Artificial neural network and response surface methodology: a comparative analysis for optimizing rice straw pretreatment and saccharification

Abstract

The present study demonstrates a comparative analysis between the artificial neural network (ANN) and response surface methodology (RSM) as optimization tools for pretreatment and enzymatic hydrolysis of lignocellulosic rice straw. The efficacy for both the processes, that is, pretreatment and enzymatic hydrolysis was evaluated using correlation coefficient (R²) & mean squared error (MSE). The values of R² obtained by ANN after training, validation, and testing were 1, 0.9005, and 0.997 for pretreatment and 0.962, 0.923, and 0.9941 for enzymatic saccharification, respectively. On the other hand, the R² values obtained with RSM were 0.9965 for cellulose recovery and 0.9994 for saccharification efficiency. Thus, ANN and RSM together successfully identify the substantial process conditions for rice straw pretreatment and enzymatic saccharification. The percentage of error for ANN and RSM were 0.009 and 0.01 for cellulose recovery and for 0.004 and 0.005 for saccharification efficiency, respectively, which showed the authority of ANN in exemplifying the non-linear behavior of the system.     

Biodegradation Kinetic Studies on Phenol in Internal Draft Tube (Inverse Fluidized Bed) Biofilm Reactor Using Pseudomonas fluorescens: Performance Evaluation of Biofilm and Biomass Characteristics

The bioremediation potential of Pseudomonas fluorescens was studied in an internal draft tube (inverse fluidized bed) biofilm reactor (IDTBR) under batch recirculation conditions using synthetic phenol of various concentrations (400, 600, 800, 1000, and 1200 mg/L). The performance of IDTBR was investigated and the characteristics of biomass and biofilm were determined by evaluating biofilm dry density and thickness, bioparticle density, suspended and attached biomass concentration, chemical oxygen demand, and phenol removal efficiency. Biodegradation kinetics had been studied for the suspended biomass culture and biofilm systems. Suspended biomass followed substrate inhibition kinetics, and the experimental data fitted well with the Haldane model. The correlation coefficient, R ², and root-mean-square error (RMSE) obtained for the Haldane model with respect to specific growth rate were .9389 and .00729, respectively, and with respect to specific phenol consumption rate were .9259 and .00972, respectively. It was also observed experimentally that biofilm overcame substrate inhibition effect and fitted the same to the Monod model (R ² = .9831, RMSE = .00884 for specific growth rate and R ² = .9686, RMSE = .00912 for specific phenol consumption rate).     

Modeling daily chlorophyll a dynamics in a German lowland river using artificial neural networks and multiple linear regression approaches

Phytoplankton biomass is an important indicator for water quality, and predicting its dynamics is thus regarded as one of the important issues in the domain of river ecology and management. However, the vast majority of models in river systems have focused mostly on flow prediction and water quality with very few applications to biotic parameters such as chlorophyll a (Chl a). Based on a 1.5-year measured dataset of Chl a and environmental variables, we developed two modeling approaches [artificial neural networks (ANN) and multiple linear regression (MLR)] to simulate the daily Chl a dynamics in a German lowland river. In general, the developed ANN and MLR models achieved satisfactory accuracy in predicting daily dynamics of Chl a concentrations. Although some peaks and lows were not predicted, the predicted and the observed data matched closely by the MLR model with the coefficient of determination (R ²), Nash–Sutcliffe efficiency (NS), and the root mean square error (RMSE) of 0.53, 0.53, and 2.75 for the calibration period and 0.63, 0.62, and 1.94 for the validation period, respectively. Likewise, the results of the ANN model also illustrated a good agreement between observed and predicted data during calibration and validation periods, which was demonstrated by R ², NS, and RMSE values (0.68, 0.68, and 2.27 for the calibration period, 0.55, 0.66 and 2.12 for the validation period, respectively). According to the sensitivity analysis, Chl a concentration was highly sensitive to dissolved inorganic nitrogen, nitrate–nitrogen, autoregressive Chl a, chloride, sulfate, and total phosphorus. We concluded that it was possible to predict the daily Chl a dynamics in the German lowland river based on relevant environmental factors using either ANN or MLR models. The ANN model is well suited for solving non-linear and complex problems, while the MLR model can explicitly explore the coefficients between independent and dependent variables. Further studies are still needed to improve the accuracy of the developed models.     

