monitoring variables. Thus, the integration of other environmental monitoring variables may lead to long-term benefits  and shape the future of REDD+ monitoring and implementation efforts . 4.4. Future Research Directions
In situ forest fire occurrence data provided by the Korea Forest Service were used as reference data in this study (Appendix B ). Each forest fire case contains information about the starting and extinguishing date/time, location (specific address), damaged area, and cause. When a forest fire occurs in a region, the public officials in charge of the region confirm and report the fire in detail. Damaged areas are calculated by trained forest fire experts based on visual observations, actual measurements using Global Positioning System (GPS) survey equipment, aerial photographs, and/or topographic maps with a scale of 1:25,000 [ 28 ]. Small forest fires damaging less than 0.7 ha of land were not considered in this study because most of them did not show little spectral difference in the Himawari-8 time-series databased on visual inspection of the images. It should be noted, though that pixel radiance is affected by not only a fire, but also many other factors. Among the 114 forest fires that resulted in damaged areas of over 0.7 ha during the study period, 64 cases that were clearly distinguishable from the satellitedata without being blocked by clouds were selected as reference data, resulting in 2165 fire pixels and 18,085 non-fire pixels between 2015 and 2017. Note that the non-fire pixels were randomly extracted from the forested areas from the images after excluding fire and cloud pixels.
early in the development of the technology. As satellite remote sensing simplified the initial inventory of forest resources, GIS provided the ability to monitor and record the changes. At present, most forestmonitoring focuses on activity data, i.e., data on forest cover changes  and two approaches are used: top–down and bottom– up. The top–down approach utilizes satellite systems , and the bottom–up approach employs ground observation . Satellitedata provide systematic coverage and a higher frequency of acquisition at a low cost, which is crucial for forestmonitoring -. However, the operational use of these systems is influenced by several factors such as cloud cover, seasonality, and the limited spatial, spectral, and temporal resolution of satelliteobservations that can lead to an inevitable lag in forest change detection -. The ability to use multiple spatial resolutions of imagery, in conjunction with field data, in multiscale analyses is a burgeoning area for research and applications. However, a perceived mismatch exists between the data needed by ecologists and the data collected with remote sensing instruments . This perception is declining due to the availability of high- resolution data that can be directly linked to traditional field-based ecological measurements . The nesting of data from different scales in an information hierarchy provides opportunities for gathering detailed, site- specific information over increasingly large areas. A new method to detect, characterize, and monitor forest change is by integrating remote sensing and GIS data . Forest change detection analysis, which employs both GIS thematic data and remotely sensed data obtained prior to and following a disturbance, is often conducted to assess speciﬁc types of forest damage ,. Integration of satellite and GIS data substantially improves impact/ damage assessment and map accuracy. The present study is conducted with the objective to design an interactive way that combines Web-based GIS tech-nologies and open-source satellitedata to support forestmonitoring.
for changing plankton community composition. The challenge is that these parameterizations need to be functions of remotely sensible variables while having predictive power. The particle size distribution satellitedata products used here are potentially useful for this purpose as should remote-sensing methodologies that quantify biomass in different phytoplankton functional types [e.g., Alvain et al., 2005; Bracher et al., 2009]. These improvements require data sets that span food-web dynamics to ocean optics observations. Last, the end-to-end validation of the model performed here is clearly limited. The data sources were selected by Buesseler and Boyd  based upon their ability to estimate TotEZ, the particle export ﬂ ux at the base of the euphotic zone, and to span a range of conditions. Most available sediment trap or 234 Th ﬂ ux observations are at ﬁ xed depths (100 or 150 m), which is often deeper than the optically determined euphotic zone depth (Figure 2a). Depth-resolved ﬂ ux pro ﬁ les are most commonly available from 234 Th disequilibrium pro ﬁ les but also require knowledge of the 234 Th:POC ratio as a function of depth in order to calculate the ﬂ ux at Z eu [e.g.,
