Feasibility of PSO–ANFIS model to estimate rock fragmentation produced by mine blasting

Desired rock fragmentation is the main goal of the blasting operation in surface mines, civil and tunneling works. Therefore, precise prediction of rock fragmentation is very important to achieve an economically successful outcome. The primary objective of this article is to propose a new model for forecasting the rock fragmentation using adaptive neuro-fuzzy inference system (ANFIS) in combination with particle swarm optimization (PSO). The proposed PSO–ANFIS model has been compared with support vector machines (SVM), ANFIS and nonlinear multiple regression (MR) models. To construct the predictive models, 72 blasting events were investigated, and the values of rock fragmentation as well as five effective parameters on rock fragmentation, i.e., specific charge, stemming, spacing, burden and maximum charge used per delay were measured. Based on several statistical functions [e.g., coefficient of correlation (R ²) and root-mean-square error (RMSE)], it was found that the PSO–ANFIS (with R ² = 0.89 and RMSE = 1.31) performs better than the SVM (with R ² = 0.83 and RMSE = 1.66), ANFIS (with R ² = 0.81 and RMSE = 1.78) and nonlinear MR (with R ² = 0.57 and RMSE = 3.93) models. Finally, the sensitivity analysis shows that the burden and maximum charge used per delay have the least and the most effects on the rock fragmentation, respectively.

     

Applications of soft computing techniques for prediction of energy dissipation on stepped spillways

In this study, numbers type of soft computing including artificial neural network (ANN), support vector machine (SVM), multivariate adaptive regression splines (MARS), and group method of data handling (GMDH) were applied to model and predict energy dissipation of flow over stepped spillways. Results of ANN indicated that this model including hyperbolic tangent sigmoid as transfer function obtained coefficient of determination (R ² = 0.917) and root-mean-square error (RMSE = 6.927) in testing stage. Results of development of SVM showed that developed model consists of radial basis function as kernel function achieved R ² = 0.98 and RMSE = 2.61 in validation stage. Developed MARS model with R ² = 0.99 and RMSE = 0.65 has suitable performance for predicating the energy dissipation. Results of developed GMDH model show with R ² = 0.95 and RMSE = 5.4 has suitable performance for modeling energy dispersion. Reviewing of results of prepared models showed that all of them have suitable performance to predict the energy dissipation. However, MARS and SVM are more accurate than the others. Attention to structures of GMDH and MARS models declared that Froude number, drop number, and ratio of critical depth to height of step are the most important parameters for modeling energy dissipation. The best radial basis function was found that as best kernel function in developing the SVM.

     

Canopy level estimations of chlorophyll and LAI for two tropical species (teak and bamboo) from Hyperion (EO1) data

Current economic development in tropical regions (especially in India, China, and Brazil) is putting tremendous pressure on tropical forest cover. Some of the dominant and economically important species are planted at large scale in these countries. Teak and bamboo are two important species of tropical regions because of their commercial and conservation values. Accurate estimates of foliar chemistry can help in evaluating the health status of vegetation in these regions. An attempt has been made to derive canopy level estimation of chlorophyll and leaf area index (LAI) for these species utilizing Hyperion data. Partial least square (PLS) regression analysis was carried out to identify the correlation between measured parameters (chlorophyll and LAI) and Hyperion reflectance spectra. PLS regression identified 600–750 nm as a sensitive spectral region for chlorophyll and 1000–1507 nm for LAI. The PLS regression model tested in this study worked well for the estimation of chlorophyll (R ² = 0.90, root mean square error (RMSE) = 0.182 for teak and R ² = 0.84, RMSE = 0.113 for bamboo) and for the estimation of LAI (R ² = 0.87, RMSE = 0.425). The lower predictive error obtained indicates the robustness of the data set and also of the applicability of the PLS regression analysis. Wavelengths recognized by the PLS regression model were utilized for the development of vegetation indices for estimating chlorophyll and LAI. Predictive performances of the developed simple ratios (SRs) were evaluated using the cross-validation method. SR 743/692 gave the best results for the prediction of chlorophyll with the leave-one-out cross-validation (LOO-CV) method (R ² = 0.73, RMSE = 0.28 for teak and R ² = 0.71, RMSE = 0.15 for bamboo). The normalized difference ratio (ND 1457/1084) gave the best results for the prediction of LAI with LOO-CV (R ² = 0.66, RMSE = 0.57). Ratios developed here can be tested for teak and bamboo cover spread in tropical regions with similar environmental conditions.     

