 2.0 Applying your own algorithms: The Calculator Block |
| The demo showed biomass trends over several decades. Recognising that geospatial capabilities have improved greatly over the same time period, how does Earth Blox reconcile these differences in data quality so that a perceived change in forest cover/biomass is determined to either be true or just a change in technology? | Data based on radar backscatter is not as prone to variations of sensor development as multispectral sensors. The JAXA PALSAR provides well-calibrated values of sigma-nought, a robust physical measure of the target that allows intercomparison across radar sensors and over time. |
| Can we access Northen Europe forest biomass and current carbon budgets with Earth Blox? | The forest biomass example shown used a specific algorithm for Cambodia. However other algorithms are available in peer-reviewed literature. Moreover, Earth Blox provide access to global datasets, including "Biomass Carbon Density Maps (global, aboveground & belowground) ", used in the template project: Biomass Carbon Density Maps (global, aboveground & belowground) |
| Is Earth Blox also useful for grassland and biomass changes there? | Although it is difficult to assess soil carbon from space (see above), a combination of different satellite data sources and expert modelling could help in evaluating grassland carbon. Sentinel 1 and 2 combined are sensitive to low biomass variations. Talk to the Earth Blox team to develop a suitable approach. Moreover, belowground living biomass carbon stock density is provided in the dataset: "Belowground living biomass carbon stock density" in the Dataset finder. |
| How can Earth Blox measure biomass in tropical forests which cannot be measured with the L-band because of the saturation effect? | You are right that you need to be careful with the saturation effect. You should expect uncertainties to rise when you get over about 150 tonnes per ha. You can be confident that values above this are at least this value, but the uncertainties are higher. All satellite methods saturate, so as you get to higher biomass values, the sensitivity is much worse. Even with our 2020 forest biomass product using height information from GEDI, you should be careful how you interpret the results over about 300 t/ha. Since in the webinar, we were looking at average values over forest project areas, the saturation effect has less impact on comparing project areas or looking at changes in time. |
Earth Blox Features and Capabilities
| Are the templates available by user choice across the network of users and Earth Box? | Yes, all the examples you have seen today are available for Earth Blox users as template projects, which are already preloaded in Earth Blox. The workflows shown in the webinar are:
2.0 Explore Worked Examples. |
| Can we use Earth Blox for prediction analysis using the objectives we want from our project? | Of course, you can interpolate any of the trends you see in the future, but we would recommend you use an effective model to ensure you are comfortable with the estimated uncertainties. |
| Can Earth Blox be used to monitor the rate at which a country is losing its forest cover to trigger a fast response using satellites? | Earth Blox can support decision-making processes. Phenology and effects on forests by extreme weather events can be monitored by satellite data with a short revisit period. This can help in the early identification of harmful conditions for forests and in taking actions towards restoration or interventions. Our templates are based on well-established and solid algorithms and can be used at the national level to monitor the rate of change. |
| Can users share with each other the templates to collaborate? | Yes, where users from the same organisation can share templates and projects with each other, or modify a copy of the template and share that. |
| Does Earth Blox trap potential errors, e.g. you load an optical image and then try to apply a RADAR calibration? | The blocks are smart and will update based on the data that you have selected. For example, you can't even choose NDVI from the index block if it's within a radar data container block. |
| Can you download the information or maps or imageries from Earth Blox as GeoTIFF, JPEG, or Shapefile? | All the data you see on Earth Blox can be downloaded. The tables and charts can be exported as images and spreadsheets. The map outputs can be downloaded as PNG and GIF. Geotiff will be available soon. The polygons and shapes you see can also be downloaded as GIS-ready vector files. |
| How could Earth Blox be used to analyse whether a project has or has not inflated its baseline projection values? | Earth Blox allow users to create in a simple way an up-to-date time series of biomass so you can compare this historic change with the project documentation. Earth Blox can also be used to look for evidence of additionality, leakage or other indicators that a project is not working as it claims. See the chart created in the template project:
|
| Do you analyse the drivers of deforestation for the baselines? | Not directly, but as shown in the webinar, in Earth Blox it is possible in a very uncomplicated way to visualise and quantify several variables which can act as deforestation drivers. The high-resolution (5m) NICFI data from Planet is also available to visually assess the drivers. |
| Is it possible to project sea level rise? | You can make a map of the impact of sea level rise by using the DEM data layer and setting a height threshold. Get in touch with us if you want to see more detailed impacts or areas at risk of flooding. We have access to these datasets and can incorporate them into Earth Blox if there is sufficient interest from users. |
| We want to learn from you in terms of capacity building. | Please reach out to us about capacity building. Earth Blox is a fantastic tool for learning about remote sensing, and we have members of our team who have experience in capacity development in more than 12 countries worldwide. |
