reduced uncertainty of 30m north american boreal …...reduced uncertainty of 30m north american...

Reduced uncertainty of 30m North American Boreal Forest CoverChristopher S.R. Neigh1, Paul M. Montesano1,2, K. Jon Ranson1, Joe Sexton3, Saurabh Channan3, Min Feng3 and John Townshend3

1Biospheric Sciences Lab., NASA GSFC , 2Science Systems Applications Inc., 3University of Maryland College Park, Geog.

Climate warming is expected to alter thedistribution of northern forests and a validatedhigh resolution baseline is required to monitorchange.

Figure 1

Figure 2 Figure 3

Earth Sciences Division – Hydrosphere, Biosphere, and Geophysics

Name: Christopher S.R. Neigh, Biospheric Sciences, NASA GSFCE-mail: [email protected]: 301-614-6681

References:2016 Montesano, P.M.; Neigh, C.S.R.; Sexton, J.; Feng, M.; Channan, S.; Ranson, K.J. & Townshend, J.R. Calibration and Validation of Landsat Tree Cover in the Taiga−Tundra Ecotone. Remote Sensing, 8, 551. 2016 Sexton, J.O.; Noojipady, P.; Song, X.P.; Feng, M.; Song, D.X.; Kim, D.H.; Anand, A.; Huang, C.Q.; Channan, S.; Pimm, S.L. & Townshend, J.R. Conservation policy and the measurement of forests. Nature Climate Change, 6, 192-+2015 Sexton, J.O.; Noojipady, P.; Anand, A.; Song, X.P.; McMahon, S.; Huang, C.Q.; Feng, M.; Channan, S. & Townshend, J.R. A model for the propagation of uncertainty from continuous estimates of tree cover to categorical forest cover and change. Remote Sensing of Environment, 156:418-425.

Data Sources: ·Multi-resolution passive optical – MODIS, Landsat , and Quickbird-2 ·LiDAR – ICESat GLAS and the Portable Airborne Laser Scanner (PALS)

Technical Description of Images: The images show pre and post calibration and validation of Landsat 30 m percent tree canopy cover (%TCC) with the difference.

Figure 1: 2010 calibrated and validated Landsat 30 m %TCC for trees taller than 2 m in height.

Figure 2: Non 2010 calibrated Landsat 30 m %TCC.

Figure 3: Difference of non 2010 calibrated minus 2010 calibrated %TCC overlaid with PALS transects used for calibration. We found satellite derived %TCC data tended to overestimate tree canopy by up to15% in the northern limit of the North American Boreal forest. This overestimation increased the uncertainty in depictions of forest cover, where small changes may reflect critical site-level drivers of forest dynamics. This work calibrated and validated Landsat-derived TCC dataset using estimates derived from long latitudinal transects of portable airborne laser scanner data.

Scientific significance societal relevance, and relationships to future missions: Forest productivity and ecosystem carbon storage is a critical component of the carbon cycle that sequesters and offsets rising fossil fuel emissions. Northern forests are currently experiencing the greatest amount of warming, responding to climate change, natural disturbances from drought, fire, pests and pathogens etc. and human induced disturbances primarily from fire and harvest. The spatial distribution of forest cover is anticipated to change in coming decades and it is poorly understood across the circumpolar domain because ~80% of the taiga-tundra ecotone is spatially diffuse and clustered in small forest patches that are difficult to discriminate by most Earth observing satellites. Our calibrated and validated estimate of tree cover will provide a baseline estimate to inform analysis of forest cover change and vulnerability in response to climate warming . Combining wall-to-wall remote sensing based estimates of forest cover with Landsat, airborne and spaceborne LiDARs, and sub-meter commercial stereo data verified with field measurements, one could infer aboveground boreal forest carbon stock and change in the Earth’s northern forests.


Much of Greenland’s ice discharge is immune to dynamic effects of refreezingKristin Poinar1,2, Ian Joughin3, Jan Lenaerts4, Michiel van den Broeke4

1Cryospheric Sciences, NASA GSFC, 2Universities Space Research Assoc., 3Univ. Washington; 4Univ. Utrecht


Meltwater can refreeze within Greenland ice and make it flow faster into the ocean, but how muchdoes this affect sea level? Using NASA data for ice velocity, we find evidence of refreezing only inice that is exposed to meltwater for long periods of time, and that ice discharge in these areas ismuch less than in areas that are less exposed to meltwater. Thus, we find that refreezing wateraffects ice discharge to the ocean only minimally; these results hold over all of Greenland.