Monitoring yield and fruit quality parameters in open-canopy tree crops under water stress. Implications for ASTER

Work on water stress detection at tree and orchard levels using a high-spatial airborne thermal sensor is presented, showing its connection with yield and some fruit quality indicators in olive and peach commercial orchards under different irrigation regimes. Two airborne campaigns were conducted with the Airborne Hyperspectral Scanner (AHS) over olive and peach orchards located in Córdoba, southern Spain. The AHS sensor was flown at three different times on 25 July 2004 and 16 July 2005, collecting 2 m spatial resolution imagery in 80 spectral bands in the 0.43-12.5 μm spectral range. Thermal bands were assessed for the retrieval of land surface temperature using the split-window algorithm and TES (Temperature-Emissivity-Separation) method, separating pure crowns from shadows and sunlit soil pixels using the reflectance bands. Stem water potential and stomatal conductance were measured on selected trees at the time of airborne flights over the orchards. Tree fruit yield and quality parameters such as oil, weight and water content (for the olive trees), and fruit volume and weight (for the peach trees) were obtained at harvest and through laboratory analysis. Relationships between airborne-estimated crown temperature minus air temperature and stem water potential yielded r² = 0.5 (12:30 GMT) at the olive tree level, and r² = 0.81 (9:00 GMT) at the treatment level in peach trees. These results demonstrate that water stress can be detected at the crown level even under the usual water management conditions of commercial orchards. Regressions of yield and fruit quality against remote sensing estimates of crown temperature as an indicator of water stress, yielded r² = 0.95 (olive fruit water content) and r² = 0.94 (peach fruit mean diameter). These results suggest that high-spatial remote sensing thermal imagery has potential as an indicator of some fruit quality parameters for crop field segmentation and irrigation management purposes. A simulation study using ASTER spectral bands and aggregated pixels for stress detection as a function of irrigation level was conducted in order to study the applicability of medium resolution thermal sensors for the global monitoring of open-canopy tree crops. The determination coefficients obtained between the ASTER-simulated canopy temperature minus air temperature and stem water potential yielded r² = 0.58 (12:30 GMT) for olive trees, suggesting the potential for extrapolating these methods to ASTER satellite for global monitoring of open tree canopies.     

Remote sensing of the distribution and abundance of host species for spruce budworm in Northern Minnesota and Ontario

Insects and disease affect large areas of forest in the U.S. and Canada. Understanding ecosystem impacts of such disturbances requires knowledge of host species distribution patterns on the landscape. In this study, we mapped the distribution and abundance of host species for the spruce budworm (Choristoneura fumiferana) to facilitate landscape scale planning and modeling of outbreak dynamics. We used multi-temporal, multi-seasonal Landsat data and 128 ground truth plots (and 120 additional validation plots) to map basal area (BA), for 6.4 million hectares of forest in northern Minnesota and neighboring Ontario. Partial least-squares (PLS) regression was used to determine relationships between ground data and Landsat sensor data. Subsequently, BA was mapped for all forests, as well as for two specific host tree genera (Picea and Abies). These PLS regression analyses yielded estimates for overall forest BA with an R² of 0.62 and RMSE of 4.67 m² ha^? 1 (20% of measured BA), white spruce relative BA with an R² of 0.88 (RMSE = 12.57 m² ha^? 1 [20% of measured]), and balsam fir relative BA with an R² of 0.64 (RMSE = 6.08 m² ha^? 1 [33% of measured]). We also used this method to estimate the relative BA of deciduous and coniferous species, each with R² values of 0.86 and RMSE values of 9.89 m² ha^? 1 (23% of measured) and 9.78 m² ha^? 1 (16% of measured), respectively. Of note, winter imagery (with snow cover) and shortwave infrared-based indices – especially the shortwave infrared/visible ratio – strengthened the models we developed. Because ground measurements were made largely in forest stands containing spruce and fir, modeled results are not applicable to stands dominated by non-target conifers such as pines and cedar. PLS regression has proven to be an effective modeling tool for regional characterization of forest structure within spatially heterogeneous forests using multi-temporal Landsat sensor data.     