| Is it possible to use other classified products (for example with longer temporal series, or national Land Use / Land Cover maps)? If so, does it have the capability to connect to cloud repositories (Google Earth Engine, AWS etc.)? | Earth Blox has a direct connection to Earth Engine and we can incorporate any dataset that it hosts. We carefully select which EE datasets we use so that they are fit for purpose. This includes several classified land use maps, in addition to the WRI-Google Dynamic World, which is updated on every Sentinel 2 pass. If we don't have what you need, please get in touch and we will endeavour to include it. |
| Can you speak about the data collection in the field and its entry into the program? Is any prior analysis needed? | You can do a great deal of high-quality mapping using satellite data alone. We do not collect field data for customers, but we are connected to a wide range of organisations that can and we can introduce you. If you have your own field data, get in touch with us so we can discuss the most effective way to include it in your version of Earth Blox. |
| What if I have a project in an area where no algorithm is available? What is your approach there? Do we collaborate with field data to develop new algorithms? | There is a range of options available to us. Including using other datasets if necessary. This is where our team of experts can advise further depending on the specifics of your project site. If you are able to collect your own field data, we can help you convert that to a relationship that can be included in an Earth Blox workflow. |
| Is it possible to access forest degradation? | Forest degradation can take many different forms (e.g. selective logging, small-scale encroachment, etc). The best method to monitor forest degradation will depend on the nature of the degradation itself. Talk to the Earth Blox team to develop a suitable approach to monitoring forest degradation in your area of interest. |
| Can you comment on the potential of using Earth Blox to calculate/monitor dynamic baseline for a specific project? | A dynamic baseline is based on the observation with satellite data of control areas relative to a given carbon project. Earth Blox allow users to create in a simple way an up-to-date time series of biomass for the reference project as well as the active project. Biomass can be evaluated by using near-real satellite radar data on any area on the planet, given the knowledge of the appropriate coefficients for the radar backscatter-biomass equation, which are reported in peer-reviewed literature. See the chart created in the template project:
|
| Does Earth Blox find the methodology for the customers? | We can help you find the methodology that answers your question, yes. Please get in touch and let us know what you are trying to achieve and where. |
| Is Earth Blox capable of combining and exporting tabular results across different geospatial datasets (e.g., outlining landcover class, forest loss, soil types, etc. across forest subtypes)? | All the data you see in the maps can be tabulated in various ways and can be downloaded as .csv. See 2.0 Creating Tables in the Knowledge Hub. Template projects with an example of tables and a bar chart are:
|
| Is the algorithm being developed specifically as per the request of customers or does it exist for the whole globe? | The algorithm for biomass has come from the scientific literature. Avtar et al 2013, PLOS ONE. "PALSAR 50 m Mosaic Data Based National Level Biomass Estimation in Cambodia for Implementation of REDD+ Mechanism" and is specific to Cambodia. We also provide a 2020 global snapshot of biomass based on GEDI and Sentinel 2 data. See also the answer below. |
| Is Earth Blox working to analyse the whole planet? | We prioritise including datasets that cover the whole planet. We have users across the globe with interests in every corner of the planet, so we want to ensure that Earth Blox works for all of them. |
| Can analysis also be done in urban environments? | Of course! Please refer to the template projects:
|
Integrations and APIs
| Do you have integration with ESRI? | No direct integration, but always keen to understand if this is important to our customers and what they want to achieve from an integration. |
| Do you also have an API to allow developers to access the abstracted functionality/templates/datasets via code? | We are exploring API development, come and talk to us so we can understand your integration needs. |
| Can we integrate GEE Asset data into Earth Blox? | Yes, we can work with you to integrate your GEE assets into Earth Blox. |
Resolution
| Is the resolution of the calculations dependent on the zoom level? For the resolution of the download of the GeoTIFF, can it be as high as possible? | The calculations are independent of the zoom level (which only controls how it is visualised on the map as you view it). The GeoTIFF download would be available up to the native resolution of the data. |
Risk Assessments
| What is the source of flooding and precipitation? | The dataset for flooding used in the template project NbS: Analyse flood risk, comes from the JRC, which has been producing data on surface water on a monthly basis since the mid-1980s. It is based on Landsat data acquired between 1984 and 2021. In the Dataset Finder you can choose among:
You can select the total_precipitation layer using the Filter Bands block. On how to use the input blocks, please refer to the following article in the Knowledge Hub: 2.0 Explanation of the INPUT blocks in the Toolbox Examples of template projects in Earth Blox include:
|
| Do you also provide the possibility of using climate change projections to evaluate the risks in future? | This is something that we can include if there is a user need. Please get in touch with the team to discuss this. |
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article