Estimating fish community diversity from environmental features in the Tagus estuary (Portugal): Multiple Linear Regression and Artificial Neural Network approaches

Relationships between environmental variables and diversity (Shannon‐Weaver index) of the fish communities in the Tagus estuary and adjacent coastal areas were analyzed. The focus was on the linearity or nonlinearity of these abiotic/biotic characteristics, with the aim to obtain an accurate short–medium term time‐scale diversity prediction from habitat variables alone. Multiple Linear Regressions (MLR) were used for the linear approach and Artificial Neural Networks (ANNs) for the nonlinear approach. MLR results in the external validation phase indicated a lack of model accuracy (R² = 0.0710; %SEP = 47.5868; E = ?0.0217; ARV = 1.0217; N = 43). Results of the best of the Artificial Neural Networks used in this study (12‐15‐15‐1 architecture) in the external validation phase (ANN: R² = 0.9736; %SEP = 7.8499; E = 0.9722; ARV = 0.0278; N = 43) were more accurate than those obtained with MLR. This indicates a clear nonlinear relationship between variables. In the best ANN model, nitrate concentration, depth, dissolved oxygen and temperature were the most important predictors of fish diversity in the Tagus estuary. The sensibility analysis indicated that the remaining variables (silicate, nitrite, transparency, salinity, slope, phosphate, water particulate organic matter, and chlorophyll a) played lesser roles in the model.     

Artificial neural networking to estimate the leaf area for invasive plant Wedelia trilobata

Leaf area are very important parameter for the understanding of growth and physiological responses of invasive plant species under different environmental factors. This study was conducted to build non-destructive leaf area model of Wedelia trilobata that were grown in greenhouse. Regression analysis and artificial neural network (ANN) approaches were used for the development of leaf area model with the help of leaf length and width of 262 plants samples. In selection of best method under both techniques, the lower value of mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) and higher value of R² were considered. According to the results it was found that ANN have higher value of (R² = 0.96) and lower value of error (MAE = 0.023, RMSE = 0.379, MAPE = 0.001) than regression analysis (R² = 0.94, MAE = 0.111, RMSE = 1.798, MAPE = 0.0005). It was concluded that error between predicted and actual value was less under ANN. Therefore, ANN model approach can be used as an alternating method for the estimation of leaf area. Through estimation of leaf area, invasive plant growth can predict under different environment conditions.     

Statistical optimization of the phytoremediation of arsenic by Ludwigia octovalvis- in a pilot reed bed using response surface methodology (RSM) versus an artificial neural network (ANN)

In this study, the removal of arsenic (As) by plant, Ludwigia octovalvis, in a pilot reed bed was optimized. A Box-Behnken design was employed including a comparative analysis of both Response Surface Methodology (RSM) and an Artificial Neural Network (ANN) for the prediction of maximum arsenic removal. The predicted optimum condition using the desirability function of both models was 39 mg kg^?1 for the arsenic concentration in soil, an elapsed time of 42 days (the sampling day) and an aeration rate of 0.22 L/min, with the predicted values of arsenic removal by RSM and ANN being 72.6% and 71.4%, respectively. The validation of the predicted optimum point showed an actual arsenic removal of 70.6%. This was achieved with the deviation between the validation value and the predicted values being within 3.49% (RSM) and 1.87% (ANN). The performance evaluation of the RSM and ANN models showed that ANN performs better than RSM with a higher R² (0.97) close to 1.0 and very small Average Absolute Deviation (AAD) (0.02) and Root Mean Square Error (RMSE) (0.004) values close to zero. Both models were appropriate for the optimization of arsenic removal with ANN demonstrating significantly higher predictive and fitting ability than RSM.     