Predicting Eucalyptus spp. stand volume in Zululand,South Africa: an analysis using a stochastic gradient boosting regression ensemble with multi-source data sets

Accurate, reliable, and up-to-date forest stand volume information is a prerequisite for a detailed evaluation of commercial forest resources and their sustainable management. Commercial forest responses to global climate change remain uncertain, and hence the mapping of stand volume as carbon sinks is fundamentally important in understanding the role of forests in stabilizing climate change effects. The aim of this study was to examine the utility of stochastic gradient boosting (SGB) and multi-source data to predict stand volume of a Eucalyptus plantation in South Africa. The SGB ensemble, random forest (RF), and stepwise multiple-linear regression (SMLR) were used to predict Eucalyptus stand volume and other related tree-structural attributes such as mean tree height and mean diameter at breast height (DBH). Multi-source data consisted of SPOT-5 raw spectral features (four bands), 14 spectral vegetation indices, rainfall data, and stand age. When all variables were used, the SGB algorithm showed that stand volume can be accurately estimated (R² = 0.78 and RMSE = 33.16 m³ ha^?1 (23.01% of the mean)). The competing RF ensemble produced an R² value of 0.76 and a RMSE value of 37.28 m³ ha^?1 (38.28% of the mean). SMLR on the other hand, produced an R² value of 0.65 and an RMSE value of 42.50 m³ ha^?1 (42.50% of the mean). Our study further showed that Eucalyptus mean tree height (R² = 0.83 and RMSE = 1.63 m (9.08% of the mean)) and mean diameter at breast height (R² = 0.74 and RMSE = 1.06 (7.89% of the mean)) can also be reasonably predicted using SGB and multi-source data. Furthermore, when the most important SGB model-selected variables were used for prediction, the predictive accuracies improved significantly for mean DBH (R² = 0.81 and RMSE = 1.21 cm (6.12% of the mean)), mean tree height (R² = 0.86 and RMSE = 1.39 m (7.02% of the mean)), and stand volume (R² = 0.83 and RMSE = 29.58 m³ ha^?1 (17.63% of the mean)). These results underscore the importance of integrating multi-source data with remotely sensed data for predicting Eucalyptus stand volume and related tree-structural attributes.     

Consensus scoring model for the molecular docking study of mTOR kinase inhibitor

The discovery of mammalian target of rapamycin (mTOR) kinase inhibitors has always been a research hotspot of antitumor drugs. Consensus scoring used in the docking study of mTOR kinase inhibitors usually improves hit rate of virtual screening. Herein, we attempt to build a series of consensus scoring models based on a set of the common scoring functions. In this paper, twenty-five kinds of mTOR inhibitors (16 clinical candidate compounds and 9 promising preclinical compounds) are carefully collected, and selected for the molecular docking study used by the Glide docking programs within the standard precise (SP) mode. The predicted poses of these ligands are saved, and revaluated by twenty-six available scoring functions, respectively. Subsequently, consensus scoring models are trained based on the obtained rescoring results by the partial least squares (PLS) method, and validated by Leave-one-out (LOO) method. In addition, three kinds of ligand efficiency indices (BEI, SEI, and LLE) instead of pIC₅₀ as the activity could greatly improve the statistical quality of build models. Two best calculated models 10 and 22 using the same BEI indice have following statistical parameters, respectively: for model 10, training set R² = 0.767, Q² = 0.647, RMSE = 0.024, and for test set R² = 0.932, RMSE = 0.026; for model 22, raining set R² = 0.790, Q² = 0.627, RMSE = 0.023, and for test set R² = 0.955, RMSE = 0.020. These two consensus scoring model would be used for the docking virtual screening of novel mTOR inhibitors.     

