Olofsson (CMS 2015) Project Profile   (updated 21-Sep-2020)
Project Title:Tracking carbon emissions and removals by time series analysis of the land surface: prototype application in tropical MRV systems compliant with IPCC Tier 3

Science Team
Members:

Pontus Olofsson, NASA MSFC (Project Lead)

Project Duration: 2016 - 2019
Solicitation:NASA: Carbon Monitoring System (2015)
Successor Projects: Woodcock (CMS 2018)  
Abstract: Many tropical countries are experiencing high rates of forest disturbance with of cycles of degradation, cultivation and recovery, but neither the activities nor the terrestrial carbon dynamics associated with the activities are properly tracked in existing REDD+ related Measurement, Reporting and Verification (MRV) systems. This situation is especially true for post-disturbance landscapes and degraded forests, as the trajectories of the land surface activities and carbon dynamics following disturbance are gradual in nature and inherently difficult to monitor. We propose to improve modeling of the carbon dynamics of areas that have experienced disturbance by combining a time series-based approach for monitoring changes on the land surface with a spatially and temporally explicit carbon bookkeeping approach. We have developed algorithms that track the land surface by analyzing time series of all available observations from the Landsat sensors complemented by data from space-borne radar instruments and other optical sensors. Implementations are currently underway across the United States and the Colombian Amazon. Additionally, we have developed open source software tools and educational materials that provide detailed hands-on instructions in support of capacity building efforts in collaboration with SilvaCarbon. We propose a novel framework for estimation of carbon emissions and removals by including detailed information on the fate of the landscape. We will modify a recently developed bookkeeping model so that it runs at the pixel-level (spatially explicit) by directly integrating the results of time series information on conversion between land categories and forest degradation. The characterization of post-disturbance tropical landscapes is critical for accurate accounting of terrestrial carbon pools and fluxes because of the high productivity and carbon density of forests in this region. Therefore, in addition to the time series analysis of the land surface, the temporal dynamics of vegetation structure and recovery following disturbance will be investigated using existing space-borne lidar data in combination with data from upcoming NASA lidar missions. Following best practices protocols for statistical inference of change in area and carbon emissions, unbiased estimates with the uncertainty quantified in the form of confidence intervals will be constructed. Prototype applications of the proposed methodology will be implemented in Colombia and Cambodia, two tropical countries representing different levels of capacity and different types of forest disturbance. A SilvaCarbon effort is underway to complete a comprehensive time series-based analysis of the conversions between the land categories and post-disturbance landscapes across the Colombian Amazon that will be used together with a set of existing field measurements of biomass in a prototype application of the proposed methodology. The methodology will be implemented in Cambodia, where field-measured data on biomass are scarcer, capacity needs greater and the rate of deforestation and forest degradation higher. Engagement with stakeholders and countries will be enhanced by collaboration with in-country SilvaCarbon activities focused on enhancing and supporting systems of MRV for REDD+ activities (including the provision of input for designing field measurement programs). In addition, a spatially and temporally explicit model for estimating the carbon dynamics related to land surface activities will be added to the open source suite of software to provide a more complete framework for the enhancement of MRV systems in the tropics.
CMS Primary Theme:
  • Land Biomass
CMS Science Theme(s):
  • Land Biomass
  • Land-Atmosphere Flux
  • MRV

Participants:

Gustavo Galindo, IDEAM, Ministry of Environment and Sustainable Development, Colombia Government
Lucy Hutyra, Boston University
Pontus Olofsson, NASA MSFC
Andrew Reinmann, Boston University
Curtis Woodcock, Boston University

Contact Support to request an email list of project participants.

Project URL(s): None provided.
 
