Integrating multiple scales of remote sensing measurement

Abstract

I Introduction

Analysing terrestrial and ocean responses to environmental change and quantifying the drivers of Earth system dynamics and their feedbacks remains a critical priority in the disciplines of geography and environmental science (e.g. Landschützer et al., 2015; Schimel et al., 2015). Remote sensing has long been recognised as having the potential to provide a rich source of information on the state of the environment at a variety of spatial and temporal resolutions and extents and for monitoring regions of the world where system understanding and field observations are limited or missing.

Progress in Physical Geography has a long history of publishing work describing the application of remote sensing to geographical, spatial, environmental and ecological research. Indeed, Professor Paul Mather had a longstanding involvement in the journal as an editor, overseeing much of the spatial and remote sensing research published therein. One of the earliest articles published in the journal (Allan, 1978) summarised the key benefits of remote sensing data to the geographer, alongside the challenges facing the remote sensing scientist. More recently in 2009 a special issue on remote sensing in the twenty-first century (issue 33(4); see Boyd, 2009) captured the state-of-the-art in the discipline and highlighted critical areas where future remote sensing developments would benefit key science areas. The articles contained in this special issue build on the long history of remote sensing manuscripts in Progress in Physical Geography and provide a small sample of the rapidly growing field of remote sensing research as relevant to fields within physical geography. The issue contains articles spanning research areas of terrestrial, marine, atmospheric and cryospheric science and includes articles across the scales of remote sensing observation – from satellite observations to proximal (near-ground) approaches. In this editorial I want to provide an introduction to these diverse manuscripts, whilst also offering a forward look showing how I think remote sensing will support environmental science and geographical research in the future.

II The age of operational remote sensing

I want to begin by referring back to Allan’s (1978) manuscript in which he raised a key point: that of remote sensing as a specialist discipline which needs to “advise on such aspects as the applicability of a system in terms of actual or anticipated object signatures” and to “determine the relevance [of remote sensing systems] to specific problems of interpretation”. I believe that there is substantial evidence in contemporary remote sensing work that points to the operational benefits of remote sensing in practical settings, e.g. agriculture (Atzberger, 2013), aquaculture (Saitoh et al., 2011), forestry (Croft et al., 2014; Zhang et al., 2014) and in ecological studies (Miller et al., 2015). I am delighted that this special issue includes a contribution from Brown et al. (based at the UK’s Environment Agency) who describe the operational use of remote sensing data from both aircraft and satellite platforms to address the need for near-real-time data in operational flood mapping and management. In the UK, flood management, prediction and attribution to climatic change is emerging as a key policy priority area for the UK government (Schaller et al., 2016) so I am certain that Brown et al.’s operational approach to flood mapping using rapidly acquired remote sensing data will have broad impact.

III A new model for remote sensing from space

Next I want to look back upon the last ‘remote sensing special issue’ published in Progress in Physical Geography (issue 33(4)) and reflect on developments in the field since then. I will begin by considering advances in space-borne remote sensing systems. Three years ago on 11 February 2013, NASA launched the latest Landsat satellite from Vandenberg air force base in California (Landsat 8, carrying the operational land imager, OLI). Soon after launch, Landsat 8 began delivering freely available operational data, making this the longest continuous mission for land surface remote sensing with a legacy dating back to 1972. In Europe there have been great scientific developments, with the European Space Agency (ESA) recently launching the first satellites in the Copernicus constellation aimed at delivering global monitoring data for a broad range of applications. Sentinel 1, a mission that will eventually comprise two identical platforms equipped with C-band synthetic aperture RADAR for sea ice and topographic monitoring, now has the first satellite (1a) in orbit. The Sentinel 2 mission (for land applications and vegetation monitoring) will also feature identical satellites (in the same orbit 180° apart), and is underway with Sentinel 2a launched and 2b due for launch later in 2016. Sentinel 3, ESA’s ocean mission (also capable of delivering products for land and atmospheric science) has just launched its first satellite (3a) from the Russian Cosmodrome of Plesetsk. These new ESA missions show a clear change in strategy for space-borne remote sensing, with dual platforms to improve data provision and temporal revisit capabilities. Furthermore, the parallel complementary missions offer improved opportunities for data inter-comparison, cross-calibration and interdisciplinary science. Importantly ESA have adopted a free, full and open data policy for all satellite missions within Copernicus, which is great news for science and operational remote sensing globally.

