A MODIS direct broadcast algorithm for mapping wildfire burned area in the western United States

Improved wildland fire emission inventory methods are needed to support air quality forecasting and guide the development of air shed management strategies. Air quality forecasting requires dynamic fire emission estimates that are generated in a timely manner to support real-time operations. In the regulatory and planning realm, emission inventories are essential for quantitatively assessing the contribution of wildfire to air pollution. The development of wildland fire emission inventories depends on burned area as a critical input. This study presents a Moderate Resolution Imaging Spectroradiometer (MODIS) - direct broadcast (DB) burned area mapping algorithm designed to support air quality forecasting and emission inventory development. The algorithm combines active fire locations and single satellite scene burn scar detections to provide a rapid yet robust mapping of burned area. Using the U.S. Forest Service Fire Sciences Laboratory (FiSL) MODIS-DB receiving station in Missoula, Montana, the algorithm provided daily measurements of burned area for wildfire events in the western U.S. in 2006 and 2007. We evaluated the algorithm's fire detection rate and burned area mapping using fire perimeter data and burn scar information derived from high resolution satellite imagery. The FiSL MODIS-DB system detected 87% of all reference fires > 4 km², and 93% of all reference fires > 10 km². The burned area was highly correlated (R² = 0.93) with a high resolution imagery reference burn scar dataset, but exhibited a large over estimation of burned area (56%). The reference burn scar dataset was used to calibrate the algorithm response and quantify the uncertainty in the burned area measurement at the fire incident level. An objective, empirical error based approach was employed to quantify the uncertainty of our burned area measurement and provide a metric that is meaningful in context of remotely sensed burned area and emission inventories. The algorithm uncertainty is ± 36% for fires 50 km² in size, improving to ± 31% at a fire size of 100 km². Fires in this size range account for a substantial portion of burned area in the western U.S. (77% of burned area is due to fires > 50 km², and 66% results from fires > 100 km²). The dominance of these large wildfires in burned area, duration, and emissions makes these events a significant concern of air quality forecasters and regulators. With daily coverage at 1-km² spatial resolution, and a quantified measurement uncertainty, the burned area mapping algorithm presented in this paper is well suited for the development of wildfire emission inventories. Furthermore, the algorithm's DB implementation enables time sensitive burned area mapping to support operational air quality forecasting.     

Regionally adaptable dNBR-based algorithm for burned area mapping from MODIS data

Recent advances in instrument design have led to considerable improvements in wildfire mapping at regional and global scales. Global and regional active fire and burned area products are currently available from various satellite sensors. While only global products can provide consistent assessments of fire activity at the global, hemispherical or continental scales, the efficiency of their performance differs in various ecosystems. The available regional products are hard-coded to the specifics of a given ecosystem (e.g. boreal forest) and their mapping accuracy drops dramatically outside the intended area. We present a regionally adaptable semi-automated approach to mapping burned area using Moderate Resolution Imaging Spectroradiometer (MODIS) data. This is a flexible remote sensing/GIS-based algorithm which allows for easy modification of algorithm parameterization to adapt it to the regional specifics of fire occurrence in the biome or region of interest. The algorithm is based on Normalized Burned Ratio differencing (dNBR) and therefore retains the variability of spectral response of the area affected by fire and has the potential to be used beyond binary burned/unburned mapping for the first-order characterization of fire impacts from remotely sensed data. The algorithm inputs the MODIS Surface Reflectance 8-Day Composite product (MOD09A1) and the MODIS Active Fire product (MOD14) and outputs yearly maps of burned area with dNBR values and beginning and ending dates of mapping as the attributive information. Comparison of this product with high resolution burn scar information from Landsat ETM+ imagery and fire perimeter data shows high levels of accuracy in reporting burned area across different ecosystems. We evaluated algorithm performance in boreal forests of Central Siberia, Mediterranean-type ecosystems of California, and sagebrush steppe of the Great Basin region of the US. In each ecosystem the MODIS burned area estimates were within 15% of the estimates produced by the high resolution base with the R² between 0.87 and 0.99. In addition, the spatial accuracy of large burn scars in the boreal forests of Central Siberia was also high with Kappa values ranging between 0.76 and 0.79.     