Data
Products:
Product Title:  Data cube of Landsat data, land cover and land cover conversion of the country of Cambodia
Time Period:  2000-2016
Description:  Data cube constructed using data from Landsats 5,7,8 to which CCDC has been applied for classification and change detection ("activity data"). Surface reflectance predictions for each spectral band allows for generation of synthetic Landsat for any date during the time period
Status:  Planned
CMS Science Theme(s):  Decision Support; MRV
Keywords:  MRV, activity data, REDD, IPCC, data cube, Landsat, land cover, land use, land change
Spatial Extent:  Cambodia
Spatial Resolution:  30 m
Temporal Frequency:  Daily
Input Data Products:  Surface reflectance from Landsat TM/ETM+/OLI; training data
Algorithm/Models Used:  CCDC
Evaluation:  Stratified estimation protocol implemented for construction of unbiased estimators of area and accuracy with confidence interval
Intercomparison Efforts/Gaps:  No
Uncertainty Estimates:  Each mapped category (i .e. each category of activity data) is estimated from sample data including 95% confidence intervals
Uncertainty Categories:  1. Ensemble (design-based inference)
Application Areas:  MRV, deforestation and land use related policies
Relevant Policies/Programs:  UNFCCC, REDD, REDD+
Potential Users:  Governments, NGOs, academia
Stakeholders:  
Current Application Readiness Level:  6
Start Application Readiness Level:  4
Target Application Readiness Level:  9
Future Developments:  Currently being constructed
Limitations:  High processing and storage requirements
Date When Product Available:  Summer 2017
Metadata URL(s):
Data Server URL(s):
Archived Data Citation:  
Bounding Coordinates:
West Longitude:0.00000 East Longitude:0.00000
North Latitude:0.00000 South Latitude:0.00000

Product Title:  Spatiotemporal estimates of carbon emissions and removals resulting land change in Cambodia
Time Period:  2000-2016
Description:  Carbon dynamics esimated by applying bookkeeping model to data cube of land conversion
Status:  Planned
CMS Science Theme(s):  Decision Support; MRV
Keywords:  Carbon, emissions factors, MRV, activity data, REDD, IPCC, data cube, Landsat, land cover, land use, land change
Spatial Extent:  Cambodia
Spatial Resolution:  30 m
Temporal Frequency:  Daily
Input Data Products:  Data cube, carbon content variables, growth curves, emssion curves
Algorithm/Models Used:  C-CCDC (carbon bookkeeping model coupled with CCDC)
Evaluation:  None yet
Intercomparison Efforts/Gaps:  No
Uncertainty Estimates:  None yet
Uncertainty Categories:  1. Ensemble (design-based inference) planned
Application Areas:  MRV, deforestation and land use related policies
Relevant Policies/Programs:  UNFCCC, REDD, REDD+
Potential Users:  Governments, NGOs, academia
Stakeholders:  
Current Application Readiness Level:  4
Start Application Readiness Level:  2
Target Application Readiness Level:  9
Future Developments:  Currently being constructed
Limitations:  How to translate population-scale parameters of bias and uncertainty to the pixel level
Date When Product Available:  2018
Metadata URL(s):
Data Server URL(s):
Archived Data Citation:  
Bounding Coordinates:
West Longitude:0.00000 East Longitude:0.00000
North Latitude:0.00000 South Latitude:0.00000