Isunnguata Glacier

Jakobshavn Glacier

Store Glacier

Rink Glacier (14%)

(10%)

(28%)

(1%)

-375 -275 -175 -75 25Easting (km)

-2700

-2600

-2500

-2400

-2300

-2200

-2100

-2000

Nor

thin

g (k

m)

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

Leng

th o

f tim

e th

at b

are

ice

is e

xpos

ed to

mel

twat

er (y

ears

)0 0.5 1 1.5 2Rate of ice discharge to ocean (km3 yr-1 km-1)

(fraction of icedischarge in

this area)

Glacier name

b

Rink Glacier

Store Glacier

Jakobshavn Glacier

Isunnguata Glacier

bare land

ocean

Greenland Ice Sheet

-375 -275 -175 -75Easting (km)

10 m/yr

100 m/yr

1 km/yr

10 km/yr

Ice

velo

city

(NA

SA M

EaSU

REs

)

Figure 1 Figure 2 Figure 3

Name: Kristin Poinar, Cryospheric Sciences, NASA GSFCE-mail: [email protected]: 301-614-7041

References: Poinar, K., I. Joughin, J. Lenaerts, and M. van den Broeke (2016), Englacial latent-heat transfer has limited influence on seaward ice flux in western Greenland, Journal of Glaciology, 62(235), doi:10.1017/jog.2016.103.

Data Sources: NASA Making Earth System Data Records for Use in Research Environments (MEaSUREs) program provided the ice-sheet surface velocity product that formed the basis of the results presented in these figures.

Technical Description of Figures: These images show the measured ice velocity in Greenland and a quantity that we derive from these measurements: the amount of time that bare ice is exposed to meltwater. In general, the longer the ice is in contact with meltwater, the greater amount of refreezing and consequent flow speedup can occur.

Figure 1: The NASA MEaSUREs data product for ice velocity at the surface of the Greenland Ice Sheet.

Figure 2: Our study area in western Greenland has large areas of ice exposed to meltwater for long time periods (yellow and red) and is thus susceptible to refreezing and speedup. However, even with the refreezing-induced speedup, this ice contributes minimally to ice discharge (blue curve on left axis) and sea-level rise.

Figure 3: Over the entire Greenland Ice Sheet, slow-moving areas (yellow and red) such as we see in western Greenland are rare; fast-moving glaciers (purple) that do not experience refreezing are more common. Thus, our conclusion that refreezing only minimally affects ice discharge also holds over all of Greenland.

Scientific significance, societal relevance, and relationships to future missions: The most recent IPCC report (2013) identified the refreezing of meltwater (“cryo-hydrologic warming”) as a source of uncertainty in the sea-level contribution of the Greenland Ice Sheet. Here we quantify, for the first time, the relative contribution of this process to Greenland ice dynamics: ~63% of the ice discharged across our study area in western Greenland does not experience appreciable cryo-hydrologic warming. This directly informs NASA Earth Science’s goal of understanding and forecasting patterns of sea-level rise from the earth’s ice sheets (2007 Decadal Survey, Executive Summary).


Landslides and Precipitation for Hurricane Matthew: Status from 10-4-2016Dalia Kirschbaum1, Thomas Stanley1,2

1Hydrological Sciences, NASA GSFC; 2USRA

The Global Precipitation Measurement (GPM) Mission provides timely observations of rainfallaccumulation that are now being used together with susceptibility information to generate“nowcasts” of potential landslide activity in near real-time. This information is available every 30minutes to 1 day to improve situational awareness of potential impacts from extreme events suchas Hurricane Matthew.