Evaluation of hawthorns maturity level by developing an automated machine learning-based algorithm

Marketability of agricultural products depends heavily on appearance attributes such as color, size, and ripeness. Sorting plays an important role in increasing marketability by separating crop classes according to appearance attributes, thus reducing waste. As an expert technique, image processing and artificial intelligence (AI) techniques have been applied to classify hawthorns based on maturity levels (unripe, ripe, and overripe). A total of 600 hawthorns were categorized by an expert and the images were taken by an imaging box. The geometric properties, color and, texture features were extracted from segmented hawthorns using the Gray Level Co-occurrence Matrix (GLCM) and evaluation of various color spaces. The efficient feature vector was created by QDA feature reduction method and then classified using two classical machine learning algorithms: Artificial Neural Network (ANN) and Support Vector Machine (SVM). The obtained results indicated that the efficient feature-based ANN model with the configuration of 14–10-3 resulted in the accuracy of 99.57, 99.16, and 98.16% and the least means square error (MSE) of 1 × 10⁻³, 8 × 10⁻³, and 3 × 10⁻³ for training, validation and test phases, respectively. The machine vision system combined with the machine learning algorithms can successfully classify hawthorns according to their maturity levels.     

Impacts of mussel (<Emphasis Type="Italic">Mytilus galloprovincialis</Emphasis>) farming on oxygen consumption and nutrient recycling in a eutrophic coastal lagoon

Fluxes of oxygen, nitrogen and phosphorus were determined in two areas of the Sacca di Goro lagoon, at a site influenced by the farming of the mussel Mytilus galloprovincialis and a control site. Mussel farming induced intense biodeposition of organic matter to the underlying sediments, which stimulated sediment oxygen demand, and inorganic nitrogen and phosphorus regeneration rates compared to the nearby control station. Overall benthic fluxes (–11.4 ± 6.5 mmol O₂ m⁻² h⁻¹; 1.59 ± 0.47 mmol NH₄⁺ m⁻² h⁻¹ and 94 ± 42 μmol PO₄³⁻ m⁻² h⁻¹) at the mussel farm are amongst the highest ever recorded for an aquaculture impacted area and question the belief that farming of filter-feeding bivalves has inherently lower impacts than finfish farming. In situ incubations of intact mussel ropes demonstrated that the mussel rope community was an enormous sink for oxygen and particulate organic matter, and an equally large source of dissolved inorganic nitrogen and phosphate to the water column. Overall, a one meter square area of␣mussel farm (mussel ropes and underlying sediment) was estimated to have an oxygen demand of 46.8 mmol m² h⁻¹ and to regenerate inorganic nitrogen and phosphorus at rates of 8.5 and 0.3 mmol m² h⁻¹, with the mussel ropes accounting for between 70 and more than 90% of the overall oxygen and nutrient fluxes. Even taking into account that within the farmed area of the Sacca di Goro lagoon, there are 15–20 m⁻² of open water for each one covered with mussel ropes, the mussel ropes would account for a large and often dominant part of overall oxygen and nutrient fluxes. These results demonstrate that it is essential to take into account the activity of the cultivated organisms and their epiphytic community when assessing the impacts of shellfish farming. Overall, whilst grazing by the mussel rope community could act as a top-down control on the phytoplankton, most of the ingested organic matter is rapidly recycled to the water column as inorganic nutrients, which would be expected to stimulate phytoplankton growth. Consequently, the net effect of the mussel farming on phytoplankton dynamics, may be to increase phytoplankton turnover and overall production, rather than to limit phytoplankton biomass.     