Quantifying indicators of riparian condition in Australian tropical savannas: integrating high spatial resolution imagery and field survey data

The objectives of this research were: (1) to quantify indicators of riparian condition; and (2) to assess these indicators for detecting change in riparian condition. Two multi‐spectral QuickBird images were acquired in 2004 and 2005 for a section of the Daly River in north Australia. These data were collected coincidently with vegetation and geomorphic field data. Indicators of riparian condition, including percentage canopy cover, organic litter, canopy continuity, bank stability, flood damage, riparian zone width and vegetation overhang, were then mapped. Field measurements and vegetation indices were empirically related using regression analysis to develop algorithms for mapping organic litter and canopy cover (R ² = 0.59–0.78). Using a standard nearest‐neighbour algorithm, object‐oriented supervised image classification provided thematic information (overall accuracies 81–90%) for mapping riparian zone width and vegetation overhang. Bank stability and flood damage were mapped empirically from a combination of canopy cover information and the image classification products (R ² = 0.70–0.81). Multi‐temporal image analysis of riparian condition indicators (RCIs) demonstrated the advantages of using continuous and discrete data values as opposed to categorical data. This research demonstrates how remote sensing can be used for mapping and monitoring riparian zones in remote tropical savannas and other riparian environments at scales from 1 km to 100s km of stream length.     

Effects of adaxial and abaxial surface on the estimation of leaf chlorophyll content using hyperspectral vegetation indices

Vegetation indices are frequently used for the non-destructive assessment of leaf chemistry, especially chlorophyll content. However, most vegetation indices were developed based on the statistical relationship between the spectral reflectance of the adaxial leaf surface and chlorophyll content, even though abaxial leaf surfaces may influence reflectance spectra because of canopy structure or the inclination of leaves. In the present study, reflectance spectra from both adaxial and abaxial leaf surfaces of Populus alba and Ulmus pumila var. pendula were measured. The results showed that structural differences of the two leaf surfaces may result in differences in reflectance and hyperspectral vegetation indices. Among 30 vegetation indices tested, R₆₇₂/(R₅₅₀ × R₇₀₈) had the smallest difference (4.66% for P. alba, 2.30% for U. pumila var. pendula) between the two blade surfaces of the same leaf in both species. However, linear regression analysis showed that several vegetation indices (R₈₅₀ ? R₇₁₀)/(R₈₅₀ ? R₆₈₀), VOG₂, D₇₃₀, and D₇₄₀, had high coefficients of determination (R² > 0.8) and varied little between the two leaf surfaces of the plants we sampled. This demonstrated that these four vegetation indices had relatively stable accuracy for estimating leaf chlorophyll content. The coefficients of determination (R²) for the calibration of P. alba leaves were 0.92, 0.98, 0.93, and 0.95 on the adaxial surfaces, and 0.88, 0.87, 0.88, and 0.92 on the abaxial surfaces. The coefficients of determination (R²) for the calibration of U. pumila var. pendula leaves were 0.85, 0.91, 0.86, and 0.90 on adaxial surface, and 0.80, 0.80, 0.84, and 0.88 on abaxial surface. These four vegetation indices were readily available and were little influenced by the differences in the two leaf surfaces during the estimation of leaf chlorophyll content.     