Figure 1

Name: Dalia Kirschbaum, Hydrological Sciences Laboratory, NASA GSFCE-mail: [email protected]: 301-614-5810

References:Stanley, T., and D. B. Kirschbaum, 2016. A heuristic approach to global landslide susceptibility mapping. Nat. Hazards, in review.Kirschbaum, D., T. Stanley, and S. Yatheendradas, 2016: Modeling landslide susceptibility over large regions with fuzzy overlay. Landslides, 13, 485–496,

doi:10.1007/s10346-015-0577-2. http://dx.doi.org/10.1007/s10346-015-0577-2.Huffman, G. J., D. T. Bolvin, D. Braithwaite, K. Hsu, R. J. Joyce, and P. Xie, 2015: Algorithm Theoretical Basis Document (ATBD) for NASA Global

Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievals for GPM (IMERG). 30 pp. http://pmm.nasa.gov/sites/default/files/document_files/IMERG_ATBD_V4.5.pdf.

Kirschbaum, D., T. Stanley, and Y. Zhou, 2015: Spatial and temporal analysis of a global landslide catalog. Geomorphology, 249, 4–15, doi:10.1016/j.geomorph.2015.03.016. http://linkinghub.elsevier.com/retrieve/pii/S0169555X15001579.

Data Sources: GPM Integrated Multi-satellite Retrievals for GPM (IMERG; https://pmm.nasa.gov/data-access/downloads/gpm), a Global Landslide Hazard for Situational Awareness (LHASA; http://ojo-bot.herokuapp.com/opensearch/classic) that uses IMERG and a Global Landslide Susceptibility Map (derived using SRTM, Landsat, Global Faults and Geology Datasets, and Roads Data)

Technical Description of Figures:

Figure 1: The graphic shows several near real-time datasets that are available to improve situational awareness during extreme rainfall events such as those that occurred during Hurricane Matthew in October, 2016. GPM IMERG data shows rainfall from 9/29-10/4. Reported landslides from 10/3 and 10/4 derived from media reports are shown in green stars. These reports likely underestimate the extent of actual landslide activity, particularly in less populated, inland locations. Lastly, a landslide “nowcast” for the area from a NASA Global Landslide Hazard for Situational Awareness (LHASA) Model is shown in red and yellow. The LHASA Nowcasts highlight areas with high landslide susceptibility and high rainfall totals that have potential to experience landslide activity. The model is currently updated daily and provided in vector or raster format. The IMERG data is available every 30 minutes with a latency of 4 hours, the landslide nowcasts are provided each day considering the previous days rainfall. All of the data are accessible via an Applications Programming Interface (API) at https://pmmpublisher.pps.eosdis.nasa.gov. The Hurricane Matthew track is from the NOAA Best Tracks dataset (http://www.nhc.noaa.gov/gis/). Updates were provided routinely for this disaster event and were coordinated with the NASA Disaster Response Program and other emergency response agencies.

Scientific significance, societal relevance, and relationships to future missions: This work demonstrates how data and products available in near real-time, such as the GPM IMERG data, is critical for rapidly providing information during an evolving disaster. The data was used by FEMA and others to assess the potential impact immediately after the event. This information is one piece of the puzzle to better understand and provide relevant, actionable information to decision makers and emergency responders. Future missions will contribute to better understanding the magnitude and impact of these events, including CYGNSS and others.


Near Real Time MODIS observations improve impact assessments for flood events reported byinternational agencies in the flood-prone Lower Mekong region of Southeast Asia.

Detecting Recent Southeast Asia Flooding in Near Real Time from MODISAakash Ahamed1,2 and John D. Bolten1

1Hydrological Sciences, NASA GSFC; 2USRA


Figure 1: MODIS-NDVI Composite (10/27/2016 – 11/06/2016)

Provinces Experiencing Flooding (Red Cross)

Real Time Surface Water (MODIS; 250m)

Thailand

Vietnam

Cambodia

Laos

Figure 2Figure 3

Figure 4 Figure 5

Name: John Bolten, Hydrological Sciences, NASA GSFCE-mail: [email protected]: 301-614-6529

References:Ahamed, A., Bolten, J.D. (In Prep). Remote Sensing Information Systems for Real-Time and Historic Flood Monitoring in Southeast Asia.

Ahamed, A., Bolten, J.D., Doyle, C.S., Fayne, J., (2017). Near Real Time Flood Monitoring and Impact Assessment Systems. In Remote Sensing ofHydrological Extremes.