Influence of sediment organic enrichment and water alkalinity on growth of aquatic isoetid and elodeid plants

1. Lake eutrophication has increased phytoplankton blooms and sediment organic matter. Among higher plants, small, oligotrophic rosette species (isoetids) have disappeared, while a few tall, eutrophic species (elodeids) may have persisted. Despite recent reduction of nutrient loading in restored lakes, the vegetation has rarely regained its former composition and coverage. Patterns of recovery may depend on local alkalinity because HCO₃^? stimulates photosynthesis of elodeids and not of isoetids. In laboratory growth experiments with two isoetids (Lobelia dortmanna and Littorella uniflora) and two elodeids (Potamogeton crispus and P. perfoliatus), we test whether organic enrichment of lake sediments has a long‐lasting influence by: (i) reducing plant growth because of oxygen stress on plant roots and (ii) inhibiting growth more for isoetids than elodeids. We also test whether (iii) increasing alkalinity (from 0.17 to 3.20 meq. L^?1) enhances growth and reduces inhibition of organic sediment enrichment for elodeids but not for isoetids. 2. In low organic sediments, higher oxygen release from roots of isoetids than elodeids generated oxic conditions to greater sediment depth for Lobelia (4.3 cm) and Littorella (3.0 cm) than for Potamogeton species (1.6–2.2 cm). Sediment oxygen penetration depth fell rapidly to 0.4–1.0 cm for all four species at even modest organic enrichment and oxygen consumption in the sediments. Roots became shorter and isoetid roots became thicker to better supply oxygen to apical meristems. 3. Growth of elodeids was strongly inhibited across all levels of organic enrichment of sediments being eight‐fold lower at the highest enrichment compared to the unenriched control. Leaf biomass of isoetids increased three‐fold by moderate organic enrichment presumably because of greater CO₂ supply from sediments being their main CO₂ source. At higher organic enrichment, isoetid biomass was reduced, leaf chlorophyll declined up to 10‐fold, root length declined from 7 to <2 cm and mortality rose (up to 50%) signalling high plant stress. 4. Lobelia was not affected by HCO₃^? addition in accordance with its use of sediment CO₂. Biomass of elodeids increased severalfold by rising alkalinity from 0.17 to 3.20 meq. L^?1 in accordance with their use of HCO₃^? for photosynthesis, while the negative impact of organically enriched sediments remained. 5. Overall, root development of all four species was so strongly restricted in sediments enriched with labile organic matter that plants if growing in situ may lose root anchorage. Other experiments demonstrate that this risk is enhanced by greater water content and reduced consolidation in organically rich sediments. Therefore, formation of more muddy and oxygen‐demanding sediments during eutrophication will impede plant recovery in restored lakes while high local alkalinity will help elodeid recovery.     

The response of Chironomidae (Diptera) to a long-term exclusion of terrestrial organic matter

We examined the effects of a seven-year detrital exclusion on chironomid assemblages in an Appalachian headwater stream. We hypothesized that litter exclusion would lead to a reduction in all chironomids at both the subfamily and generic levels because organic matter serves as both food and habitat in these headwater streams. Tanytarsini total abundance and biomass significantly declined after litter exclusion. Before litter exclusion, Tanytarsini average abundance was 4271 ± 1135 S.E. m⁻² and 625 ± 98 after litter exclusion. Biomass was 3.57 ± 0.96 mg AFDM m⁻² before litter exclusion and 1.03 ± 0.9 after exclusion. In contrast, Orthocladiinae abundance and biomass did not change because a psammanophilic chironomid, Lopescladius sp., and other Orthocladiinae genera did not decline significantly. Overall chironomid taxa richness and diversity did not change as a result of litter exclusion. However, Canonical Correspondence Analysis (CCA) of genus-level biomass did show a clear separation between the litter exclusion stream and a reference stream. Separation of taxa between the two streams was due to differences in fine (r ² = 0.39) and coarse (r ² = 0.36) organic matter standing stocks and the proportion of small inorganic substrates (r ² = 0.39) present within a sample. As organic matter declined in the litter exclusion stream, overall chironomid biomass declined and the chironomid community assemblage changed. Tanytarsini were replaced by Orthocladiinae in the litter exclusion stream because they were better able to live and feed on biofilm associated with inorganic substrates. Handling editor: K. Martens     

Effects of organic matter availability on the life history and production of a top vertebrate predator (Plethodontidae: Gyrinophilus palleucus) in two cave streams

