Improved Model Structure Research Articles

Radiograph recognition of COVID-19 based on DenseNet

This paper presents a neural network approach for pneumonia image classification. Utilizing thousands of actual X-ray images from both pneumonia patients and healthy individuals, a classification model is developed. This neural network model, a prevalent machine learning technique, is employed as an adjunct tool to aid pneumonia diagnosis and pre-screening. The algorithms utilized in the study are comprehensively analyzed and detailed. Through an examination of the network's architecture, several effective methods for enhancing the model are proposed, leading to improved model structure and classification performance. The novel model structure proposed in this study is demonstrated and evaluated. The final training accuracy exceeds 95%, demonstrating the model's proficiency in learning from the training data. Moreover, the test accuracy hovers around 92%, indicating the model's robust performance in generalizing to unseen data.

Measuring Morality: An Examination of the Moral Foundation Questionnaire's Factor Structure.

Moral foundations theory proposes five domains of morality-harm, fairness, loyalty, purity, and authority. Endorsement of these moral domains is assessed by the Moral Foundations Questionnaire (MFQ), a 30-item scale that has undergone intense measurement scrutiny. Across five samples (N = 464,229), we show greatly improved model fit using a Bifactor model that accounts for two kinds of items in the MFQ: judgment and relevance. We add to this space by demonstrating how using this improved measurement structure changes the strength of correlations of the moral foundations with numerous attitudes, cognitive styles, and moral decision-making. Future research should continue to identify what, if anything, the relevance and judgment factors might substantively capture over and above the substantive domains of moral foundations. In the meantime, we recommend that researchers use the Bifactor model for its improved model structure, rather than dropping the relevant items as some have proposed.

Improved model structure of direct contact membrane distillation for saline water purification

A pilot plant comprising membrane distillation is tested for coordinated inlet temperature for the feed and cold streams with a fixed difference. Selected inlet temperature differences of 30, 40, and 50 °C were examined. It is found that water production increases with both temperature difference and inlet feed temperature. Specifically, the water production at a temperature difference of 50 °C can be 13%, and 30% higher than that at a temperature difference of 40, and 30 °C, respectively. The measurements of the mass flux were used to validate the internal mass and heat transfer equations which proved to provide an excellent prediction of the water production with up to 3.7% average error. The measurements of the exit temperature of the brine and permeate streams were used to confirm the macroscale heat balance equations. They revealed a considerable discrepancy between the heat supplied by the feed stream and the heat transferred between the two channels as well as between the heat gained by the cold stream and the heat supplied. The first discrepancy is ascribed to the existence of environmental heat loss at the feed side. The second discrepancy is attributed to the existence of environmental heat loss and the contribution of the heat of condensation to additionally warm up the cold stream.

Num2vec: Pre-Training Numeric Representations for Time Series Forecasting in the Sensing System

Time series forecasting in the sensing system aims to predict future values based on historical records that sensors have collected. Previous works, however, usually focus on improving model structure or algorithm for better performance but the perspective of learning proper numeric representations is overlooked. The inappropriate and coarse numeric representations are not expressive enough to capture the intrinsic characteristics of numbers, which will obviously degrade the prediction performance. In this article, we propose Num2vec, an algorithmic framework to learn numeric representations. Specifically, Num2vec lists three main logic characteristics of numbers: arithmetic, direction, and periodicity. By representing numbers into a transition space, Num2vec can translates numbers agilely to different Internet of Things tasks through selecting the corresponding characteristics. According to such a design, Num2vec enjoys flexible numeric representations to fit different Internet of Things time series tasks. Extensive experiments on four real-world datasets show that the approach achieves the best performance when compared with state-of-the-art baselines.