For each reference period, the 5 processed Landsat TM/ETM+ images were merged together to maximize the spatial coverage. Figure 2.7 visualizes the merging process for 2010 by the NDFI. Due to masked clouds and ETM+ SLC-off areas for which no valid information was available from any of the 5 Landsat input images, gaps remained for 5.1% (382 km²) and 3.1% (237 km²) of the entire study area for 2007 and 2010, respectively (Table 2.5). It has to be mentioned here that open mining pits were partly masked as clouds because of a strong spectral overlap (Landsat fractions) between clouds and the very bright sandy soil in the open mining pits. In addition, few very dark cloud shadow areas were not detected by the masking algorithm and cause artifacts within the degraded forest class. These problems, however, are well known and can be solved by a cloud detection algorithm that additionally incorporates thermal information (e.g. FMASK, (Zhu & Woodcock 2012)). Combining the Landsat data set of 2007 and 2010 only a marginal part of the data gaps are overlapping (0.3%, 22 km²) and for 7.9% (597 km²) of the study area no dual-date information is available (Table 2.5). The results shows that in a heavy cloud contaminated environment, a series of Landsat images is required to produce a dataset with a reasonable spatial coverage. In our case, Landsat images acquired throughout an entire year had to be merged to get 95% coverage. This prohibits a meaningful intra-annual analysis using only Landsat information. Although Landsat images acquired over a period of up to 11 month (reference year 2007) were combined, only dry season observations were selected to avoid major seasonal variations that otherwise need to be corrected to provide consistent data. SAR layover and shadow effects in mountainous parts of the study area lead to data gaps that account for 0.05% (3.81 km²) of the study area. In contrast to the Landsat data gaps, SAR layover and shadow areas have always the same location for a certain SAR acquisition specification (sensor, orbit, mode and associated incidence angle). In the present case, almost the entire SAR layover and shadow areas are covered by Landsat information and only a negligible part of ~0.01% is missing in both datasets (Table 2.5).
The main challenge of studying drought is the lack of precipitation measurements over large geographic areas. The low spatial density of rain gauges, the accuracy of available measurements and the lack of rainfall archive at large spatial scale are the mains limiting problems. Such constraints are not specific to Morocco but concern almost every country in the world but in various importance levels. To overcome these problems, the use of remote sensing data appears to be the most efficient tool in terms of accuracy, spatial coverage as well as economical cost. In last twenty years, several precipitation data sets derived from various remote sensing products have been released. According to the approaches and algorithms applied to estimate precipitation amount, three sources of data can be considered: There are models that provide rainfall estimates using infrared satellite imagery, such as PERSIANN (Precipitation Estimation from Remote Sensing Information using Artificial Neural Network, ) and technics like CMORPH (CPC MORPHing technique,  that estimates rainfall based on passive microwave and infrared satellitedata. There is finally TRMM (Tropical Rainfall Measurement Mission, a space mission that provides several products of rain estimates from a combination of passive microwave, visible/infrared and a rainfall radar data . Among these sources of precipitation data at large geographic scale, TRMM, which was primarily designed to monitor and study tropical rainfall, have proved to be an important source of information in many application fields, such as monitoring global hydrological cycle [9-11], floods [12-14], and drought assessment [15-17] Indeed, the evaluation of TRMM data against rain gauges has been conducted in various previous studies around the world. Some studies have evaluated the rainfall estimates from TRMM in parallel with other satellite products [18-21]. Almazroui (2011) and Mantas et al. (2015) have focused their work on the TRMM product by evaluating its accuracy on different time steps and in different geographical and topographical contexts [22-23]. In addition to the direct comparison to rain gauges, Collischonn et al. (2008) provided an assessment based on hydrological modeling at a daily time step . Islam and Uyeda (2007) compared daily rainfall from TRMM 3B42 to rain-gauge measurements over Bangladesh . Huang et al. (2014) evaluated the TMPA V7 products with a relatively dense rain gauge network in Beijing and adjacent regions for an extreme precipitation event .
Mapping and monitoring carbon stocks in forested regions of the world, particularly the tropics, has attracted a great deal of attention in recent years as deforestation and forest degradation account for up to 30% of anthropogenic carbon emissions, and are now included in climate change negotiations. We review the potential for satellites to measure carbon stocks, specifically aboveground biomass (AGB), and provide an overview of a range of approaches that have been developed and used to map AGB across a diverse set of conditions and geographic areas. We provide a summary of types of remote sensing measurements relevant to mapping AGB, and assess the relative merits and limitations of each. We then provide an overview of traditional techniques of mapping AGB based on ascribing field measurements to vegetation or land cover type classes, and describe the merits and limitations of those relative to recent data mining algorithms used in the context of an approach based on direct utilization of remote sensing measurements, whether optical or lidar reflectance, or radar backscatter. We conclude that while satellite remote sensing has often been discounted as inadequate for the task, attempts to map AGB without satellite imagery are insufficient. Moreover, the direct remote sensing approach provided more coherent maps of AGB relative to traditional approaches. We demonstrate this with a case study focused on continental Africa and discuss the work in the context of reducing uncertainty for carbon monitoring and markets.