Prototyping a global algorithm for systematic fire-affected area mapping using MODIS time series data

The remote sensing of Earth surface changes is an active research field aimed at the development of methods and data products needed by scientists, resource managers, and policymakers. Fire is a major cause of surface change and occurs in most vegetation zones across the world. The identification and delineation of fire-affected areas, also known as burned areas or fire scars, may be considered a change detection problem. Remote sensing algorithms developed to map fire-affected areas are difficult to implement reliably over large areas because of variations in both the surface state and those imposed by the sensing system. The availability of robustly calibrated, atmospherically corrected, cloud-screened, geolocated data provided by the latest generation of moderate resolution remote sensing systems allows for major advances in satellite mapping of fire-affected area. This paper describes an algorithm developed to map fire-affected areas at a global scale using Moderate Resolution Imaging Spectroradiometer (MODIS) surface reflectance time series data. The algorithm is developed from the recently published Bi-Directional Reflectance Model-Based Expectation change detection approach and maps at 500 m the location and approximate day of burning. Improvements made to the algorithm for systematic global implementation are presented and the algorithm performance is demonstrated for southern African, Australian, South American, and Boreal fire regimes. The algorithm does not use training data but rather applies a wavelength independent threshold and spectral constraints defined by the noise characteristics of the reflectance data and knowledge of the spectral behavior of burned vegetation and spectrally confusing changes that are not associated with burning. Temporal constraints are applied capitalizing on the spectral persistence of fire-affected areas. Differences between mapped fire-affected areas and cumulative MODIS active fire detections are illustrated and discussed for each fire regime. The results reveal a coherent spatio-temporal mapping of fire-affected area and indicate that the algorithm shows potential for global application.     

Retrieving middle-infrared reflectance for burned area mapping in tropical environments using MODIS

The ephemeral character of the radiative signal together with the presence of aerosols imposes severe limitations on the use of classical approaches, e.g. based on red and near-infrared, to discriminate between burned and unburned surfaces in tropical environments. Surface reflectance in the middle-infrared (MIR) has been used to circumvent these difficulties because the signal is virtually unaffected by the presence of aerosols associated to biomass burning. Retrieval of the MIR reflected component from the total signal is, however, a difficult problem because of the presence of a diversity of radiance sources, namely the surface reflected solar irradiance and the surface emitted radiance that may reach comparable magnitude during daytime. The method proposed by Kaufman and Remer (1994) to retrieve surface MIR reflectance presents the advantage of not requiring auxiliary datasets (e.g. atmospheric profiles) nor major computational means (e.g. for solving radiative transfer models). Nevertheless, the method was specifically designed to retrieve MIR reflectance over dense dark forests in the middle latitudes and, as shown in the present study, severe problems may arise when applying it beyond the range of validity, namely for burned area mapping in tropical environments. The present study consists of an assessment of the performance of the method for a wide range of atmospheric, geometric and surface conditions and of the usefulness of extracted surface reflectances for burned area discrimination. Results show that, in the case of tropical environments, there is a significant decrease in performance of the method for high values of land surface temperature, especially when associated with low sun elevation angles. Burned area discrimination is virtually impaired in such conditions, which are often present when using data from instruments on-board polar orbiters, namely MODIS in Aqua and Terra, to map burned surfaces over the Amazon forest and “cerrado” savanna regions.     