Time-series modeling of fishery landings using ARIMA models and Fuzzy Expected Intervals software

Forecasting, using historic time-series data, has become an important tool for fisheries management. ARIMA modeling, Modeling for Optimal Forecasting techniques and Decision Support Systems based on fuzzy mathematics may be used to predict the general trend of a given fish landings time-series with increased reliability and accuracy. The present paper applies these three modeling methods to forecast anchovy fish catches landed in a given port (Thessaloniki, Greece) during 1979–2000 and hake and bonito total fish catches during 1982–2000. The paper attempts to assess the model's accuracy by comparing model results to the actual monthly fish catches of the year 2000. According to the measures of forecasting accuracy established, the best forecasting performance for anchovy was shown by the DSS model (MAPE = 28.06%, RMSE = 76.56, U-statistic = 0.67 and R² = 0.69). The optimal forecasting technique of genetic modeling improved significantly the forecasting values obtained by the selected ARIMA model. Similarly, the DSS model showed a noteworthy forecasting efficiency for the prediction of hake landings, during the year 2000 (MAPE = 2.88%, RMSE = 13.75, U-statistic = 0.19 and R² = 0.98), as compared to the other two modeling techniques. Optimal forecasting produced by combined modeling scored better than application of the simple ARIMA model. Overall, DSS results showed that the Fuzzy Expected Intervals methodology could be used as a very reliable tool for short-term predictions of fishery landings.     

Imaging chlorophyll fluorescence with an airborne narrow-band multispectral camera for vegetation stress detection

Progress in assessing the feasibility for imaging fluorescence using the O₂-A band with 1 nm full-width half-maximum (FWHM) bands centered at 757.5 and 760.5 nm is reported in this paper. Multispectral airborne data was acquired at 150 m above ground level in the thermal, visible and near infrared regions yielding imagery at 15 cm spatial resolution. Simultaneous field experiments conducted in olive, peach, and orange orchards (water stress trials), and an olive orchard (variety trial) enabled the detected variability in fluorescence emission to be examined as function of stress status. In a parallel modelling activity the coupled leaf-canopy reflectance-fluorescence model, FluorMOD, was used to assess fluorescence retrieval capability by the in-filling method, as well as by fluorescence indices from the published literature. Fluorescence retrievals using the in-filling method, the derivative index D702/D680 and reflectance indices R690/R630, R761-R757, and R761/R757 yielded the best results in the simulation study, while demonstrating insensitivity to leaf area index (LAI) variation. The fluorescence in-filling method, derivative index D702/D680, and R761-R757 were the indices least affected by chlorophyll a + b (Cab) variation. On the other hand, other published indices for fluorescence detection at leaf and canopy levels exhibited high sensitivity to variations in Cab and LAI, and therefore were considered less suitable for in-field fluorescence detection. The fluorescence signal extraction from airborne imagery using the in-filling method was validated through comparisons with field-measured steady-state fluorescence (Fs) using the PAM-2100 and GFS-3000 instruments, confirming simulation predictions. The water stress experiments conducted on olive and peach orchards demonstrated the feasibility of chlorophyll fluorescence (F) extraction at the tree level from the airborne imagery, yielding determination coefficients r² = 0.57 (olive), and r² = 0.54 (peach). Consistent results were obtained between airborne F and ground truth assimilation (A) measured in the olive variety field experiment under no water stress levels, yielding r² = 0.71.     

Extracting structural attributes from IKONOS imagery for Eucalyptus plantation forests in KwaZulu-Natal,South Africa,using image texture analysis and artificial neural networks