In order to validate the estimated velocity, we have computed the instantaneous velocity field using the time se- ries altimeter-derived velocity anomaly field and the mean velocity field and compared it with the available in-situ ve- locities measured by the shipboard Acoustic Doppler Cur- rent Profiler (ADCP) data obtained from the WOCE cruises (WOCE Global Data version 3.0) of R/V Knorr (cruise id 180, 373, 500, 501) and R/V Franklin (id 339, 340). The depth of ADCP data used varies from 20 - 30 m and we as- sume that it represents the near surface geostrophic current. The instantaneous velocity estimated in this study is in good agreement with the ADCP derived velocities (Fig. 7). The regression lines fitted have slopes 0.89 and 0.87 for the u and v velocity components respectively and have correla- tion co-efficients of 0.89 and 0.88 respectively.
Abstract. The performance of the three cloud products cloud fractional cover, cloud type and cloud top height, derived from NOAA AVHRR data and produced by the EUMETSAT Climate MonitoringSatellite Application Fa- cility, has been evaluated in detail over the Arctic re- gion for four months in 2007 using CALIPSO-CALIOP observations. The evaluation was based on 142 selected NOAA/Metop overpasses allowing almost 400 000 individ- ual matchups between AVHRR pixels and CALIOP mea- surements distributed approximately equally over the studied months (June, July, August and December 2007). Results suggest that estimations of cloud amounts are very accurate during the polar summer season while a substantial loss of detected clouds occurs in the polar winter. Evaluation re- sults for cloud type and cloud top products point at specific problems related to the existence of near isothermal condi- tions in the lower troposphere in the polar summer and the use of reference vertical temperature profiles from Numerical Weather Prediction model analyses. The latter are currently not detailed enough in describing true conditions relevant on the pixel scale. This concerns especially the description of near-surface temperature inversions which are often too weak leading to large errors in interpreted cloud top heights.
Abstract—This paper proposes a method which describes the information precision with a soft fusion model, instead of the traditional rigid fusion method. The method is divided into two steps, the pretreatment model and fusion center model. Each forms a relative independent model, and the two models have a progressive relationship. The former is used for consistency evaluation, data cleaning and invalid data eliminating, while the latter provides fusion results and variable precision fusion expression by the adaptive threshold clustering algorithm. Experimental results show that the fusion method can not only give every SST data a different precision, but also carry more information to describe precision multiple distribution, which make users get high-quality data and enjoy more rights.
A few critical challenges of these community-basedmonitoring systems include the efficacy of high-intensity annual monitoring using permanent sample plots, lack of spatial extrapolation power to address changes beyond the sample plots, such as forest cover loss and degrada- tion, and the reliable quantification of carbon dynam- ics over the entire study area . Danielsen et al.  stressed the need for reliable third-party evaluations with a detailed explanation of the potential disadvan- tages of communitymonitoring, such as biased report- ing by communities, intimidation of communities for biased reporting and over-burdening communities with workloads from state-owned systems . Skutsch et al.  and Bavikatte and Jonas  stressed the need for the rational integration of geospatial technologies to improve the efficacy of communitymonitoring systems. Remote-sensing-basedmonitoring of forest degradation/ enhancement essentially involves two main approaches : first, detection indicated by a change in canopy cover or proxies and second, the quantification of gain or loss in AGB. Both approaches provide spatially explicit estimates with wall to wall coverage, enabling the under- standing of dynamics beyond point-based estimates [4, 19]. The reliable remote-sensing-based visual indicators and systematic information systems could also extend objectivity to verification and third-party evaluations [1, 20]. The multi-resolution satellite systems help to address biomass estimations at the species, stand and forest type levels, enabling the spatial linking of different scales of information to reach from the community level to national estimates [21, 22].