1. Surface ecosystems provide the primary source of organic matter to many cave communities. Variation in the strength of connectivity to the surface suggests that some caves may be more resource‐limited than others. To test this, we examined diet, prey availability and production of an obligate cave salamander Gyrinophilus palleucus (Plethodontidae), a top predator, in two south‐eastern U.S.A. caves with different levels of organic matter (Tony Sinks cave, 165 g AFDM m^?2; Bluff River cave, 62 g AFDM m^?2). 2. We quantified density, biomass, growth rate, production and diet of G. palleucus monthly for 21 months. Diet composition, differences in prey communities and seasonal patterns in prey consumption were also analysed. 3. Salamander density, biomass and secondary production were significantly greater in the high organic matter cave (0.10 m^?2, 0.18 g AFDM m^?2, 0.12 g AFDM m^?2 year^?1) than in the low organic matter cave (0.03 m^?2, 0.03 g AFDM m^?2, 0.01 g AFDM m^?2 year^?1). Although growth rates were not statistically different between the two cave salamander populations, low recaptures probably influenced this result. 4. Isopoda prey were the major contributor to salamander production in the high organic matter cave (69%). In the low organic matter cave, production was provided by isopods (41%) and oligochaetes (20%). The lower number of prey taxa contributing to salamander production in the high organic matter cave suggests the ability to forage more selectively. 5. The differences in foraging strategy, density, biomass and secondary production were probably related to differences in the strength of surface connectivity, which controls organic matter supply. Links between basal resource level and top predator performance show the importance of bottom‐up limitation in the food webs of caves and other detritus‐based ecosystems.     

Modeling of polygalacturonase enzyme activity and biomass production by <Emphasis Type="Italic">Aspergillus sojae</Emphasis> ATCC 20235

Aspergillus sojae, which is used in the making of koji, a characteristic Japanese food, is a potential candidate for the production of polygalacturonase (PG) enzyme, which of a major industrial significance. In this study, fermentation data of an A. sojae system were modeled by multiple linear regression (MLR) and artificial neural network (ANN) approaches to estimate PG activity and biomass. Nutrient concentrations, agitation speed, inoculum ratio and final pH of the fermentation medium were used as the inputs of the system. In addition to nutrient conditions, the final pH of the fermentation medium was also shown to be an effective parameter in the estimation of biomass concentration. The ANN parameters, such as number of hidden neurons, epochs and learning rate, were determined using a statistical approach. In the determination of network architecture, a cross-validation technique was used to test the ANN models. Goodness-of-fit of the regression and ANN models was measured by the R ² of cross-validated data and squared error of prediction. The PG activity and biomass were modeled with a 5-2-1 and 5-9-1 network topology, respectively. The models predicted enzyme activity with an R ² of 0.84 and biomass with an R ² value of 0.83, whereas the regression models predicted enzyme activity with an R ² of 0.84 and biomass with an R ² of 0.69.     

Influence of natural amphipod (<Emphasis Type="Italic">Victoriopisa</Emphasis><Emphasis Type="Italic">australiensis</Emphasis>) (Chilton, 1923) population densities on benthic metabolism,nutrient fluxes,denitrification and DNRA in sub-tropical estuarine sediment

The influence of natural populations of the sub-surface deposit-feeding amphipod Victoriopisa australiensis on sediment biogeochemistry was assessed by randomly collecting 21 sediment cores in a zone of Coombabah Lake, southern Moreton Bay, Australia, where the benthic infauna was dominated by this species. Cores were incubated sequentially to determine sediment–water column fluxes of oxygen, dissolved inorganic carbon and inorganic N species, followed by incubations to determine rates of denitrification and dissimilatory nitrate reduction to ammonium (DNRA) using the isotope pairing technique. Finally, each core was sieved in order to determine the population and biomass of amphipods present. Whilst all measures of overall benthic metabolism (sediment oxygen demand, and effluxes of inorganic carbon and nitrogen) showed increased with amphipod density, with rates being stimulated 70–220% at the highest categorised density range of 2,500–3,500 ind m⁻², only the correlation with dissolved inorganic carbon was statistically significant. In contrast, there were no discernable trends between amphipod densities and any of the N-cycle processes with the slopes of all correlations being very close to zero. These results highlight the differences in mesocosm simulations of fauna effects, which primarily relate to shifts in rates of organic matter turnover, compared to natural sediments where fauna effects relate more to induced changes in rates of organic matter deposition. Therefore, while mesocosms represent a powerful tool to investigate the mechanisms by which fauna influences microbial metabolism in the sediment, only studies of natural sediments can determine to what extent these mechanisms function in situ. Handling editor: Pierluigi Viaroli     