The suitability of optical IKONOS satellite data (multispectral and panchromatic) for the estimation of forest structural attributes – for example, stems per hectare (SPHA), diameter at breast height (DBH), mean tree height (MTH), basal area (BA) and volume in plantation forest environments – was assessed in this study. The relationships of these forest structural attributes to statistical image texture from IKONOS imagery were analysed. The coefficients of determination (R ²) of multilinear regression models developed for the estimation of SPHA, DBH, MTH, BA and volume using statistical texture features from multispectral data were 0.63, 0.68, 0.81, 0.86 and 0.86, respectively. When the statistical texture features from panchromatic data were applied, the R ² for the respective forest structural attributes increased by 25%, 31%, 6%, 0.2% and 0.2%, respectively. Artificial neural network (ANN) models produced strong and significant relationships between estimated and actual measures of SPHA, DBH, MTH, BA and volume with an R ² of 0.83, 0.83, 0.90, 0.90 and 0.92, respectively, based on multispectral IKONOS data. Based on panchromatic IKONOS imagery, the R ² for the respective forest structural attributes increased by 18%, 12%, 5%, 3% and 6%, respectively. Results such as these bode well for the application of high spatial resolution imagery to forest structural assessment.     

Comparison of artificial neural network and combined models in estimating spatial distribution of snow depth and snow water equivalent in Samsami basin of Iran

Snow water equivalent (SWE) is a key parameter in hydrological cycle, and information on regional SWE is required for various hydrological and meteorological applications, as well as for hydropower production and flood forecasting. This study compares the snow depth and SWE estimated by multivariate linear regression (MLR), discriminant function analysis, ordinary kriging, ordinary kriging-multivariate linear regression, ordinary kriging-discriminant function analysis, artificial neural network (ANN) and neural network-genetic algorithm (NNGA) models. The analysis was performed in the 5.2 km² area of Samsami basin, located in the southwest of Iran. Statistical criteria were used to measure the models’ performances. The results indicated that NNGA, ANN and MLR methods were able to predict SWE at the desirable level of accuracy. However, the NNGA model with the highest coefficient of determination (R ² = 0.70, P value < 0.05) and minimum root mean square error (RMSE = 0.202 cm) provided the best results among the other models. The lower SWE values were registered in the east of study area and higher SWE values appeared in the west of study area where altitude was higher.     

Detecting water stress effects on fruit quality in orchards with time-series PRI airborne imagery

A methodology for the assessment of fruit quality in crops subjected to different irrigation regimes is presented. High spatial resolution multispectral and thermal airborne imagery were used to monitor crown temperature and the Photochemical Reflectance Index (PRI) over three commercial orchards comprising peach, nectarine and orange fruit trees during 2008. Irrigation regimes included sustained and regulated deficit irrigation strategies, leading to high variability of fruit quality at harvest. Stem water potential was used to monitor individual tree water status on each study site. Leaf samples were collected for destructive sampling of xanthophyll pigments to assess the relationship between the xanthophyll epoxidation state (EPS) and PRI at leaf and airborne-canopy level. At harvest, fruit size, Total Soluble Solids (TSS) and Tritatable Acidity (TA) were measured to characterize fruit quality. A statistically significant relationship between EPS and PRI was found at the leaf (r² = 0.81) and canopy level (r² = 0.41). Airborne-derived crown PRI calculated from the imagery acquired during the fruit growth was related to the ratio of the total soluble solids normalized by the tritatable acidity (TSS/TA), an indicator of fruit quality measured on the same trees, yielding a coefficient of determination of r² = 0.50. The relationship between the integral of PRI time-series and TSS/TA yielded a coefficient of determination of r² = 0.72 (peach) and r² = 0.61 (nectarines). On the contrary, the relation between TSS/TA and the time-series of crown thermal imagery was very weak (r² = 0.21 and 0.25 respectively). These results suggest that a physiological remote sensing indicator related to photosynthesis, such as PRI, is more appropriate for fruit quality assessment than crown temperature, the established method of water stress detection, which is more related to crown transpiration. A radiative transfer modelling study was conducted to assess the potential validity of this methodology for fruit quality assessment when using medium spatial resolution imagery. The analysis shows important effects of soil and shadows on the PRI vs EPS relationship used for fruit quality assessment if non-pure crown reflectance was extracted from the imagery.     