Integration of geostationary FRP and polar-orbiter burned area datasets for an enhanced biomass burning inventory

Earth Observation (EO) sensors play an important role in quantifying biomass burning related fuel consumption and carbon emissions, and capturing their spatial and temporal dynamics. Typically, biomass burning emissions inventories are developed by exploiting either burned area (BA) or active fire (AF) measures of fire radiative energy (FRE). These approaches have both advantages and limitations. For example, methods based on burned area data typically require hard-to-obtain estimates of fuel load and combustion completeness, and the accuracy of the BA algorithm may deteriorate for small fires or those in densely forested terrain. Conversely, ‘raw’ FRE-based methods are typically low-biassed due to the non-detection of low intensity fires, and are also hindered by cloud cover. Here we develop a methodology integrating these two EO data types to deliver a high temporal resolution emissions inventory, maximising the benefit of each data type without requiring additional information. We focus on Africa, the most fire affected continent, and combine daily FRE observations provided by Meteosat Spinning Enhanced Visible and Infrared Imager (SEVIRI) with BA measures delivered by Moderate Resolution Imaging Spectroradiometer (MODIS). For individual fires detected by both types of data, we estimate fuel consumption per unit area (FC_A: g·m^− 2) via the ratio of FRE-derived total fuel consumption (FC_T) to BA. These values are then extrapolated to fires that were mapped using the BA data but which remained undetected in the SEVIRI AF product, thus correcting for the ‘low spatial resolution bias’ inherent in geostationary AF datasets. Calculated daily continental scale FC_T for Africa varies between 0.3 and 20 Tg for the period February 2004-January 2005. We estimate annual continental FC_T to be 1418 Tg, far closer to the 2272 Tg provided by the widely used Global Fire Emissions Database (version 3; GFEDv3) than is obtained when using ‘raw’ FRE data alone. This synergistic approach has substantially narrowed the gap between GFEDv3 and FRE-derived emissions inventories, whilst the geostationary FRP observations offer the advantage that the daily emissions estimates can be distributed more accurately over the diurnal fire cycle if required for linking to atmospheric transport models.     

Evaluating spectral indices for burned area discrimination using MODIS/ASTER (MASTER) airborne simulator data

Wildland fires are an annually recurring phenomenon in many terrestrial ecosystems. Accurate burned area estimates are important for modeling fire-induced trace gas emissions and rehabilitating post-fire landscapes. High spatial and spectral resolution MODIS/ASTER (MASTER) airborne simulator data acquired over three 2007 southern California burns were used to evaluate the sensitivity of different spectral indices at discriminating burned land shortly after a fire. The performance of the indices, which included both traditional and new band combinations, was assessed by means of a separability index that provides an assessment of the effectiveness of a given index at discriminating between burned and unburned land. In the context of burned land applications results demonstrated (i) that the highest sensitivity of the longer short wave infrared (SWIR) spectral region (1.9 to 2.5 μm) was found at the band interval from 2.31 to 2.36 μm, (ii) the high discriminatory power of the mid infrared spectral domain (3 to 5.5 μm) and (iii) the high potential of emissivity data. As a consequence, a newly proposed index which combined near infrared (NIR), longer SWIR and emissivity outperformed all other indices when results were averaged over the three fires. Results were slightly different between land cover types (shrubland vs. forest-woodland). Prior to use in the indices the thermal infrared data were separated into temperature and emissivity to assess the benefits of using both temperature and emissivity. Currently, the only spaceborne sensor that provides moderate spatial resolution (< 100 m) temperature and emissivity data is the Advanced Spaceborne and Thermal Emission Radiometer (ASTER). Therefore, our findings can open new perspectives for the utility of future sensors, such as the Hyperspectral Infrared (HyspIRI) sensor. However, further research is required to evaluate the performance of the newly proposed band combinations in other vegetation types and different fire regimes.     