Product Title:  Data cube of Landsat data, land cover and land cover conversion of the Colombia Amazon
Time Period:  2000-2016
Description:  Data cube constructed using data from Landsats 5,7,8 to which CCDC has been applied for classification and change detection ("activity data"). Surface reflectance predictions for each spectral band allows for generation of synthetic Landsat for any date during the time period
Status:  Public
CMS Science Theme(s):  Decision Support; MRV
Keywords:  MRV, activity data, REDD, IPCC, data cube, Landsat, land cover, land use, land change
Spatial Extent:  The Colombian Amazon
Spatial Resolution:  30 m
Temporal Frequency:  Daily
Input Data Products:  Surface reflectance from Landsat TM/ETM+/OLI; training data
Algorithm/Models Used:  CCDC
Evaluation:  Stratified estimation protocol implemented for construction of unbiased estimators of area and accuracy with confidence interval
Intercomparison Efforts/Gaps:  No
Uncertainty Estimates:  Each mapped category (i .e. each category of activity data) is estimated from sample data including 95% confidence intervals
Uncertainty Categories:  1. Ensemble (design-based inference)
Application Areas:  MRV, deforestation and land use related policies
Relevant Policies/Programs:  UNFCCC, REDD, REDD+
Potential Users:  Governments, NGOs, academia
Stakeholders:  
Current Application Readiness Level:  8
Start Application Readiness Level:  4
Target Application Readiness Level:  9
Future Developments:  Currently being tested by implementing agency in Colombia
Limitations:  High processing and storage requirements
Date When Product Available:  Spring 2017
Metadata URL(s):
Data Server URL(s):
Archived Data Citation:  
Bounding Coordinates:
West Longitude:0.00000 East Longitude:0.00000
North Latitude:0.00000 South Latitude:0.00000

Product Title:  Data cube of pixel-level omission and commission error probabilities of land cover conversion of the Colombia Amazon
Description:  
Status:  Public
CMS Science Theme(s):  
Keywords:  
Spatial Extent:  
Spatial Resolution:  
Temporal Frequency:  
Input Data Products:  
Algorithm/Models Used:  
Evaluation:  
Intercomparison Efforts/Gaps:  
Uncertainty Estimates:  
Uncertainty Categories:  
Application Areas:  
Relevant Policies/Programs:  
Potential Users:  
Stakeholders:  
Current Application Readiness Level:  6
Start Application Readiness Level:  4
Target Application Readiness Level:  9
Future Developments:  
Limitations:  
Date When Product Available:  
Metadata URL(s):
Data Server URL(s):
Archived Data Citation:  
Bounding Coordinates:
West Longitude:0.00000 East Longitude:0.00000
North Latitude:0.00000 South Latitude:0.00000

Product Title:  Spatiotemporal estimates of carbon emissions and removals resulting land change in the Colombia Amazon
Time Period:  2000-2016
Description:  Carbon dynamics esimated by applying bookkeeping model to data cube of land conversion
Status:  Public
CMS Science Theme(s):  Decision Support; MRV
Keywords:  Carbon, emissions factors, MRV, activity data, REDD, IPCC, data cube, Landsat, land cover, land use, land change
Spatial Extent:  The Colombian Amazon
Spatial Resolution:  30 m
Temporal Frequency:  Daily
Input Data Products:  Data cube, carbon content variables, growth curves, emssion curves
Algorithm/Models Used:  C-CCDC (carbon bookkeeping model coupled with CCDC)
Evaluation:  None yet
Intercomparison Efforts/Gaps:  No
Uncertainty Estimates:  None yet
Uncertainty Categories:  1. Ensemble (design-based inference) planned
Application Areas:  MRV, deforestation and land use related policies
Relevant Policies/Programs:  UNFCCC, REDD, REDD+
Potential Users:  Governments, NGOs, academia
Stakeholders:  IDEAM, Ministry of Environment and Sustainable Development, Colombia Government (Point of Contact: Gustavo Galindo (gusgalin@gmail.com))
Current Application Readiness Level:  6
Start Application Readiness Level:  2
Target Application Readiness Level:  9
Future Developments:  Currently being constructed
Limitations:  How to translate population-scale parameters of bias and uncertainty to the pixel level
Date When Product Available:  Fall 2017
Metadata URL(s):
Data Server URL(s):
Archived Data Citation:  
Bounding Coordinates:
West Longitude:0.00000 East Longitude:0.00000
North Latitude:0.00000 South Latitude:0.00000