In her editorial for the last remote sensing issue, Boyd (2009) suggested that there would be an increase in the “number and diversity” of remote sensing systems “particularly as a result of growth in commercial outfits and additional nations moving into the Earth observation sector”. Seven years on, it is certainly true that the number and diversity of systems has increased. Let me focus first on the space sector, where the strategy of constellation based remote sensing (beyond the ESA Sentinels) has a new, commercial focus. Planet Labs, a new satellite start-up company emerging from Silicon Valley is using ‘swarms’ of small, light cubesat satellites to generate daily data with global coverage at fine spatial resolution (Butler, 2014; Hand, 2015a, 2015b). Speaking in a TED talk on the philosophy of Planet Labs’ approach, Will Marshall (CEO and co-founder) said, “what we want are images of the whole planet every day” and to do this, “we need lots of satellites” because the large and expensive “one satellite, one rocket” model “is not scalable” (Marshall, 2014). Planet Labs’ ‘doves’ satellites are compact allowing multiple platforms to be launched at the same time for a fraction of the cost of larger single missions. The first ‘flock’ of 28 dove satellites (3U cubesats featuring commercial off-the-shelf components, and with a mass of ∼5 kg and a size of 10 cm × 10 cm × 34 cm) are now in orbit, comprising the largest constellation of Earth orbiting satellites in human history. These have been placed into orbit from the international space station (ISS), and are delivered via regular rocket supply missions to the ISS. Planet Labs are not alone and other similar companies exist, for example Skybox Imaging who plan a 24-strong ‘swarm’ of cubesats in space (Butler, 2014). The small sensors on board these lightweight cubesats typically have basic capabilities (visible and infra-red bands) but their ability to capture daily data at fine spatial resolution across a global extent is unprecedented in remote sensing history. This approach is designed to transcend the age-old challenge of remote sensing from satellites. With these systems, we can potentially have our cake (fine spatial resolution) and eat it (daily data)! The major benefit of using the International Space Station (ISS) as a launch platform is the regularity of visits from cargo resupply vehicles. Another benefit is offered by the ISS astronauts who can perform quality checks on the hardware to ensure the miniature satellites are not damaged before their deployment into orbit (Niles, 2014).

IV A new democratic age for satellite remote sensing?

All of the above leads me to ponder whether we are entering, or perhaps have already emerged into, a new democratic age in space, which will change how we carry out remote sensing in the future. With lightweight technologies and clever ‘piggybacking’ on other missions, there are a now new ways of reaching the critical orbital zone around the Earth from where a synoptic view and critical measurements of the Earth system can be captured. Access to this orbital space is no longer restricted to agencies and organisations with million dollar budgets and the capability of launching large, heavy satellites. As a remote sensing scientist I see this as providing a revolutionary new capacity for distributed, global sensing in the future. Of course, in space, this approach does not come without its risks. Larger numbers of smaller satellites exacerbate the ‘space debris’ problem which poses threats to long term monitoring from orbital positions, but these concerns are beginning to be addressed, for example by cubesats that can de-orbit themselves using balloons (Lücking et al., 2011; Niles, 2014). A secondary challenge comes with the high data volumes and how data are stored and processed efficiently. This was also a challenge highlighted by Allan (1978) and remains a key question. Can computer processing now keep pace with data volumes? Finally, we must also not lose sight of the importance of remote sensing data validation and sensor calibration with these new cube-sat missions. I cannot miss my opportunity to raise two critical questions – firstly, how is long-term data quality ensured, and second, how reliable are radiometric comparisons between different sensors in the swarm? Knowing the answers to these questions is crucial if data from the missions are to be useful for supporting scientific investigations.