Assessment of multitemporal compositing techniques of MODIS and AVHRR images for burned land mapping

The performance of several criteria to generate multitemporal composites of daily Moderate Resolution Imaging Spectrometer (MODIS) and Advanced Very High Resolution Radiometer (AVHRR) images for burned land mapping was tested using data acquired over the Iberian Peninsula in 2001, 2003 and 2004. The experiment was based on four tests that assessed the discriminability between burned and unburned areas, the presence of artifacts (clouds and clouds shadows), the verticality of the sensor viewing angle, and the spatial coherency of the composite images. Seven different compositing techniques were tested, based on maximizing normalized difference vegetation index (NDVI) and brightness/surface temperature, and minimizing reflectance and sensor zenith angles. The composite criterions that provide the most accurate images for burned land mapping were based on maximizing brightness/surface temperatures, either as the only criterion or in conjunction with minimizing sensor zenith angle or near infrared (NIR) reflectance. These composites present high discrimination capacity between burned and unburned areas, remove most clouds and cloud shadows, offer high spatial coherency and present middle-to-low sensor zenith angles. Traditional compositing criterion based on maximizing NDVI values provided poor results in most tests. Finally, standard NASA MODIS composite provides close to nadir observation angles, and good spatial coherency, but it offered lower discrimination between burned and unburned areas that those composites based on thermal data.     

Characterization of the tropical diurnal fire cycle using VIRS and MODIS observations

Seven years of data from the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) were used to characterize the average diurnal fire cycle in 15 regions of the tropics and sub-tropics. Bias errors in the resulting diurnal cycles were either avoided or removed through a combination of judicious region selection and the application of corrections to compensate for cloud obscuration and time-dependent “blind spots” in the fire-detection capability of the VIRS sensor. Supplementary data from the Moderate Resolution Imaging Spectroradiometer (MODIS) on board NASA's Terra satellite aided this process. In all regions, the local time of peak burning fell between 13:00 and 18:30, with fire activity peaking distinctly earlier for the heavily forested regions. The time period of the central 50% of total daily fire activity varied from a minimum of 1.3 h in North Central Africa to a maximum of 5.5 h in Eastern Australia. In general, shorter periods of burning were associated with greater tree cover. Using the diurnal cycles obtained for each region, an analysis of the drift in the local overpass times of the NOAA-7 through NOAA-14 afternoon satellites was performed. Results show that very large, spurious trends are likely to occur in a long-term Advanced Very High Resolution Radiometer (AVHRR) fire record due to differences in diurnal sampling over time.     

Fire radiative energy for quantitative study of biomass burning: derivation from the BIRD experimental satellite and comparison to MODIS fire products

A major focus in global change research is to quantify the amount of gaseous and particulate pollutants emitted from terrestrial vegetation fires. Determination of the emitted radiant energy released during biomass combustion episodes (the so-called fire radiative energy or FRE) has been suggested as a new tool for determining variations in biomass combustion rates and the rate of production of atmospheric pollutants. We review the physical principals behind the remote determination of FRE and present an alternative method for its derivation via analysis of ‘fire pixel’ radiances in the middle infrared spectral region. We compare our method to the existing FRE retrieval approach used in the EOS Moderate Resolution Imaging Spectro-radiometer (MODIS) fire products, and to retrievals of FRE based on derived fire temperature and area made via the so-called Bi-spectral method. We test each FRE retrieval method using both simulated data and imagery from a new experimental space mission, the Bi-spectral InfraRed Detection (BIRD) small satellite, which has sensors specifically designed for the study of active fires. We analyse near simultaneous MODIS and BIRD data of the fires that burned around Sydney, Australia in January 2002. Despite the markedly different pixel size and spectral coverage of these sensors, where the spatial extent of the fire pixel groups detected by MODIS and BIRD are similar, the derived values of FRE for these fires agree to within ±15 %. However, in certain fires, the lower spatial resolution of MODIS appears to prevent many of the less intensely radiating fire pixels being detected as such, meaning MODIS underestimates FRE for these fires by up to 46% in comparison to BIRD. Though the FRE release of each of these low intensity fire pixels is relatively low, their comparatively large number makes their overall FRE significant. Thus, total FRE release of the Sydney fires on 5 January 2002 is estimated to be 6.5×10⁹ J s⁻¹ via BIRD but 4.0×10⁹ J s⁻¹ via MODIS. The ability of BIRD to resolve individual fire fronts further allows the first accurate calculation of ‘radiative’ fireline intensity from spaceborne measurements, providing values of 15-75 kJ s⁻¹ m⁻¹ for fire fronts that are up to 9 km in length. Finally, we analyse the effectiveness of the satellite-based FRE retrieval methods in estimating the FRE from the active flaming and smouldering components only (FRE_Active, believed to be proportional to the rate of biomass combustion), despite the sensor receiving additional radiance from the ‘cooling ground’. The MIR radiance method appears particularly strong in this regard, allowing FRE_Active to be estimated to within ±30% in the range 100-100,000 J s⁻¹ m⁻². These results provide further confidence in the ability of spaceborne missions to derive physically meaningful values of FRE that could be used to support biomass burning emissions inventories. Future comparisons between FRE derived via MODIS and those from higher spatial resolution BIRD or airborne imagery may allow the MODIS-derived FRE values to be ‘calibrated’ for any systematic underestimation. We therefore expect FRE to become an important tool for enhancing global studies of terrestrial vegetation fires with infrared remote sensing, particularly as the majority of large fires are now imaged four times per day via the MODIS instruments on the Terra and Aqua spacecraft.     