This study explores the ability to assimilate frequent satellite-based leaf area index (LAI) data with an individual- based DGVM known as the SEIB-DGVM, which stands for the spatially explicit individual-based DGVM (Sato et al., 2007). We developed a non-Gaussian ensemble DA system with the SEIB-DGVM based on a particle filter approach. Although the particle filter is an existing, well-known ap- proach, this is the first attempt to apply it to an individual- based DGVM with frequent LAI data. Therefore, we focus on the methodological development in this study and per- form a series of numerical experiments at a single location with only a couple of plant functional types (PFTs) as the first step. It would be numerically straightforward to extend it to the global scale in future studies, since the local-scale experiments can be performed in parallel for different loca- tions. In the present study, we first perform idealized simula- tion experiments to investigate how well we can estimate the model parameters associated with phenology by assimilating the LAI data every 4 days, simulating the satellite-based LAI product from the Moderate Resolution Imaging Spectrora- diometer (MODIS) aboard the Terra and Aqua spacecrafts. We also investigate to what extent assimilating the LAI data could improve the estimates of the state variables such as GPP (gross primary production), RE (ecosystem respiration), NEE (net ecosystem exchange), and biomass, the most fun- damental variables for carbon cycle and vegetation states. Sensitivities to the filter settings such as the random perturba- tion sizes and particle sizes are also investigated. Following the idealized experiments, we perform an experiment using the real MODIS LAI observation data to see how well the proposed approach performs in the real world.
A forest section of the Management Plan Unit (MPU) in a mountain area of the High Tatras (Cen- tral Slovakia) was chosen as test area. The area of MPU is relatively multiple with the range of heights above sea level from 980 to 2,052 m. Different forest types occur there, mainly Sombreto-Piceetum, Cem- breto-Piceetum with dominance of spruce (Picea abies L.), also Cembreto-Mughetum and Mughetum acidofilum with dominance of dwarf pine (Pinus mugo T.). Mountain crests of MPU are covered with the meadow community where Calamagrostis vil- losa, Vaccinium myrtillus, Vaccinium vitis-idea and Juncus trifidus are dominant.
Harsh conditions and/or distance to transport infrastructure can make it dif ﬁcult/impossible to visit sites that are remote, resulting in a desire for remote data backhaul and autonomous system operation. The technological advances in sensor and communication technology have enabled development of long-term intelligent monitoring systems. This technology extends to a range of application scenarios , including surveillance, emergency communications, support and environmental monitoring. This paper focuses on an emerging application to support long-term environmental monitoring using cam- eras and sensor networks. This can be used in ecological research monitoring natural resources, biodi- versity and ecological and social sustainability [2 –4]. Many areas of interest to this type of research are geographically isolated and require systems able to operate unattended for prolonged periods. The absence of local communications infrastructure motivates the use of satellite communications services. The architecture presented in this paper was developed for the WiSE project  at the University of Aberdeen, Scotland, UK. The system supports long-term monitoring of sporadic events. While it is cru- cial, the system could report all events of interest, and the system must also carefully manage its use of power resources to ensure availability over a long-term deployment. A commercial-off-the-shelf broad- band satellite terminal provides backhaul communications to locations where there is little or no other communications, although the platform can also incorporate general packet radio service/Global System for Mobile communication [6,7] where there is coverage by commercial mobile access networks. A wired/wireless Local Area Networks connects motion and networked environmental sensors.
Remote sensing provides a useful source of data from which updated land cover information can be extracted for assessing and monitoring vegetation changes. Remote sensing is one of a suite of tools available to land managers that provide up-to-date, detailed information about land condition. Remote sensing uses instruments mounted on satellites or in planes to produce images or 'scenes' of the Earth's surface. Remote sensing satellitedata aims for the achieving higher accuracy and more detailed results for classifications [1-5] . Vegetation change may be a terminology of rather comprehensive definitions, which ranges from the in growth of a single tree to the entire deforestation by clear-cut. Whether we can detect and monitor vegetation changes by remote sensing data depends on the spatial and temporal characteristics of the change and the type of remote sensor data to be used. Therefore, it is important for us to understand the nature of vegetation changes prior to analyzing remote sensing data. In the past several decades, air photo interpretation has played an important role in detailed vegetation mapping [6-7] .
