Skip to main content
European Commission logo print header

Multi-frequency RADAR imaging for the analysis of tropical forest structure in the Amazon

Periodic Reporting for period 1 - MF-RADAR (Multi-frequency RADAR imaging for the analysis of tropical forest structure in the Amazon)

Reporting period: 2016-01-01 to 2017-12-31

The tropical rainforests of the Amazon basin are one of the world’s areas that are richest in biodiversity. They cover the largest terrestrial tropical biome in the world, store significant amounts of carbon and stabilise the regional and global climate. Deforestation, forest degradation and climate change impacts are posing a threat to the future of this unique region. This project helped in the capability for vegetation monitoring of the tropics. It represented an advance in the state-of-the-art because it developed innovative mapping and monitoring methods for estimating vegetation structural parameters based on the synergy between different data sources and methods. These new approaches are important in the context of carbon markets and REDD+ (Reducing emissions from deforestation and forest degradation). The importance of the project to society is that it has improved the monitoring of tropical areas to mitigate deforestation and forest degradation, and estimated the biomass and carbon stock with more accuracy, thus providing quantitative evidence needed for a better understanding of the Amazon region.
The main research goal of this project was to develop a methodology to derive information on forest structure and floristics based on the polarimetric and interferometric variables from TanDEM-X (X band), PALSAR/ALOS (L band) and Sentinel 1 (C band) integrated with geomorphometric data derived from the Shuttle Radar Topography Mission, SRTM (C band), to benefit the monitoring of plant communities in the Amazon, estimate biomass and carbon stocks.
The specific research objectives were:
1. To determine which variables/attributes (geomorphometric and SAR variables) are related to the floristic composition and structure of forest plant communities;
2. To develop a methodology to derive information on forest floristic structure from the integration of multi-frequency (2 to 8 GHz) SAR data and geomorphometry;
3. To develop and validate a predictive model for forest structure, especially the above ground biomass and floristics based on the integration of different data sources;
4. To quantify the relationship between biomass and the rate of carbon stocks in tropical areas.
The fellow developed a multi-frequency SAR method for mapping forest structure and biomass in the Amazon. Open access publications describe the progress made for each objective. From the secondment, courses, conferences and meetings, the fellow learned new skills and improved her existing skills in radar remote sensing. She also increased her professional network and has secured another research position at the University of Leicester on a project called Forest 2020 ( Forests 2020 is a major investment by the UK Space Agency as part of the International Partnerships Programme (IPP) to help protect and restore up to 300 million hectares of tropical forests by improving forest monitoring in six partner countries through advanced uses of satellite data.
"From the beginning of the project to the end (from 01/01/2016 to 31/12/2017) the main results achieved so far are:

1. The SAR images (TanDEM-X, Sentinel 1, SRMT e ALOS-PALSAR 2) were processed and analysed.
2. Dr. Bispo leaded a fieldwork in September 2016 in Tapajos National Forest.
3.The results of the project are being published in several different papers to international journals instead of to be disseminated in just one large paper. The first two papers published are focused in understand which geomorphometrics variables derived from SRTM have more correlation with floristic composition and structure of forest plant communities (see in the final report).
4. At the moment we have 3 papers in preparation:
The goal of the first paper in preparation ""Mapping forest structural types and change detection in the Brazilian Amazon with interferometric height from TanDEM-X SAR data"" is to analyse the potential application of the interferometric height estimated from TanDEM-X data to discriminate the horizontal patterns of forest structural types (primary and secondary forest in different successional stages) and monitoring the changes in height.

The goal of the second paper ""Multifrequency approach to carbon and biomass estimation at Brazilian Amazon"" is the integration SAR multifrequency based on the polarimetric and interferometric variables from TanDEM-X (X band), PALSAR/ALOS (L band), Sentinel 1 (C band) integrated with geomorphometric data derived from SRTM (C band) to benefit the monitoring of forest communities, estimates of biomass and carbon stock. The results show that the multifrequency approach provide a better result when compared when the images from the individual sensors.

The goal of the third paper ""Aboveground carbon stocks estimates in different types of tropical forests using full polarimetric ALOS-PALSAR 2""the goal of the present study is to evaluate the aboveground carbon stock differences, among four forests in tropical areas using coherent and incoherent polarimetry attributes derived from ALOS PALSAR 2. The ALOS/PALSAR-2 is a continuation of the ALOS-1 mission and has significantly improved resolution capability and shorter time of revisit (Jaxa, 2014). For reach the purpose of our study we evaluate the potential of the ALOS PALSAR 2 to estimate AGB, map the spatial distribution of AGB and the associated uncertainties to those estimates.

The dissemination was done using the project website, conferences, meetings and publishing papers. The complete list is in the final report attached."
The project represents an advance in the analysis of satellite radar images for studying the forest structure and floristics, focusing especially on biomass estimation in tropical areas, which is a field of growing global importance. It is being created the first estimates of vegetation structure using topographic information integrated with SAR data acquired at different frequencies.
It is also expected that this project contributes to monitoring programs of the Amazon, e.g. PRODES (Amazon Deforestation Monitoring Project). Within PRODES deforestation is monitored by satellites since 1988 in order to retrieve the annual rates of deforestation in the region, which are used by the Brazilian government for the establishment of public policies. The main challenge of the Amazon monitoring programme is the cloud cover and complexity of floristic and forest structure. Therefore, our project can contribute to the monitoring of the Amazon program by adding information derived from RADAR data. This can result in economic and social impacts in the medium and long term, especially given the availability of RADAR Sentinel 1 (C band) with high temporal resolution and free acess. This contribution is of great importance for the initiative to reduce emissions from deforestation and forest degradation (REDD +) and also emphasizing the ecosystem services that these kind of forests provide. Among them we can highlight: genetic resources, biochemicals, natural medicines, pharmaceuticals, regulation of air quality, climate (global, regional and local), water, erosion , water purification, spiritual and religious value, aesthetic value , recreation and ecotourism, soil formation and minerals, photosynthesis, nutrient cycling etc. Thus our project shows that it has good potential contributing and promoting positive social and economic impacts.