Object-based image classification for burned area mapping of Creus Cape, Spain, using NOAA-AVHRR imagery

Due to the ability of the NOAA-AVHRR sensor to cover a wide area and its high temporal frequency, it is possible to quickly obtain a general overview of the prevailing situation over a large area of terrain and, more specifically, quickly assess the damage caused by a recent large forest fire by mapping the extent of the burned area. The aim of this work was to map a large forest fire that recently took place on the Spanish Mediterranean coast using innovative image classification techniques and low spatial resolution imagery. The methodology involved developing an object-based classification model using spectral as well as contextual object information. The burned area map resulting from the image classification was compared with the fire perimeter provided by the Catalan Environmental Department in terms of spatial overlap and size in order to determine to what extent they were compatible. Results of the comparison indicated a high degree (≈90%) of spatial agreement. The total burned area of the classified image was found to be 6900 ha, compared to a fire perimeter of 6000 ha produced by the Catalan Environmental Department. It was concluded that, although the object-oriented classification approach was capable of affording very promising results when mapping a recent burn on the Spanish Mediterranean coast, the method in question required further assessment to ascertain its ability to map other burned areas in the Mediterranean.     

Addressing the spatiotemporal sampling design of MODIS to provide estimates of the fire radiative energy emitted from Africa

Satellite-based estimates of the fire radiative power (FRP) and energy (FRE) emitted from open biomass burning are affected by the spatiotemporal resolution of polar-orbiting and geostationary sensors. Here the impacts of the MODIS sampling design on estimates of FRE are characterized by superimposing the timing and extents of the Terra and Aqua granules onto the SEVIRI active fire product. Results for different land-cover types across Africa indicate that the FRE measured by SEVIRI during eight days is linearly related to the sum of FRP measured by SEVIRI within the MODIS granules. These relationships are least variable during the height of the fire season when diurnal cycles of FRP measured by SEVIRI are most consistent. Relationships between FRE and the sum of FRP developed using the SEVIRI active fire product are directly applied to the sum of FRP retrieved from the MODIS Terra and Aqua Climate Modeling Grid (CMG) fire products. Estimates of FRE from MODIS herein agree within 5% of those obtained from previously published methods, but remain a factor of 0.72 times those obtained by adjusting SEVIRI measurements of FRE to account for low spatial resolution detection biases. An examination of the MODIS scan geometry suggests that the latter underestimation is attributed to the coupling between a MODIS imaging artefact referred to as the “bow-tie” effect and the typical calculation used to retrieve the sum of FRP from the MODIS CMG fire products. Depending on the availability of MODIS scan angle information, we offer rigorous and simplified calculations to account for the bow-tie effect. Applying the simplified adjustment to the MODIS CMG fire products yields national estimates of monthly FRE that are 1.44 times greater than originally predicted.     

