Remote sensing to establish forest dynamics and structure

Remote sensing is an important tool in monitoring the environment and environmental impact of climate change. Many of these changes occur naturally while others are instigated through anthropogenic activity. This ongoing research project uses available remote sensing data to not only monitor change but also to establish relationships between environments and the remote sensing data. Ultimately enriching our understanding of the natural environment through increasing understanding of electromagnetic interactions.

Of particular interest is the terrestrial biosphere and within this, forest environments. Establishing links between forest dynamics and remote sensing data from systems such as Synthetic Aperture Radar (SAR) (e.g. PALSAR 1&2, UAVSAR) and Lidar (e.g. LVIS, GLAS) is crucial to enable the remote detection of forest parameters such as vertical forest structure, areal extent, leaf area index, and aboveground biomass. Successful implementation of theorised and applied remote sensing methods allows for impact on policy decisions, climate change and carbon modelling, deforestation and clear cut regulation among others.

The aim of the project was to establish links between remote sensing data and the structure and dynamics of the terrestrial biosphere to inform climate modelling and international policy. In this project use is made of modelling and empirical data to establish linkages between SAR backscatter interactions and the physical structure of the world’s forests to better map, monitor, and quantify their contribution to the terrestrial carbon cycle.

The project was designed to inform on or determine the origin of identified active remote sensing scattering trends within forest environments through independent and synergistic use of synthetic aperture radar (SAR), lidar systems, modelling, and field study. Trends such as the so called “saturation” effect, Leaf Area Index (LAI) estimation from waveform lidar, and the links between SAR interferometric coherence and the extinction rates of both SAR and lidar energy through the forest canopy are all examined with the aim of improving the determination of vertical and horizontal forest structure remotely.

To date the project has provided impact through international publications and also through presentations at international conferences such as AGU and IGARSS. The project has fostered collaborative partnerships between researchers at the University of Brighton, University of Maryland, NASA Goddard Space Flight Centre, and NASA Jet Propulsion Lab, with collaborative work influencing current (UAVSAR, LVIS, GLAS) and future (NISAR, GEDI, ICESat 2) active remote sensing missions and in particular their modes and methods of data acquisition.

Project findings have highlighted the impact forest structure has upon various active remote sensing methods and their collected data. The greatest use of these results has been to challenge accepted beliefs of the capabilities of the studied sensors and to provide further insight into the relationships between synthetic aperture radar, lidar, and forest biomass both through direct and indirect association.

To date the project has shown how forest parameters related to aboveground biomass, such as vertical canopy structure, can be related to or derived from radar backscatter trends, dynamics, and interactions both through modelling and empirical measurements; how vertical structure can be determined from waveform lidar data, including leaf area index through mapping forests in California and the UK (potentially globally), and of how radar and lidar data can be used in combination to overcome limitations existing with each sensor’s ability in the horizontal and vertical ranges, using data acquired in Quebec, Canada.

The outputs of this project improve understanding of forest interactions with active remote sensing and encourage wider application of the techniques and methods presented in the work for future ecosystem and environmental studies.

Brolly, Matthew, Simard, Marc, Tang, Hao, Dubayah, Ralph, Fisk, Justin A lidar-radar framework to assess the impact of vertical structure on interferometric coherence IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

Michelakis, Dimitrios, Stuart, Neil, Brolly, Matthew, Woodhouse, Iain, Lopez, German and Linares, Vinicio (2015) Estimation of Woody Biomass of Pine Savanna Woodlands From ALOS PALSAR Imagery IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8 (1). pp. 244-254. ISSN 1939-1404

Tang, Hao, Dubayah, Ralph, Brolly, Matthew, Ganguly, Sangram and Zhang, Gong(2014) Large-scale retrieval of leaf area index and vertical foliage profile from the spaceborne waveform lidar (GLAS/ICESat) Remote Sensing of Environment, 154. pp. 8-18. ISSN 0034-4257

Brolly, Matthew and Woodhouse, I.H. (2014) Long Wavelength SAR Backscatter Modelling Trends as a Consequence of the Emergent Properties of Tree Populations Remote Sensing, 6. pp. 7081-1709. ISSN 2072-4292

Tang, Hao, Brolly, Matthew, Zhao, Feng, Strahler, Alan, Schaaf, Crystal, Ganguly, Sangram, Zhang, Gong and Dubayah, Ralph (2014) Deriving and validating Leaf Area Index (LAI) at multiple spatial scales through lidar remote sensing: a case study in Sierra National Forest, CA Remote Sensing of Environment, 143. pp. 131-141. ISSN 0034-4257

Brolly, Matthew and Woodhouse, I.H. (2013) Vertical backscatter profile of forests predicted by a macroecological plant model International Journal of Remote Sensing, 34 (4). pp. 1026-1040. ISSN 0143-1161

Woodhouse, Iain, Mitchard, Edward, Brolly, Matthew, Maniatis, Danae and Ryan, Casey (2012) Radar backscatter is not a ‘direct measure’ of forest biomass Nature Climate Change, 2. pp. 556-557. ISSN 1758-678X

Brolly, Matthew and Woodhouse, I.H. (2012) A “Matchstick Model” of microwave backscatter from a forest Ecological Modelling, 237-238. pp. 74-87. ISSN 0304-3800

Brolly, Matthew, Woodhouse, I.H., Niklas, K.J. and Hammond, S.T. (2012) A macroecological analysis of SERA derived forest heights and implications for forest volume remote sensing PLoS ONE, 7 (3). ISSN 1932-6203