1,721,439 research outputs found
A review of the application of BRDF models to infer land cover parameters at regional and global scales
No full textThis paper presents a review of the application of Bi-directional Reflectance Distribution Function (BRDF) models in the inference of land surface parameters at regional and global scales using remotely sensed data. Information on land surface parameters, such as Leaf Area Index (LAI), fraction of Absorbed Photosynthetically Active Radiation (fAPAR), aerodynamic surface roughness and albedo, are valuable for understanding the transfer of energy and mass between terrestrial ecosystems and the atmosphere (e.g., carbon, nitrogen and methane cycling) and for ingestion into the lower boundary condition of global circulation models (GCM)s. Conventional techniques for acquiring information on land surface parameters do not account for or utilize the directional nature of surface reflectance. This paper reviews empirical, semi-empirical and, to a lesser extent, physical BRDF models that describe the surface BRDF. In each case examples are given of their application in inferring land surface parameters. The review concludes by discussing the future prospects of BRDF modelling using spaceborne sensors. <br/
Scalable importance tempering and Bayesian variable selection
Full text linkWe propose a Monte Carlo algorithm to sample from high-dimensional probability distributions that combines Markov chain Monte Carlo (MCMC) and importance sampling. We provide a careful theoretical analysis, including guarantees on robustness to high-dimensionality, explicit comparison with standard MCMC and illustrations of the potential improvements in efficiency. Simple and concrete intuition is provided for when the novel scheme is expected to outperform standard ones. When applied to Bayesian Variable Selection problems, the novel algorithm is orders of magnitude more efficient than available alternative sampling schemes and allows to perform fast and reliable fully Bayesian inferences with tens of thousand regressors
Multilevel linear models, Gibbs Samplers and multigrid decompositions (with discussion)
Full text linkWe study the convergence properties of the Gibbs Sampler in the context of posterior distributions arising from Bayesian analysis of conditionally Gaussian hierarchical models. We develop a multigrid approach to derive analytic expressions for the convergence rates of the algorithm for various widely used model structures, including nested and crossed random effects. Our results apply to multilevel models with an arbitrary number of layers in the hierarchy, while most previous work was limited to the two-level nested case. The theoretical results provide explicit and easy-to-implement guidelines to optimize practical implementations of the Gibbs Sampler, such as indications on which parametrization to choose (e.g. centred and non-centred), which constraint to impose to guarantee statistical identifiability, and which parameters to monitor in the diagnostic process. Simulations suggest that the results are informative also in the context of non-Gaussian distributions and more general MCMC schemes, such as gradient-based ones
Estimating of rice crop yield in Thailand using satellite data
No full textRice is the world’s major staple food crop occupying over 12% of global cropland area which produces around 800 million tons. Nearly 90% of the world’s rice is produced and consumed in Asian countries. Therefore, information on agricultural plantation area, yield, and production are essential to ensure food security of nearly 3 billion people. At the moment this information is either lacking in many countries or only available post-harvest, this is too late to input into any effecting policy in a specific year. Therefore, there is a pressing need to provide accurate and reliable yield estimation well ahead of harvest. In this project we explore potential of multi source remote sensing data coupled with crop model to provide country scale yield estimation in Thailand. For optical sensor, the study utilised Landsat8 OLI/TIRS satellite data to develop common vegetation indexes (VIs) approach to derive essential crop biophysical variables such as Leaf Area Index. This is supplemented with information from microwave sensor such as Sentinel 1 to overcome issues with cloud. At the end, we produced a regular time series of crop biophysical variable across the growing season. These satellite-based estimates were validated with dedicated field campaign in three provinces covering the entire growing season. Initial results suggest a good agreement between the optical/microwave derived crop biophysical variables and ground data. Finally, these will be used as an input to the ORYZA 2000 crop model to adjust the model parameters and develop a high resolution yield prediction
Global impact of landscape fire emissions on surface level PM2.5 concentrations, air quality exposure and population mortality