Active fire detection and characterization with the advanced spaceborne thermal emission and reflection radiometer (ASTER)

We present an automated fire detection algorithm for the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor capable of mapping actively burning fires at 30-m spatial resolution. For daytime scenes, our approach uses near infrared and short-wave infrared reflectance imagery. For nighttime scenes a simple short wave infrared radiance threshold is applied. Based on a statistical analysis of 100 ASTER scenes, we established omission and commission error rates for nine different regions. In most regions the probability of detection was between 0.8 and 0.9. Probabilities of false alarm varied between 9 × 10^− 8 (India) and 2 × 10^− 5 (USA/Canada). In most cases, the majority of false fire pixels were linked to clusters of true fire pixels, suggesting that most false fire pixels occur along ambiguous fire boundaries. We next consider fire characterization, and formulate an empirical method for estimating fire radiative power (FRP), a measure of fire intensity, using three ASTER thermal infrared channels. We performed a preliminary evaluation of our retrieval approach using four prescribed fires which were active at the time of the Terra overpass for which limited ground-truth data were collected. Retrieved FRP was accurate to within 20%, with the exception of one fire partially obscured by heavy soot.     

An assessment of the MODIS collection 5 leaf area index product for a region of mixed coniferous forest

Canopy leaf area index (LAI), defined as the single-sided leaf area per unit ground area, is a quantitative measure of canopy foliar area. LAI is a controlling biophysical property of vegetation function, and quantifying LAI is thus vital for understanding energy, carbon and water fluxes between the land surface and the atmosphere. LAI is routinely available from Earth Observation (EO) instruments such as MODIS. However EO-derived estimates of LAI require validation before they are utilised by the ecosystem modelling community. Previous validation work on the MODIS collection 4 (c4) product suggested considerable error especially in forested biomes, and as a result significant modification of the MODIS LAI algorithm has been made for the most recent collection 5 (c5). As a result of these changes the current MODIS LAI product has not been widely validated. We present a validation of the MODIS c5 LAI product over a 121 km² area of mixed coniferous forest in Oregon, USA, based on detailed ground measurements which we have upscaled using high resolution EO data. Our analysis suggests that c5 shows a much more realistic temporal LAI dynamic over c4 values for the site we examined. We find improved spatial consistency between the MODIS c5 LAI product and upscaled in situ measurements. However results also suggest that the c5 LAI product underestimates the upper range of upscaled in situ LAI measurements.     

Agricultural burning in the Southeastern United States detected by MODIS

The southeastern United States, including the states of Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, and Virginia, had a high occurrence of fire activity as detected by the 1 km Moderate Resolution Imaging Spectroradiometer (MODIS) TERRA Active Fire Product (MOD 14). The analysis of the satellite data from 2001 to 2004 showed that agricultural burning in the southeastern United States accounted for an average of 16% of annual fire activity. The southeastern region contributed an average of 33% of all agricultural burning detected in the contiguous United States. Crop residues that burned in the southeast included rice, winter wheat, sugarcane, soybean and cotton. Much of the agricultural burning occurred in June and from October to January and was related to the harvest of winter wheat and rice in the spring and the harvest of sugarcane, soybean and cotton in the fall and winter. The results showed that cropland burning was spatially dependent on crop type and temporally dependent on management practices (planting/harvesting). Arkansas, Florida, and Louisiana contributed more than 75% of all agricultural burning in the southeast. A 250 m MODIS land cover map cover for 2004 was developed for these three states using a decision tree classification and validation from a field campaign in the fall of 2004. Compared to the standard MODIS 1 km Land Cover Dataset (MOD 12) product ([Friedl, M. A., McIver, D. K., Hodges, J. C. F., Zhang, X. Y., Muchoney, D., Strahler, A. H., Woodcock, C. E., Gopal, S., Schneider, A., Cooper, A., Baccini, A., Gao, F., Schaaf, C. (2002), Global land cover mapping from MODIS: algorithms and early results. Remote Sensing of the Environment, 83, 287-302.]), the 250 m classified images contained on average 50% more cropland area and improved the estimation of cropland area based on validation from ground control sites of croplands. Results from the decision tree classification for each state revealed that in 2004 agricultural burning contributed 73%, 54%, and 33% of total fires for Arkansas, Florida, and Louisiana, respectively.     