Mapping tree distribution and LAI in peatlands using field methods and imaging spectroscopy: a case study for the Haaksbergerveen,the Netherlands

Leaf area index (LAI) is among the vegetation parameters that play an important role in climate, hydrological and ecological studies, and is used for assessing growth and expansion of vegetation. The main objective of this study was to develop a methodology to map the LAI distribution of birch trees (Betula pendula) in peatland ecosystems using field-based instruments and airborne-based remote-sensing techniques. The developed mapping method was validated using field-based LAI measurements using the LAI-2000 instrument. First vegetation indices, including simple ratio (SR), normalized difference vegetation index (NDVI), and reduced simple ratio (RSR), were derived from HyMap data and related to ground-based measurements of LAI. LAI related better with RSR (R² = 0.68), followed by NDVI (R² = 0.63) and SR (R² = 0.58), respectively. Areas with birch were identified using Spectral Angle Mapper (SAM) to classify the image into 11 end members of dominant species including bare soil and open water. Next, the relationship between LAI and RSR was applied to areas with birch, yielding a birch LAI map. Comparison of the map of the birch trees and field-based LAI data was done using linear regression, yielding an R² = 0.38 and an RMSE = 0.25, which is fairly accurate for a structurally highly diverse field situation. The method may prove an invaluable tool to monitor tree encroachment and assess tree LAI in these remote and poorly accessible areas.     

A near-infrared narrow-waveband ratio to determine Leaf Area Index in orchards

An innovative index for LAI estimation over orchard canopies based on the combination of two near-infrared narrow wavebands (1050 nm and 1250 nm) is presented. The index is insensitive to physiological changes on the leaf level even under changing pheneological developmental stage. In the meantime, the index is highly sensitive for variations at the canopy level. Values of LAI estimated by means of this newly developed index, called sLAIDI (standardized LAI Determining Index), resulted in excellent agreements with in-situ measured LAI values of citrus trees, with determination coefficients (R²) of 0.83. Based on in-situ measured data and simulated data (PROSPECT and ACRM), the newly developed sLAIDI was also found to perform outstanding at high LAI values in contrary to well-known LAI related indices such as the Normalized Difference Vegetation Index (NDVI) and the Modified Chlorophyll Ratio Index 2 (MCARI2) that saturate at LAI values of 5 and 8 respectively. Moreover, the sLAIDI index was independent of changing pigment contents. Consequently, this index was regarded a valuable tool to more accurately extract biochemical parameters after normalizing the considered spectral data for LAI. By implementing sLAIDI in existing chlorophyll-related indices (e.g., GM index), the performance to estimate the leaf chlorophyll content from spectral citrus and peach canopy level data is increased (R² increases from 0.68 to 0.77). Moreover, a correction factor was proposed to enable the application of sLAIDI on fields with strong variations of water content.     

Estimating standing biomass in papyrus (Cyperus papyrus L.) swamp: exploratory of in situ hyperspectral indices and random forest regression

Accurate estimates of papyrus (Cyperus papyrus) biomass are critical for an efficient papyrus swamp monitoring and management system. The objective of this study was to test the utility of random forest (RF) regression and two narrow-band vegetation indices in estimating above-ground biomass (AGB) for complex and densely vegetated swamp canopies. The normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) were calculated from field spectrometry data and fresh AGB was measured in 82 quadrats at three different areas in the iSimangaliso Wetland Park, South Africa. NDVI was calculated from all possible band combinations of the electromagnetic spectrum (350 and 2500 nm), while EVI was calculated from possible band combinations in the blue, red, and near infrared of the spectrum. Backward feature elimination and RF regression were used as variable selection and modelling techniques to predict papyrus AGB. Results showed that the effective portions of electromagnetic spectrum for estimation AGB of papyrus swamp were located within the blue, red, red-edge, and near-infrared regions. The three best selected EVIs were computed from bands located at (i) 445, 682, and 829 nm, (ii) 497, 676, and 1091 nm, and (iii) 495, 678, and 1120 nm. These indices produced better predictive accuracies (R² = 0.90; root mean square error of prediction (RMSEP) = 0.289 kg m^?2; 7.99% of the mean) than the best selected NDVIs (R² = 0.85; RMSEP = 0.343 kgm^?2; 9.49% of the mean) that were calculated from bands located at (i) 739 and 829 nm, (ii) 739 and 814 nm, (iii) 744 and 789 nm, and (iv) 734 and 909 nm. The results of the present study demonstrate the utility of narrow-band vegetation indices (EVI and NDVI) and RF regression in estimating papyrus AGB at high density, a previously challenging task with broadband satellite sensors.     