Product Title:  CMS: Landsat-derived Annual Land Cover Maps for the Colombian Amazon, 2001-2016
Start Date:  01/2001      End Date:  12/2016     (2001-2016)
Description:  This dataset provides annual maps of land cover classes for the Colombian Amazon from 2001 through 2016 that were created by classifying time segments detected by the Continuous Change Detection and Classification (CCDC) algorithm. The CCDC algorithm detected changes in Landsat pixel surface reflectance across the time series, and the time segments were classified into land cover types using a Random Forest classifier and manually collected training data. Annual maps of land cover were created for each Landsat scene and then post-processed and mosaicked. Land cover types include unclassified, forest, natural grasslands, urban, pastures, secondary forest, water, or highly reflective surfaces. The training data are not included with this dataset.
Status:  Archived
CMS Science Theme(s):  Land Biomass
Keywords:  
Spatial Extent:  Colombian Amazon
Spatial Resolution:  30 m
Temporal Frequency:  annual
Input Data Products:  
Algorithm/Models Used:  
Evaluation:  
Intercomparison Efforts/Gaps:  
Uncertainty Estimates:  
Uncertainty Categories:  
Application Areas:  
Relevant Policies/Programs:  
Potential Users:  
Stakeholders:  
Current Application Readiness Level:  
Start Application Readiness Level:  
Target Application Readiness Level:  
Future Developments:  
Limitations:  
Date When Product Available:  
Assigned Data Center:  ORNL DAAC
Metadata URL(s):

https://doi.org/10.3334/ORNLDAAC/1783
Data Server URL(s):

https://doi.org/10.3334/ORNLDAAC/1783
Archived Data Citation:  ArĂ©valo, P. 2020. CMS: Landsat-derived Annual Land Cover Maps for the Colombian Amazon, 2001-2016. ORNL DAAC, Oak Ridge, Tennessee, USA. DOI: 10.3334/ORNLDAAC/1783

Bounding Coordinates:
West Longitude:-78.03000 East Longitude:-65.95000
North Latitude:5.38000 South Latitude:-3.88000

 
Publications: Arevalo, P., Bullock, E. L., Woodcock, C. E., Olofsson, P. 2020. A Suite of Tools for Continuous Land Change Monitoring in Google Earth Engine. Frontiers in Climate. 2. DOI: 10.3389/fclim.2020.576740

Arevalo, P., Olofsson, P., Woodcock, C. E. 2020. Continuous monitoring of land change activities and post-disturbance dynamics from Landsat time series: A test methodology for REDD+ reporting. Remote Sensing of Environment. 238, 111051. DOI: 10.1016/j.rse.2019.01.013

Bullock, E. L., Woodcock, C. E., Souza, C., Olofsson, P. 2020. Satellite-based estimates reveal widespread forest degradation in the Amazon. Global Change Biology. 26(5), 2956-2969. DOI: 10.1111/gcb.15029

Olofsson, P., Arevalo, P., Espejo, A. B., Green, C., Lindquist, E., McRoberts, R. E., Sanz, M. J. 2020. Mitigating the effects of omission errors on area and area change estimates. Remote Sensing of Environment. 236, 111492. DOI: 10.1016/j.rse.2019.111492

Tang, X., Hutyra, L. R., Arevalo, P., Baccini, A., Woodcock, C. E., Olofsson, P. 2020. Spatiotemporal tracking of carbon emissions and uptake using time series analysis of Landsat data: A spatially explicit carbon bookkeeping model. Science of The Total Environment. 720, 137409. DOI: 10.1016/j.scitotenv.2020.137409

Tang, X., Woodcock, C. E., Olofsson, P., Hutyra, L. R. 2021. Spatiotemporal assessment of land use/land cover change and associated carbon emissions and uptake in the Mekong River Basin. Remote Sensing of Environment. 256, 112336. DOI: 10.1016/j.rse.2021.112336

Archived Data Citations: Arévalo, P. 2020. CMS: Landsat-derived Annual Land Cover Maps for the Colombian Amazon, 2001-2016. ORNL DAAC, Oak Ridge, Tennessee, USA. DOI: 10.3334/ORNLDAAC/1783