Project website: http://projectmekongnasa.appspot.com

Data Sources:Earth observation datasets – MODIS Surface Reflectance MOD/MYD09GA and MOD/MYD35 for Near Real-Time product; MODIS Permanent WaterBodies (MOD44W) with MOD/MYD09 Q1 and A1 surface reflectance to train surface water classifiers.

Technical Description of Figures:Figure 1 (center): False Color NDVI composite computed from MODIS surface reflectance observations made in Southeast Asia between 10/02/2016 –11/06/2016. Red polygons indicate provinces in Thailand (left) and Vietnam (right) experiencing flooding, reported by the International Federation of RedCross and Red Crescent Societies (http://www.ifrc.org/). Flooding in Vietnam was reported on 10/30/16 – 11/01/16; flooding in Thailand was reported on10/05/2016.

Figures 2 - 5 (outer): High resolution 250m surface water extent derived from near real-time LANCE – MODIS imagery for the days on which floodingwas reported. Surface water is classified using the historic NDVI signatures (MOD09Q1) of permanent water bodies (MOD 44W). Each image is ascreenshot of the operational flood monitoring system taken on the day of the reported flood. The images show (clockwise from top-left) Lampang,Thailand on 10/5; Ha Tinh, Vietnam on 11/1; Quan Tri, Vietnam on 11/1; and Chainat, Thailand on 10/5.

Scientific significance, societal relevance, and relationships to future missions:Flood disaster events in Southeast Asia result in significant loss of life and economic damage. International agencies and governments typically reportflood disasters at the province or district level, often after the onset of an event. These reports traditionally rely on correspondence with local municipalities,census data, information from news and media agencies, and back of the envelope calculations. Remote sensing information systems designed to monitorfloods in near real-time can significantly improve the spatial resolution and accuracy of information procured by international agencies like the Red Cross,and serve as decision support tools to formulate effective response to sudden onset events. An operational near real-time monitoring system (available athttp://projectmekongnasa.appspot.com) and supporting software tools automatically assess flood impacts to population and infrastructure to provide arapid first set of impact numbers generated hours after the onset of an event. MODIS-derived surface water extent products (e.g. Figures 2-6) exhibit goodagreement (80-90%) when compared to high resolution (22m – 150m) radar data (TerraSAR-X, Envisat ASAR, Disaster Monitoring Constellation), duringboth flood and non-flood conditions. These methods can help guide radar satellite tasking in persistently cloudy or highly damaged areas, and may beextended to other sensors (e.g. VIIRS, Landsat), as well as future missions.


Cloud motion in the GOCI/COMS ocean color dataWayne D. Robinson1,2, Bryan Franz1 and Antonio Mannino1, and Jae-Hyun Ahn3

1Ocean Ecology, NASA GSFC; 2SAIC, Inc.; 3Korea Ocean Satellite Center

Due to the long time needed to acquire all bands for each portion of a GOCI scene (about 50seconds), wind-driven clouds can be displaced by 1 or more pixels from the first to the lastband. This feature, 1: can be used to determine wind at cloud height, and 2: causes degradedocean color and chlorophyll-a retrievals in regions with fast-moving clouds. Cloud motionshould be factored into the design and/or processing of data from geosynchronous ocean colorinstruments.

Figure 1

37∫ N

130∫E

A

B

C

00 400 800km

Figure 2 Figure 3

Zoom-In of region ‘A’’


Name: Wayne Robinson, Ocean Ecology, NASA GSFC and SAIC Inc.E-mail: [email protected]: 301-286-3883

References:W.D. Robinson, B.A. Franz, A. Mannino, and J.-H.Ahn, Cloud motion in the GOCI/COMS ocean color data, International Journal of Remote Sensing, 37:20, 4948-4963, DOI: 10.1080/01431161.2016.1225177.

Data Sources: True color imagery from the GOCI geostationary multispectral instrument aboard the Korean COMS satellite..


Figure 1: Top: A large portion of the GOCI true color scene of the Korean peninsula and surrounding waters. Point ‘A’ is near a fast moving cirrus cloud moving off the coast to the East. The bottom image is a close-up of the cloud feature made with bands that show the full effect of the motion – the green, 490 nm band was taken about 52 seconds later than the red, 660 nm band. The resulting image shows a mis-registration of the clouds due to their motion (note that wind speed was determined to be about 62 m s-1).