Beyond this new, and perhaps more open model for satellite remote sensing, Boyd’s (2009) suggestion that there would also be an increase in the international remote sensing effort has now come to fruition. One example is the Chinese National Space Administration (CNSA) Gaofen (meaning ‘high definition’) mission. Gaofen-1 was launched in 2013 and was first in a series of high-resolution optical Earth observation satellites. Recently Gaofen-4 was launched into a high geosynchronous orbit, measuring an area over China of approximately 7000 × 7000 km² and with a spatial resolution of 50 m in the visible region and 400 m in the near infra-red. The new model for cubesat releases from the ISS has also paved the way for a more international effort in space – recent launches from the ISS have included cubesat missions from Lithuania and Japan (Niles, 2014) and with crowd funding opportunities it looks likely that the diversity of nations able to be involved in space-based missions will increase in number and scope in future years.

V Ice, vegetation, atmosphere and ocean

Let me now focus on the collection of papers in this issue that provide reviews of the state-of-the art in global monitoring from satellite remote sensing instruments. Sam et al.’s paper shows, for the first time, the utility of time-series data from Landsat 8’s OLI for cryospheric research, and they provide a detailed assessment of how these freely available data can be used for flow velocity estimation in high altitude Indian Himalayan glaciers. The science problem tackled by this research is a key example of where remote sensing has a critical role to play in improving scientific understanding of sensitive global systems. The high altitude cryospheric systems studied by Sam et al. are highly sensitive to climatic change and their remoteness means that operational, regular monitoring is very challenging to achieve with traditional field based techniques (Gioli et al., 2013). Sam et al. highlight that the latest Landsat products from the OLI have not really been explored for their potential utility in glacier surface velocity estimation and yet they crucially show that their satellite-derived flow models show good correspondence with field data. Their results from two glacial forms – an ice glacier and a debris-covered glacier show that the Landsat panchromatic band was best used over the high albedo ice, whilst a short-wave infra-red band performed best where the glacier had a covering of debris.

Turning to terrestrial ecosystems, Dash and Ogutu provide us with an up-to-date review of the latest progress in space-borne optical remote sensing systems for monitoring change and dynamics in vegetated systems. As I have done in this editorial, they reference developments within ESA, NASA and in the new generation of nano-satellites and explain how these will revolutionise operational terrestrial monitoring in the future. They also highlight some new and exciting products coming online – for example, a new global land cover product provided at a finer spatial resolution (30 m) than has ever been available before now. This ‘GlobeLand30’ product, generated by the National Geomatics Centre of China (NGCC) using data from various satellite systems (Landsat 5 and 7 and the China Environmental Disaster Alleviation Satellite), is argued by Dash and Ogutu to provide the opportunity to improve the accurate characterisation of land cover types and their temporal and spatial dynamics. Critically, Dash and Oguto also identify three key challenges, specifically in quality control of data and resulting products, data affordability and continuity of data acquisition.

Moving to the oceans, Shutler et al. provide a comprehensive review of remote sensing for monitoring ocean systems and their interface with the atmosphere and cryosphere. They focus on critical development areas for oceanic and atmospheric science, such as sea-surface atmosphere gas fluxes and point to some interesting future avenues for research with satellite remote sensing data, in particular estimating ocean–atmosphere fluxes of critical gases such as dimethyl sulphide. They suggest that many of the research priorities in the atmosphere and ocean realm will benefit from the long-term capability of the new ESA Sentinel missions and from the distributed capacity offered by new swarms of cubesats. Interestingly, the ISS appears again as a platform for remote sensing in itself – Shutler et al. highlight how it is a platform for several remote sensing sensors which show great promise for ocean and atmospheric science, i.e. the hyperspectral imager for the coastal oceans (HICO), ISS-RapidSCAT (a scatterometer for climate research, weather predictions and hurricane monitoring) and, in the future, the cloud-aerosol transport system (CATS) which is planned for deployment on the ISS.