In terms of whether the biomass maps are “accurate enough” to be recommended for carbon management purposes in these counties, it appears that one can obtain reasonable biomass values in many, but not all, areas at the plot scale (roughly 1.5 acres). Furthermore, as mentioned above, county scale estimates were only useful for Howard County, but not Anne Arundel, where more work is needed. The current evaluations have already been considered in the process of designing more effective field collection strategies and modeling approaches for devel- oping improved biomass maps in Maryland counties. For example, newer random forest models exclude variable ra- dius plot locations that had biomass detected by LIDAR over a 30-m area (the pixel size) but that had no trees measured in them. This can occur when trees are at the edge of a pixel, too far away to be included in the variable radius plot measurement, but still being observed by the LIDAR. When these locations were excluded the resulting model had better agreement with the FIA data because there were fewer instances where biomass was predicted in the FIA plot but there was no biomass measured (R 2 = 0.59, RMSE = 82.4 Mg/ha, slope = 1.1; compare with Figure 2a). Another issue contributing to the poor agreement was probably our combination of a single plot design of the NFI and the regular FIA plot design, resulting in inconsistent plot-pixel comparisons through- out the sample. As “nonforest” biomass is important to consider in Maryland and elsewhere, plot designs and overall strategies for addressing the “nonforest” biomass gap, are discussed below.
The spectral property of the training sets along with previous knowledge, as well as the data from field studies and higher resolution images were combined to perform a supervised classification. In this research dataset from Landsat 5- TM was interpreted straight to LULC in 5 classes as follows; (F) Forest, (W) Water body, (U) Urban, (P) Paddy field, and (C) Crop field. Figure 4 illustrates sample training area of water and sample training area of forest.
It is not possible a priori to indicate which predictors will lead to the best result in the desired classification. The dependencies and correlations between them are too complex. Commonly used is the forward stepwise regression, Wilks (1995). In each step a predictor is added to the equation and based on the statistical scores it is decided if the additional predictor contributes to the overall performance. It is up to the user to decide how many steps or predictors contribute significantly to the classification performance. Using all predictors may lead to an over-fit regression, Wilks (1995). In an over-fit regression too many predictors are used in the equation to describe the observations. The regression will fit to the used observations but the equation may fail to describe other observations not used for its determination.
There has been considerable discussion in recent literature about the effect of climate change on winds and transport in the Southern Ocean. Large-scale climate models predict a poleward movement and strengthening of the westerly winds over the Southern Ocean in a warming world plus depletion of polar stratospheric ozone [e.g., Fyfe and Saenko, 2006]. Both effects have been shown to lead to increased trans- port and a southward shift of the Antarctic Circumpolar Current (ACC) in climate models [e.g., Fyfe and Saenko, 2006]. However, experiments with eddy-resolving models find increased westerlies in the Southern Ocean, leading to more energetic eddy variability with no significant trends in transport through the Drake Passage [e.g., Hallberg and Gnanadesikan, 2006]. Observations of ACC transport also suggest that the long- term trends in transport are much smaller than what large-scale models predict, and there is no observatio- nal evidence for a significant trend [Gille, 2008; B€oning et al., 2008; Cunningham et al., 2003; Rintoul and Sokolov, 2001; Rintoul et al., 2002]. The lack of an apparent trend may be due to the limited number of observations, drift in the models, the current having reached the ‘‘eddy saturation limit’’ [Hallberg and Gna- nadesikan, 2001], or from a compensating signal caused by freshwater and/or heat fluxes south of the ACC [e.g., Marshall and Radko, 2003].
perform well against other reference datasets, i.e. whether a parameter set that yields for example good soil moisture performance also yields realistic LSTs. For this purpose we assess the performance of parameter perturbations perform- ing best against particular reference datasets with respect to all other reference datasets in Fig. 3. White colours mean that parameter perturbations which perform well against par- ticular reference datasets (x-axis) also perform well against other reference datasets (y -axis). Vice versa, black colours indicate that they do not also perform well against other ref- erence datasets. Note that in the case of a perfect model and perfect observations this plot would be completely white. The many dark coloured fields in Fig. 3 indicate that the parameter perturbations performing best against particular reference datasets are different, i.e. there is no parameter perturbation that performs best in all respects. This can be explained by (1) equifinality (i.e. many different parameter sets leading to equally well-performing model simulations) as there are 25 pre-selected well-performing parameter sets among all 50 considered parameter sets, and by (2) incon- sistencies within HTESSEL, especially between hydrologi- cal and skin-temperature-related processes. This is apparent as for example HTESSEL configurations performing well in terms of LSTs yield particularly poor performance in terms of hydrology, and vice versa. These inconsistencies might be partly associated with missing processes in HTESSEL, for example the over-simplification that a single parameter rep- resents the complex energy transfers between the top of the canopy and the underlying soil.