Figure 2: A field of wind speeds derived for the scene in Figure 1.

Figure 3: A cross-section through a thin cloud in a GOCI scene (left) and a MODIS Aqua co-incident scene (right). The mis-alignment of the cloud signal in the GOCI bands (top plots) results in wildly varying water-leaving radiance and chlorophyll-a retrievals (bottom plots) while for MODIS Aqua (right), the cloud signal lines up in total radiance. The cloud signal is treated as additional aerosol and is removed, resulting in good water-leaving radiance and chlorophyll-a retrievals.

Scientific significance, societal relevance, and relationships to future missions: The ability of GOCI to detect cloud motion and thus, cloud winds is a useful piece of meteorological information. The ability to measure the motion in time periods of 50 seconds (and possibly down to 8 seconds) permits better cloud motion measurements and allows examining motion on these shorter time scales. Cloud motion between bands unfortunately has the effect of degrading retrievals of water-leaving radiance and chlorophyll-a, and thus results in fewer high quality retrievals. Several of the geostationary ocean color instruments planned to be built in the future could use the same instrument design as GOCI and thus, have the same issue. Although this design has several advantages, it also results in poorer retrievals in the presence of even small amounts of moving clouds. This effect should be considered in instrument design and/or data processing.


Improving Landsat-8 Aquatic Science Products Nima Pahlevan1,4, John R. Schott5, Bryan Franz2, Sean Bailey2, and Brian Markham3

1Terrestrial Information Systems, NASA GSFC (Code 619/616/618), 2Ocean Ecology, NASA GSFC, 3Biospheric Sciences, NASA GSFC, 4Science Systems and Applications, Inc., 5Rochester Institute of Technology

The remote-sensing reflectance (Rrs) products derived from Landsat-8 before (a) and after (b)across-track non-uniformity improvements. The Landsat-derived Rrs products can be directlyrelated to the color of near-surface waters in urban/rural and coastal/inland areas threatened byhuman/climate-change impacts.


Figure 1

Name: Nima Pahlevan, Terrestrial Information Systems Lab, NASA GSFC / SSAIE-mail: [email protected]: 301-614-6684

References:Pahlevan, N., Z. Lee, J. Wei, C. Schaff, J. Schott, and A. Berk, On-orbit radiometric characterization of OLI (Landsat-8) for applications in aquatic remote sensing. Remote Sensing of Environment, 2014. 154: p. 272–284. http://dx.doi.org/10.1016/j.rse.2014.08.001Franz, B.A., S.W. Bailey, N. Kuring, and P.J. Werdell, Ocean color measurements with the Operational Land Imager on Landsat-8: implementation and evaluation in SeaDAS. Journal of Applied Remote Sensing, 2015. 9(1): p. 096070-096070. 10.1117/1.JRS.9.096070Pahlevan, N., Schott, J. R., Franz, B., Zibordi G., Markham, B., Bailey, S., Schaaf, C., Ondrusek, M., Greb, S., Strait, C.. To be published in Remote Sensing of Environment, 201.

Data Sources: Landsat-8, MODIS, VIIRS products made available though USGS and NASA GSFC/OBPG. In-situ data were provided through the AERONET-OC program and by collaborators, including Steven Greb, Chris Strait, and Mike Ondrusek.


Figure 1: The across-track non-uniformity of Landsat-derived aquatic science products are minimized using a combination of reference ocean color measurements made at the Marine Optical Buoy (MOBY) site and cross-calibrations with MODIS/VIIRS observations. The change in the magnitude of Rrs comes from absolute calibration using the MOBY data whereas MODIS/VIIRS data were utilized for minimizing non-uniformity.

Scientific significance, societal relevance, and relationships to future missions: High-quality, high-resolution Rrs products from Landsat are critical in providing precise information regarding the distribution of algal blooms and increases in volume of suspended sediments induced by episodic events and/or dredging activities in coastal/inland waters.


reduced uncertainty of 30m north american boreal …...reduced uncertainty of 30m north american...

Documents