VI Working with remote sensing data at different grains and extents

So far, I have focused on the state-of-the-art broad-extent global monitoring from remote sensing missions in space, but in parallel there have been recent developments in proximal sensing closer to Earth. The field of proximal sensing is changing – a variety of new work is using sensors fitted to drones and kites to improve fine-grained understanding of environmental processes (Bryson et al., 2013; Burkart et al., 2014; Colomina and Molina, 2014; Dandois and Ellis, 2013; Floreano and Wood, 2015; Murray et al., 2013). Alongside, there are exciting new developments in terrestrial laser scanning and waveform LiDAR (Anderson et al., 2015; Calders et al., 2015; Danson et al., 2014) and in airborne and field spectroscopy (Meroni et al., 2009; Rascher et al., 2015). The grain size of data produced by these proximal approaches is often very different from those captured by satellite instruments but these two models for remote sensing measurement are in fact, highly complementary.

One area where the integration of remote sensing data from satellite and proximal methods can address critical science priorities is in the terrestrial system. Here, critical ‘tipping elements’ (Lenton et al., 2008) exist. These are places where “a small change in forcing triggers a strongly nonlinear response in the internal dynamics of part of the climate system” (Lenton et al., 2008) and include high latitude areas with permafrost cover, for example. Schimel et al. (2015), looking at these tipping element areas, have commented how “important gaps exist in our observations of the terrestrial carbon cycle, resulting from sparse and biased sampling of high flux and high storage regions”. This is because current networks of in situ terrestrial flux observations (e.g. FLUXNET; Baldocchi et al., 2001) are concentrated across Europe and North America, and are conversely sparse within the ‘tipping element’ regions. Schimel et al. (2015) argue that satellite observations will be critical in providing quantitative ecosystem information to fill these gaps, and such data will undoubtedly come from the systems already discussed and also from more specialised systems such as ESA Flex (a planned vegetation fluorescence mission (European Space Agency, 2015)) or from NASA’s Global Ecosystem Dynamics Investigation (GEDI) LiDAR (NASA, 2014). But satellite data are not the only critical piece of the puzzle - other work has shown that remote sensing estimates of important land surface variables such as biomass, require more robust and comprehensive uncertainty estimates from in situ sampling across diverse ecosystems (Hill et al., 2013). On this same theme, and in this special issue, the paper by Li and Guo reviews the range of multi-sensor approaches that can be used for mapping non-photosynthetic vegetation (NPV) – this is the part of the canopy that does not photosynthesise but controls uptake of carbon, water, and nutrients, and the frequency and intensity of natural fire. In this review, the authors describe a wide range of remote sensing approaches at varying scales, from satellite observations to airborne LiDAR and hyperspectral measurements and they conclude that the major challenge lies in “developing integrated models to combine the advantages of multiple sensors while minimising the disadvantages of each data source”. Returning to the work of Schimel et al. (2015), it is clear that a key priority for the future success of remote sensing in addressing questions of global change lies critically in the “co-ordination of in situ and remote observations”. I think that the in situ component to which they refer should explicitly include proximal remote sensing observations describing fine-grained spatial patterns in landscape form and function. As stated earlier, proximal sensing approaches are a key (and growing) source of fine-grained spatial, and increasingly, volumetric information (e.g. in vegetation canopy structural modelling; Dandois and Ellis, 2010; Zahawi et al., 2015). Imagining the ways in which such data can be used creatively to address the issues so far highlighted would represent a creative leap forward for remote sensing as a discipline. I shall return to this theme at the end of my editorial.