Prototyping of Land-SAF leaf area index algorithm with VEGETATION and MODIS data over Europe

The Satellite Application Facility on Land Surface Analysis (Land-SAF) aims to provide land surface variables for the meteorological and environmental science communities from EUMETSAT satellites. This study assesses the performance of a simplified (i.e. random distribution of vegetation is assumed) version of the Land-SAF algorithm for the estimation of Leaf Area Index (LAI) when prototyped with VEGETATION (processed in CYCLOPES program) and MODIS reflectances. The prototype estimates of LAI are evaluated both by comparison with validated CYCLOPES and MODIS LAI products derived from the same sensors and directly through comparison with ground-based estimates. Emphasis is given on evaluating the impact of the algorithm and input data on LAI retrieval discrepancies. Analysis is achieved over Europe for the 2000-2003 period. The results demonstrate the capacity of the Land-SAF algorithm to retrieve consistent LAI estimates from multiple optical sensors even when their reflectances present systematic differences. High spatial and temporal consistencies between Land-SAF prototype estimates and existing LAI products are found. The differences between Land-SAF and CYCLOPES LAI are lower than their uncertainties (RMSE (relative RMSE) within 0.4 (30%)). Land-SAF prototype estimates and MODIS LAI show larger discrepancies mainly due to differences in the vegetation structure representation and algorithm assumptions (RMSE ranging from 0.2 (30%) up to 0.8 (40%)). Land-SAF prototype provides higher LAI values than MODIS for herbaceous canopies (i.e. shrubs, grasses and crops) and lower values for woody biomes (i.e. savannas and forests). Direct validation indicates that LAI estimates from prototyping of the Land-SAF algorithm with CYCLOPES and MODIS reflectances achieve similar performances (differences with ground measurements are lower than 0.5 LAI units in 60% and 50% of the cases, respectively) as CYCLOPES and MODIS LAI products. Results from this prototyping exercise appear useful for improved retrieval of LAI and constitute a step forward for refinement, validation and consolidation of the Land-SAF algorithm.     

秸秆焚烧遥感监测进展分析与展望

秸秆焚烧是生物质燃烧的重要组成部分,不仅导致秸秆资源浪费,而且还会对环境造成严重危害。传统秸秆焚烧监测方法以人工巡查为主,监测范围受限且人力物力资源耗费大。遥感技术作为新兴的地表信息监测手段,给秸秆焚烧大范围监测带来了发展契机。介绍了遥感技术在秸秆焚烧火点监测、过火面积估算和焚烧迹地监测3个方面的基本原理、监测方法和研究进展,并分析了遥感技术在秸秆焚烧监测应用中存在的不足。在此基础上,从多源数据融合互补、监测方法优化集成、监测信息深入挖掘和时空信息决策服务等4个方面对秸秆焚烧遥感监测的未来发展进行了展望。     

Global characterization of biomass-burning patterns using satellite measurements of fire radiative energy