No full textAirborne fine particulate matter (PM2.5) represents the greatest ambient air pollution risk to health. Wildfires and managed burns, together referred to hereafter as ‘landscape’ fires, are a significant PM2.5 source in many regions worldwide, able to affect air quality in areas far away from the fires themselves. We use 0.125°, 3 hourly outputs (2016-2019) from the Copernicus Atmospheric Monitoring System (CAMS) to investigate patterns of global population exposure to ambient surface level PM2.5, and specifically to the contribution coming from landscape fires. We calculate both the air quality impacts and annual average mortality at the level of the nation state, and our analysis highlights both the burden of poor air quality and the significance of landscape fire sources in developing nations in particular. We find 143 countries to have an average population weighted (PW) total PM2.5 surface level concentration exceeding the 10 µg.m-3 guideline recommended by WHO, with 67.2 million people annually exposed to PM2.5 levels classed as ‘hazardous’ (> 250.5 µg.m-3) according to the US Environmental Protection Agency (EPA) air quality index (AQI). Landscape fires alone result in 44 million people annually being exposed to air quality considered unhealthy (PM2.5 > 55 µg.m-3), and 4 million to that considered ‘hazardous’ to health (> 250.5 µg.m-3). Populations in central and west Africa and south and south east Asia are most affected by the landscape fire smoke, and eight countries exceed the WHO annual mean 10 µg.m-3 guideline from this source only - with the contribution from fires highest in Laos (61% of the total PM2.5), Democratic Republic of Congo (45%) and Sierra Leone (44%). Combining published dose-response functions with these landscape-fire PM2.5 contributions, we estimate that 677,745 premature deaths annually result from exposure to landscape fire smoke, with almost 39% of these in children under five. This mortality represents between 8 and 21% of the estimated 3.2 to 8.9 million people dying annually from outdoor air pollution exposure, highlighting landscape fires as a significant contributor. Our results indicate that environmental programmes aimed at lessening the use of fire in land clearance and agricultural residue management in developing nations would very likely result in significant co-benefits for health
Evaluation of satellite dust detection algorithms in the Middle East region
No full textIn the last 15 years, the frequency, spatial extent, and intensity of dust storms have increased and it is one of the main continuously occurring environmental hazard in the Middle East region. Since dust storms generally cover a large spatial extent and are highly dynamic, satellite Earth Observation (EO) is a key tool for detecting their occurrence, identifying their origin, and monitoring their transport and state. A variety of satellite dust detection algorithms have been developed to identify dust emissions sources and dust plumes once entrained in the atmosphere. This paper evaluates the performance of five widely applied dust detection algorithms: the Brightness Temperature Difference (BTD), D-parameter, Normalized Difference Dust Index (NDDI), Thermal-Infrared Dust Index (TDI) and the Middle East Dust Index (MEDI). These algorithms are applied to Moderate Resolution Imaging Spectroradiometer (MODIS) data to detect dust-contaminated pixels during three significant dust events in 2007 in the Middle East region that originated from sources in Iraq, Syria and Saudi Arabia. The results indicate that all methods have a comparable performance in detecting dust-contaminated pixels during the three dust events with an average detection rate (between all algorithms) of 85%. However, substantial differences exist in their ability to distinguish dust from clouds and the land surface, which resulted in large errors of commission. Direct validation of these algorithms with observations from seven Aerosol Robotic Network (AERONET) stations in the region found an average false detection rate (between all algorithms) of 89.6%. Although the algorithms performed well in detecting the dust-contaminated pixels their high false detection rate means it is challenging to apply these algorithms in operational context
Classified earth observation data between 1990 and 2015 for the Perth Metropolitan Region, Western Australia using the Import Vector Machine algorithm
No full textThis dataset represents land cover for 7 sequential snapshots (1990, 2000, 2003, 2005, 2007, 2013 and 2015) over the Perth Metropolitan Region, Western Australia (WA) derived from medium resolution Landsat data. Cloud free imagery was acquired in or close to the month of July coinciding with WA's winter months coinciding with peak green-up facilitating the greatest contrast between spectrally similar surfaces (e.g. bare earth and urban). Imagery was first standardised and normalised to remove inherent residual noise (e.g. differences in modelled atmospheric correction parameters) whilst permitting classification of all imagery based upon a single classification model. The model was computed from the 2005 image representing the month post maximum rainfall of all considered imagery associated with peak greenness and maximum spectral separability. Classification of the normalised data was achieved with the Import Vector Machine (IVM) algorithm following a hybrid forward/backward strategy that adds import vectors whilst continuously testing validity in each step, producing a sparse and more accurate classification solution. Classified land cover data is provided in raster format (.tif) and divided into the classes: bare earth (1), grassland (2), low urban albedo (e.g. asphalt (3)), water (4), forest (5) and high urban albedo (e.g. concrete (6)). Please see MacLachlan et al. (2017) for further details.