VII The dawn of a new proximal sensing era

As I have suggested above, the past five years have seen a transformation in the way that proximal remote sensing data are captured. At this juncture I want to give particular focus to the now complementary fields of photogrammetry and computer vision, which have combined recently into one of the de facto tools for fine-grained modelling from proximal remote sensing data. This approach is commonly referred to as Structure from Motion (SfM) photogrammetry (e.g. Westoby et al., 2012) and it allows the conversion of basic spatial two dimensional image data into sophisticated spatial models in three dimensions. Combined with the miniaturisation of sensors, modern digital imaging capabilities (including cameras on ubiquitous mobile phone and tablet devices), enhancements in computing technology and GPS, these approaches allow some of the earliest techniques in remote sensing to be revisited, revived and re-born. Duffy et al.’s paper in this issue revisits a classic paper from the turn of the last century (Dines, 1903) where kites were first positioned as platforms for scientific data capture. Duffy et al. demonstrate the renaissance of kite-based aerial imaging as part of the modern physical geographer’s remote sensing toolkit (e.g. Lorenz and Scheidt, 2014), uniquely supported (and probably brought about) by the advent of SfM image processing tools. An exciting prospect for me is that with a simple kite and imaging device (e.g. a simple digital camera with an intervalometer) it is now possible to build centimetric resolution orthomosaics and even more impressively, point clouds and surface models.

Also in this issue, Smith et al. provide a comprehensive review of SfM approaches within physical geography and they chart the chronological developments in both SfM and multi-view stereo (MVS) approaches, showing how these approaches have “revolutionised three-dimensional topographic surveys”. Critically they show how the basic data requirements for SfM/MVS modelling approaches can be gathered from a wide range of ‘democratic’ platforms including drones, kites, microlights, balloons, blimps, and from photographs collected on the ground from varying positions around a feature. Most importantly, they explain how the basic data required can be captured from simple digital cameras and they refer to other work that shows how simple low-cost cameras (including those on smartphones) and more expensive digital SLRs deliver models with similar point cloud quality at short ranges. They highlight the importance of model validation as a key future step and state, “the lack of a consistent validation methodology or a systematic campaign to validate SfM-MVS-derived models inhibits precise determination of expected errors”. Furthermore, they discuss that the flexibility with which SfM can be applied means that the spatial accuracy of resultant products and isolation of individual error sources is likely to prove “extremely challenging”. There is certainly work to be done, for example, in resolving issues around SfM accuracy in landscapes with varied and complex terrain. That said, I believe that these approaches offer a truly revolutionary approach for land surface monitoring, whether that be for vegetation remote sensing (Dandois and Ellis, 2013; Puttock et al., 2015), in fluvial environments (Javemick et al., 2014; Woodget et al. 2015) or for monitoring evolution of geomorphic features (Eltner et al., 2015; Lucieer et al., 2014) and believe we are on the brink of a huge expansion in the use of this technique in future years. One only has to look online at the range of free tools with which to explore SfM approaches (e.g. 123Dcatch http://www.123dapp.com/catch) to see this happening in real-time.

VIII Conclusion

To conclude I want to finish by returning to the importance of improving the co-ordination and integration of fine-scale proximal remote sensing data with satellite observations. I believe that the emerging suite of new proximal spatial modelling approaches may prove exceptionally useful in tackling the broader issues raised by Schimel et al. (2015) by virtue of the now global distribution of basic camera systems. The ability to crowd-source the analysis of citizen-gathered photographs from drones, kites or on the ground and process them using SfM could produce revolutionary data for validation of models and satellite data. This is an exciting time to be a remote sensing scientist. We must of course, retain a critical focus on the important issues of ensuring high data quality and product validation, but with this in mind I am very excited to see what the future holds and to be a part of the fine-grained data revolution. Let’s go fly a kite!

References

Allan

(1978) Remote sensing in physical geography. Progress in Physical Geography 2(1): 36–54.

Anderson

Hancock

Disney

. (2015) Is waveform worth it? A comparison of LiDAR approaches for vegetation and landscape characterization. Remote Sensing in Ecology and Conservation. Epub ahead of print 1 November 2015. Available at: http://onlinelibrary.wiley.com/doi/10.1002/rse2.8/abstract.

Atzberger

(2013) Advances in remote sensing of agriculture: context description, existing operational monitoring systems and major information needs. Remote Sensing 5(2): 949–981.