The role of plankton biomass in controlling fluctuations of suspended matter in Lake Constance

The concentrations of particulate matter, expressed as dry weight (DW), particulate organic (POM), and inorganic material were measured at regular intervals in Lake Constance between February 1980 and December 1982. Maximum particle concentrations were recorded for the euphotic zone in summer (7 mg l⁻¹), while minima occurred during the early summer and in winter. Annual mean concentrations of DW within the entire water column varied between 0.6 and 0.7 mg l⁻¹. In the euphotic zone nearly 70% of DW is organic material. The inorganic particles originate either from phytoplankton (diatomaceous silicon, biogenic decalcification) or from the tributaries. Although phytoplankton biomass only comprises a relatively small proportion (i.e. 30% at maximum) of organic matter, it is the primary source of POM. Therefore, seasonal variations in phytoplankton control epilimnetic concentrations of POM in Lake Constance. Inorganic material comprises smaller proportions of suspended matter. Seasonal variations are related predominantly to fluctuations in biomass and therefore particulate inorganic matter is suggested to originate mainly from autochthonous sources. At the sampling station concentrations of inorganic particles supported by the main tributary, the Alpenrhein, only occasionally vary concomitantly with runoff.     

基于HJ1B和ALOS/PALSAR数据的森林地上生物量遥感估算

森林地上生物量的精确估算能够减小碳储量估算的不确定性。为了探寻一种有效地提高森林生物量估算精度的方法,探讨了基于遥感物理模型和经验统计模型估算山地森林地上生物量的方法。首先,基于Li-Strahler几何光学模型和多元前向模式(MFM)进行模型模拟,结合查找表算法(LUT)从多光谱图像HJ1B估算贺兰山研究区的森林地上生物量。其次,采用统计方法建立了2种回归模型:(1)多光谱图像HJ1B进行混合像元分解(SMA),并与雷达图像ALOS/PALSAR进行图像融合建立生物量回归模型;(2)雷达图像ALOS/PALSAR后向散射系数和实测生物量建立了生物量回归模型。用实测数据对3种算法估算结果进行精度验证。研究结果表明:采用几何光学模型和MFM算法估算的森林地上生物量精度最好(决定系数R2=0.61,均方根误差RMSE=8.33 t/hm2,P0.001),其估算地上生物量与实测值一致性较好,估算生物量精度略优于SMA估算的精度(R2=0.60,RMSE=9.417 t/hm2);ALOS/PALSAR多元回归估算的精度最差(R2=0.39,RMSE=14.89 t/hm2)。由此可见,采用几何光学模型和混合像元分解SMA适合估算森林地上生物量,利用这2种方法进行森林地上生物量遥感监测研究具有一定的应用潜力。     