Remote sensing is the most practical means of measuring energy release from large open-air biomass burning. Satellite measurement of fire radiative energy (FRE) release rate or power (FRP) enables distinction between fires of different strengths. Based on a 1-km resolution fire data acquired globally by the MODerate-resolution Imaging Spectro-radiometer (MODIS) sensor aboard the Terra and Aqua satellites from 2000 to 2006, instantaneous FRP values ranged between 0.02 MW and 1866 MW, with global daily means ranging between 20 and 40 MW. Regionally, at the Aqua-MODIS afternoon overpass, the mean FRP values for Alaska, Western US, Western Australia, Quebec and the rest of Canada are significantly higher than these global means, with Quebec having the overall highest value of 85 MW. Analysis of regional mean FRP per unit area of land (FRP flux) shows that at peak fire season in certain regions, fires can be responsible for up to 0.2 W/m² at peak time of day. Zambia has the highest regional monthly mean FRP flux of ~ 0.045 W/m² at peak time of day and season, while the Middle East has the lowest value of ~ 0.0005 W/m². A simple scheme based on FRP has been devised to classify fires into five categories, to facilitate fire rating by strength, similar to earthquakes and hurricanes. The scheme uses MODIS measurements of FRP at 1-km resolution as follows: category 1 (< 100 MW), category 2 (100 to < 500 MW), category 3 (500 to < 1000 MW), category 4 (1000 to < 1500 MW), category 5 (≥ 1500 MW). In most regions of the world, over 90% of fires fall into category 1, while only less than 1% fall into each of categories 3 to 5, although these proportions may differ significantly from day to day and by season. The frequency of occurrence of the larger fires is region specific, and could not be explained by ecosystem type alone. Time-series analysis of the proportions of higher category fires based on MODIS-measured FRP from 2002 to 2006 does not show any noticeable trend because of the short time period.     

Operational active fire mapping and burnt area identification applicable to Mexican Nature Protection Areas using MODIS and NOAA-AVHRR direct readout data

Since 1999, the National Commission for the Knowledge and Use of the Biodiversity (CONABIO) in Mexico has been developing and managing the “Operational program for the detection of hot-spots using remote sensing techniques”. This program uses images from the MODerate resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites and from the Advanced Very High Resolution Radiometer of the National Oceanic and Atmospheric Administration (NOAA-AVHRR), which are operationally received through the Direct Readout station (DR) at CONABIO. This allows the near-real time monitoring of fire events in Mexico and Central America. In addition to the detection of active fires, the location of hot spots are classified with respect to vegetation types, accessibility, and risk to Nature Protection Areas (NPA). Besides the fast detection of fires, further analysis is necessary due to the considerable effects of forest fires on biodiversity and human life. This fire impact assessment is crucial to support the needs of resource managers and policy makers for adequate fire recovery and restoration actions. CONABIO attempts to meet these requirements, providing post-fire assessment products as part of the management system in particular for satellite-based burnt area mapping. This paper provides an overview of the main components of the operational system and will present an outlook to future activities and system improvements, especially the development of a burnt area product. A special focus will also be placed on the fire occurrence within NPAs of Mexico.     

A hybrid inversion method for mapping leaf area index from MODIS data: experiments and application to broadleaf and needleleaf canopies

Leaf area index (LAI) is an important variable needed by various land surface process models. It has been produced operationally from the Moderate Resolution Imaging Spectroradiometer (MODIS) data using a look-up table (LUT) method, but the inversion accuracy still needs significant improvements. We propose an alternative method in this study that integrates both the radiative transfer (RT) simulation and nonparametric regression methods. Two nonparametric regression methods (i.e., the neural network [NN] and the projection pursuit regression [PPR]) were examined. An integrated database was constructed from radiative transfer simulations tuned for two broad biome categories (broadleaf and needleleaf vegetations). A new soil reflectance index (SRI) and analytically simulated leaf optical properties were used in the parameterization process. This algorithm was tested in two sites, one at Maryland, USA, a middle latitude temperate agricultural area, and the other at Canada, a boreal forest site, and LAI was accurately estimated. The derived LAI maps were also compared with those from MODIS science team and ETM+ data. The MODIS standard LAI products were found consistent with our results for broadleaf crops, needleleaf forest, and other cover types, but overestimated broadleaf forest by 2.0-3.0 due to the complex biome types.