Baldocchi

Falge

. (2001) FLUXNET: a new tool to study the temporal and spatial variability of ecosystem-scale carbon dioxide, water vapor, and energy flux densities. Bulletin of the American Meteorological Society 82(11): 2415–2434.

Boyd

(2009) Remote sensing in physical geography: a twenty-first century perspective. Progress in Physical Geography 33(4): 451–456.

Bryson

Johnson-Roberson

Murphy

. (2013) Kite aerial photography for low-cost, ultra-high spatial resolution multi-spectral mapping of intertidal landscapes. PLoS One 8(9): e73550.

Burkart

Cogliati

Schickling

. (2014) A novel UAV-based ultra-light weight spectrometer for field spectroscopy. IEEE Sensors Journal 14(1): 62–67.

Butler

(2014) Many eyes on Earth. Nature 505(7482): 143–144.

Calders

Newnham

Burt

. (2015) Nondestructive estimates of above-ground biomass using terrestrial laser scanning. Methods in Ecology and Evolution 6(2): 198–208.

10.

Colomina

Molina

(2014) Unmanned aerial systems for photogrammetry and remote sensing: a review. ISPRS Journal of Photogrammetry and Remote Sensing 92: 79–97.

11.

Croft

Chen

Noland

(2014) Stand age effects on Boreal forest physiology using a long time-series of satellite data. Forest Ecology and Management 328: 202–208.

12.

Dandois

Ellis

(2010) Remote sensing of vegetation structure using computer vision. Remote Sensing 2(4): 1157–1176.

13.

Dandois

Ellis

(2013) High spatial resolution three-dimensional mapping of vegetation spectral dynamics using computer vision. Remote Sensing of Environment 136: 259–276.

14.

Danson

Gaulton

Armitage

. (2014) Developing a dual-wavelength full-waveform terrestrial laser scanner to characterize forest canopy structure. Agricultural and Forest Meteorology 198: 7–14.

15.

Dines

(1903) Scientific kite flying. Nature 68: 144–145.

16.

Eltner

Baumgart

Maas

. (2015) Multi-temporal UAV data for automatic measurement of rill and interrill erosion on loess soil. Earth Surface Processes and Landforms 40(6): 741–755.

17.

European Space Agency (2015) N° 42–2015: FLEX mission to be next ESA Earth Explorer. Available at: www.esa.int/For_Media/Press_Releases/FLEX_mission_to_be_next_ESA_Earth_Explorer (accessed 7 February 2016).

18.

Floreano

Wood

(2015) Science, technology and the future of small autonomous drones. Nature 521(7553): 460–466.

19.

Gioli

Khan

Scheffran

(2013) Climatic and environmental change in the Karakoram: making sense of community perceptions and adaptation strategies. Regional Environmental Change 14(3): 1151–1162.

20.

Hand

(2015a) Startup liftoff. Science 348(6231): 172–177.

21.

Hand

(2015b) Thinking inside the box. Science 348(6231): 176–177.

22.

Hill

Williams

Bloom

. (2013) Are inventory based and remotely sensed above-ground biomass estimates consistent? PLoS One 8(9): e74170.

23.

Javemick

Brasington

Caruso

(2014) Modeling the topography of shallow braided rivers using Structure-from-Motion photogrammetry. Geomorphology 213: 166–182.

24.

Landschützer

Gruber

Haumann

. (2015) The reinvigoration of the Southern Ocean carbon sink. Science 349(6253): 1221–1224.

25.

Lenton

Held

Kriegler

. (2008) Tipping elements in the Earth’s climate system. Proceedings of the National Academy of Sciences of the United States of America 105(6): 1786–1793.

26.

Lorenz

Scheidt

(2014) Compact and inexpensive kite apparatus for geomorphological field aerial photography, with some remarks on operations. GeoResJ 3 – 4: 1–8.

27.

Lucieer

De Jong

Turner

(2014) Mapping landslide displacements using Structure from Motion (SfM) and image correlation of multi-temporal UAV photography. Progress in Physical Geography 38(1): 97–116.