Interpretation and topographic compensation of conifer canopy self-shadowing

The self-shadowing of conifer canopies results from the size and arrangement of trees within a stand and is a first-order term controlling radiance from forested terrain at common pixel scales of tens of meters. Although self-shadowing is a useful attribute for forest remote-sensing classification, compensation for the topographic effects of self-shadowing has proven problematic. This study used airborne canopy LiDAR measurements of 80 Pacific Northwest, USA conifer stands ranging in development stage from pre-canopy closure to old-growth in order to model canopy self-shadowing for four solar zenith angles (SZA). The shadow data were compared to physical measurements used to characterize forest stands, and were also used to test and improve terrain compensation models for remotely sensed images of forested terrain. Canopy self-shadowing on flat terrain strongly correlates with the canopy's geometric complexity as measured by the rumple index (canopy surface area/ground surface area) (R² = 0.94–0.87 depending on SZA), but is less correlated with other stand measurements: 95th percentile canopy height (R² = 0.68), mean diameter at breast height (dbh) (R² = 0.65), basal area ha^? 1 (R² = 0.18), and canopy stem count ha^? 1 (R² = 0.18). The results in this paper support interpretation of self-shadowing as a function of canopy complexity, which is an important ecological characteristic in its own right. Modeling of canopy self-shadowing was used to assess the accuracy of the Sun-Canopy-Sensor (SCS) topographic correction, and to develop a new empirical Adaptive Shade Compensation (ASC) topographic compensation model. ASC used measured shadow (as an estimate of canopy complexity) and the SCS term (to describe the illumination geometry) as independent variables in multiple regressions to determine the topographic correction. The ASC model provided more accurate radiance corrections with limited variation in results across the full range of canopy complexities and incidence angles.     

Estimating northern peatland CO2 exchange from MODIS time series data

Studies using satellite sensor-derived data as input to models for CO₂ exchange show promising results for closed forest stands. There is a need for extending this approach to other land cover types, in order to carry out large-scale monitoring of CO₂ exchange. In this study, three years of eddy covariance data from two peatlands in Sweden were averaged for 16-day composite periods and related to data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and modeled photosynthetic photon flux density (PPFD). Noise in the time series of MODIS 250 m vegetation indices was reduced by using double logistic curve fits. Smoothed normalized difference vegetation index (NDVI) showed saturation during summertime, and the enhanced vegetation index (EVI) generally gave better results in explaining gross primary productivity (GPP). The strong linear relationships found between GPP and the product of EVI and modeled PPFD (R² = 0.85 and 0.76) were only slightly stronger than for the product of EVI and MODIS daytime 1 km land surface temperature (LST) (R² = 0.84 and 0.71). One probable reason for these results is that several controls on GPP were related to both modeled PPFD and daytime LST. Since ecosystem respiration (ER) was largely explained by diurnal LST in exponential relationships (R² = 0.89 and 0.83), net ecosystem exchange (NEE) was directly related to diurnal LST in combination with the product of EVI and modeled PPFD in multiple exponential regressions (R² = 0.81 and 0.73). Even though the R² values were somewhat weaker for NEE, compared to GPP and ER, the RMSE values were much lower than if NEE would have been estimated as the sum of GPP and ER. The overall conclusion of this study is that regression models driven by satellite sensor-derived data and modeled PPFD can be used to estimate CO₂ fluxes in peatlands.