星载大光斑LiDAR与HJ-1A高光谱数据联合估测区域森林地上生物量

以吉林省汪清林业局经营区为研究区,利用HJ-1A/HSI高光谱数据和ICESat-GLAS波形数据,估测区域森林地上生物量。从平滑后的GLAS波形数据中提取波形长度W和地形坡度参数TS,建立GLAS森林最大树高估测模型;从GLAS波形数据中提取能量参数I(植被回波能量Ev和回波总能量E之比),建立GLAS森林郁闭度估测模型;利用GLAS估测的森林最大树高和森林郁闭度联合建立森林地上生物量模型。由于GLAS呈离散条带状分布,无法实现区域估测,因此研究将GLAS波形数据与HJ-1A/HSI高光谱数据联合,基于支持向量回归机算法实现森林地上生物量区域估测,得到研究区森林地上生物量分布图。研究结果显示,基于W和TS建立的GLAS森林最大树高估测模型的adj.R~2=0.78,RMSE=2.51m,模型验证的adj.R~2=0.85,RMSE=1.67m。地形坡度参数TS能够有效的降低地形坡度的影响;当林下植被高度为2m时,得到的基于参数I建立的GLAS森林郁闭度估测模型效果最好,模型的adj.R~2=0.64,RMSE=0.13,模型验证的adj.R~2=0.65,RMSE=0.12。利用森林最大树高和森林郁闭度建立的森林地上生物量模型的adj.R~2=0.62,RMSE=10.88 t/hm~2,模型验证的adj.R~2=0.60,RMSE=11.52 t/hm~2。基于支持向量回归机算法,利用HJ-1A/HSI和GLAS数据建立的森林地上生物量SVR模型,生成了森林地上生物量分布图,利用野外数据对得到的分布图进行验证,验证结果显示森林地上生物量估测值与实测值存在很强的线性关系(adj.R~2=0.62,RMSE=11.11 t/hm~2),能够满足林业应用的需要。因此联合ICESat-GLAS波形数据与HJ-1A高光谱数据,能够提高区域森林地上生物量的估测精度。     

Optimization of operational conditions for adipate ester synthesis in a stirred tank reactor

Esterification of adipic acid and oleyl alcohol in a solvent-free system featuring a stirred tank reactor containing commercially immobilized Candida antarctica lipase B was performed. The process was carried out using an artificial neural network (ANN) trained by the Levenberg-Marquardt (LM) algorithm. The effects of four operative variables, temperature, time, amount of enzyme, and impeller speed, on the reaction yield were studied. By examining different ANN configurations, the best network was found to consist of seven hidden nodes using a hyperbolic tangent sigmoid transfer function. The values of the coefficient of determination (R²) and root mean squared error (RMSE) between the actual and predicted responses were determined to be 1 and 0.0058178 for training and 0.99467 and 0.622540 for the testing datasets, respectively. These results imply that the developed model was capable of predicting the esterification yield. The operative variables affected the yield, and their order of contribution was as follows: time > amount of enzyme > temperature > impeller speed. A high percentage of yield (95.7%) was obtained using a low level of enzyme (2.5% w/w), and the temperature, time, and impeller speed were 66.5°C, 354 min (about 6 h), and 500 rpm, respectively. A simple protocol for efficient substrate conversion in a solvent-free system evidenced by high enzyme stability is indicative of successful ester synthesis.     

Bacterial Growth on Allochthonous Carbon in Humic and Nutrient-enriched Lakes: Results from Whole-Lake 13C Addition Experiments

Organic carbon (C) in lakes originates from two distinct sources—primary production from within the lake itself (autochthonous supply) and importation of organic matter from the terrestrial watershed (allochthonous supply). By manipulating the ¹³C of dissolved inorganic C, thereby labeling within-lake primary production, we examined the relative importance of autochthonous and allochthonous C in supporting bacterial production. For 35 days, NaH¹³CO₃ was added daily to two small, forested lakes. One of the lakes (Peter) was fertilized so that primary production exceeded total respiration in the epilimnion. The other lake (Tuesday), in contrast, was low in productivity and had high levels of colored dissolved organic C (DOC). To obtain bacterial C isotopes, bacteria were regrown in situ in particle-free lake water in dialysis tubes. The contribution of allochthonous C to bacterial biomass was calculated by applying a two-member mixing model. In the absence of a direct measurement, the isotopic signature of the autochthonous end-member was estimated indirectly by three different approaches. Although there was excess primary production in Peter Lake, bacterial biomass consisted of 43–46% allochthonous C. In Tuesday Lake more than 75% of bacterial growth was supported by allochthonous C. Although bacteria used autochthonous C preferentially over allochthonous C, DOC from the watershed contributed significantly to bacterial production. In combination with results from similar experiments in different lakes, our findings suggest that the contribution of allochthonous C to bacterial production can be predicted from ratios of chromophoric dissolved organic matter (a surrogate for allochthonous supply) and chlorophyll a (a surrogate for autochthonous supply).