Supplement to: MacLachlan, A.; Biggs, E.; Roberts, G.; Boruff, B. Urban Growth Dynamics in Perth, Western Australia: Using Applied Remote Sensing for Sustainable Future Planning. Land 2017, 6, 9. doi:10.3390/land6010009
Also available at the pangea data publisher for earth and environmental science. doi: doi.pangaea.de/10.1594/PANGAEA.871017</span
Urbanisation-induced land cover temperature dynamics for sustainable future urban heat island mitigation
No full textUrban land cover is one of the fastest global growing land cover types which permanently alters land surface properties and atmospheric interactions, often initiating an urban heat island effect. Urbanisation comprises a number of land cover changes within metropolitan regions. However, these complexities have been somewhat neglected in temperature analysis studies of the urban heat island effect, whereby over-simplification ignores the heterogeneity of urban surfaces and associated land surface temperature dynamics. Accurate spatial information pertaining to these land cover change – temperature relationships across space is essential for policy integration regarding future sustainable city planning to mitigate urban heat impacts. Through a multi-sensor approach, this research disentangles the complex spatial heterogeneous variations between changes in land cover (Landsat data) and land surface temperature (MODIS data), to understand the urban heat island effect dynamics in greater detail for appropriate policy integration. The application area is the rapidly expanding Perth Metropolitan Region (PMR) in Western Australia (WA). Results indicate that land cover change from forest to urban is associated with the greatest annual daytime and nighttime temperature change of 0.40 °C and 0.88 °C respectively. Conversely, change from grassland to urban minimises temperature change at 0.16 °C and 0.77 °C for annual daytime and nighttime temperature respectively. These findings are important to consider for proposed developments of the city as such detail is not currently considered in the urban growth plans for the PMR. The novel intra-urban research approach presented can be applied to other global metropolitan regions to facilitate future transition towards sustainable cities, whereby urban heat impacts can be better managed through optimised land use planning, moving cities towards alignment with the 2030 sustainable development goals and the City Resilience Framework (CRF)
Estimating biomass consumed from fire using MODIS FRE
No full textBiomass burning is an important global phenomenon impacting atmospheric composition. Application of satellite based measures of fire radiative energy (FRE) has been shown to be effective for estimating biomass consumed, which can then be used to estimate gas and aerosol emissions. However, application of FRE has been limited in both temporal and spatial scale. In this paper we offer a methodology to estimate FRE globally for 2001–2007 at monthly time steps using MODIS. Accuracy assessment shows that our FRE estimates are precise (R2 = 0.85), but may be underestimated. Global estimates of FRE show that Africa and South America dominate biomass burning, accounting for nearly 70% of the annual FRE generated. Applying FRE-based combustion factors to Africa yields an annual average biomass burned of 716–881 Tg of dry matter (DM). Comparison with the GFEDv2 biomass burned estimates shows large annual differences suggesting significant uncertainty remains in emission estimate
- …