28.

Lücking

Colombo

McInnes

(2011) A passive de-orbiting strategy for high altitude CubeSat missions using a deployable reflective balloon. In: 8th IAA symposium on small satellites, Berlin, Germany, 4–8 April 2011. Available at: http://strathprints.strath.ac.uk/41233/1/McInnes_CR_Pure_A_passive_de_orbiting_strategy_for_high_altitude_CubeSat_missions_using_a_deployable_reflective_balloon_Mar_2011.pdf%5D.

29.

Marshall

(2014) TED talk: tiny satellites show us the Earth as it changes in near real time. Available at: www.ted.com/speakers/will_marshall (accessed 5 December 2015).

30.

Meroni

Rossini

Guanter

. (2009) Remote sensing of solar induced chlorophyll fluorescence: review of methods and applications. Remote Sensing of Environment 113(10): 2037–2051.

31.

Miller

Scales

Ingram

. (2015) Basking sharks and oceanographic fronts: quantifying associations in the north-east Atlantic. Functional Ecology 29(8): 1099–1109.

32.

Murray

Neal

Labrosse

(2013) Development and deployment of an intelligent kite aerial photography platform (iKAPP) for site surveying and image acquisition. Journal of Field Robotics 30(2): 288–307.

33.

NASA (2014) Global Ecosystem Dynamics Investigation LiDAR (GEDI). Available at: http://science.nasa.gov/missions/gedi/ (accessed 9 February 2016).

34.

Niles

(2014) It’s a march of the cubesats as space station deployment continues. Available at: www.nasa.gov/mission_pages/station/research/news/cubesat_deployment (accessed 9 February 2016).

35.

Puttock

Cunliffe

Anderson

. (2015) Aerial photography collected with a multirotor drone reveals impact of Eurasian beaver reintroduction on ecosystem structure. Journal of Unmanned Vehicle Systems 3(3): 123–130.

36.

Rascher

Alonso

Burkart

. (2015) Sun-induced fluorescence – a new probe of photosynthesis: first maps from the imaging spectrometer HyPlant. Global Change Biology 21(12): 4673–4684.

37.

Saitoh

S-I

Mugo

Radiarta

. (2011) Some operational uses of satellite remote sensing and marine GIS for sustainable fisheries and aquaculture. ICES Journal of Marine Science 68(4): 687–695.

38.

Schaller

Kay

Lamb

. (2016) Human influence on climate in the 2014 southern England winter floods and their impacts. Nature Climate Change. Epub ahead of print 1 February 2016. DOI:10.1038/nclimate2927.

39.

Schimel

Pavlick

Fisher

. (2015) Observing terrestrial ecosystems and the carbon cycle from space. Global Change Biology 21(5): 1762–1776.

40.

Westoby

Brasington

Glasser

. (2012) ‘Structure-from-Motion’ photogrammetry: a low-cost, effective tool for geoscience applications. Geomorphology 179: 300–314.

41.

Woodget

Carbonneau

Visser

. (2015) Quantifying submerged fluvial topography using hyperspatial resolution UAS imagery and structure from motion photogrammetry. Earth Surface Processes and Landforms 40(1): 47–64.

42.

Zahawi

Dandois

Holl

. (2015) Using lightweight unmanned aerial vehicles to monitor tropical forest recovery. Biological Conservation 186: 287–295.

43.

Zhang

Huang

Hogg

. (2014) Estimating spatial variation in Alberta forest biomass from a combination of forest inventory and remote sensing data. Biogeosciences 11(10): 2793–2808.

Integrating multiple scales of remote sensing measurement – from satellites to kites

Abstract

I Introduction

II The age of operational remote sensing

III A new model for remote sensing from space

IV A new democratic age for satellite remote sensing?

V Ice, vegetation, atmosphere and ocean

VI Working with remote sensing data at different grains and extents

VII The dawn of a new proximal sensing era

VIII Conclusion

References