Comparing Random Forest to Bayesian Networks as nitrogen management decision support systems

AbstractNitrogen (N) is notoriously difficult to manage and there are many approaches for fertilizer N rate recommendations. Existing fertilizer N rate recommendation systems can be improved by incorporating the effects of weather on sidedress economicoptimum N rates (EONR). In this study, we evaluated the performance of machine learning methods, a Bayesian Network (BN) and a Random Forest (RF) for estimating EONR for corn. BN draws relationships between variables based on assumptions about conditional independence, where the model is structured by an algorithm or, in this case, expert opinion. In contrast, RF determines model structure based on the input variables and model output. The models were trained and validated using a large database (n = 324) of corn yield response to N fertilizer collected across southern Ontario. Sixty‐six of the 324 site‐years were used for validation with success assessed by the frequency that N rate predictions that produced net returns were within CAN$25 ha−1 of the observed EONR. The success rate was 64% and 48% for the BN and RF, respectively. Both models incorporated weather from planting to sidedress and outperformed a benchmark provincial N recommendation system. We argue that BN has advantages when some input variables are unknown or uncertain and for improving model structure with stakeholder feedback. Moreover, RF is easy to implement but the model structure must use point estimates instead of probabilities for uncertain parameter values such as future weather. BN represents a more flexible modeling approach than RF for incorporating both modeling and stakeholder input.

Single Banana Appearance Grading with Ppyolo-Banana

Highlights An inspection method for grading the appearance of bananas using a target detection algorithm is proposed—by calculating the number and area ratio of different defective areas as a discriminating criterion. The Mish activation function makes the network easier to optimize and improves generalization performance. CustomPAN adds an attention mechanism, optimized for better multi-feature fusion. Optimized loss function in regression task with DIoULoss. Abstract. With the development of the fruit individual packaging industry, the appearance quality of individually packaged fruits has put forward higher requirements. Due to the dense and uneven defects on the surface of bananas, the existing detection algorithms are prone to the problem of unrecognizable or degraded recognition accuracy. In this article, we propose an efficient banana surface defect detection model, the PPYOLO-Banana model. PPYOLO-Banana is based on the PPYOLOE+-m model with improved model structure and loss function, and the optimized CustomPAN can get more multi-level features, and compared with the original network PPYOLOE+-m model, the algorithm significantly improves the accuracy, with an average accuracy improvement of 2.2% (1.3% for the original image test set). mAP of PPYOLO-Banana is 97.0% (96.1% for the original image test set), which is 14.3% higher than the PPYOLOE model, and 10.9%, 8.9%, 8.9%, and 8.1% higher than the YOLOX, YOLOX-tiny, YOLOv5, and YOLOV4 models, respectively. The detection speed of the PPYOLO-Banana model is 17.71 frames per second, which is 2.95, 2.10, 1.90, and 0.98 times higher than that of YOLOv3, YOLOv4, YOLOX, and YOLOX-tiny, respectively. The results show that the proposed PPYOLO-Banana model achieves a balance between accuracy and speed in recognizing banana surface defects, improves the quality detection capability of individually packed fruits, it can effectively grade the quality of banana appearance, and has good potential to become an intelligent sorting machine. Keywords: Banana defect recognition, Banana appearance grading, CustomPAN, DIoULoss, PPYOLOE+.

A WRF/WRF-Hydro coupling system with an improved structure for rainfall-runoff simulation with mixed runoff generation mechanism

• The structure of WRF-Hydro is improved for runoff generation with mixed mechanisms. • Empirical equations from a conceptual model are transplanted into WRF-Hydro. • Merged rainfall forcing with radar QPE and rain gauges is used for model calibration. • The structure is verified for both one-way and the fully coupled WRF/WRF-Hydro. • Increase of infiltration-excess runoff is found with decrease of saturation-excess. The coupled atmospheric-hydrologic systems help achieve deeper understanding on the interactions between the atmospheric and land-surface processes, improve the spatial and temporal accuracy of hydrologic forecasts and extend the forecast leading time. WRF-Hydro is nowadays a widely used hydrologic module to be coupled with the mesoscale numerical weather model WRF for atmospheric-hydrologic research and applications. The structure of WRF-Hydro is improved and expanded in this study to better adapt to the complicated rainfall-runoff transformation mechanism in the mixed runoff generation regions. The infiltration parameterisation is replaced by an infiltration equation that takes into account the impact of the surface soil moisture variation on the infiltration capacity, and the spatial discretisation of the infiltration capacity from a distribution curve is achieved based on the conception of the topographic index. For the runoff convergence, the river channel leakage loss is introduced into the Muskingum-Cunge (MC) streamflow calculations with variable parameters in time. Efforts are also made to reduce parameter calibration errors in WRF-Hydro. The accuracy of the WRF rainfall forcing is improved by merging weather radar and rain gauge observations using the Kriging with external drift (KED) interpolation tool. The key parameters of WRF-Hydro are then calibrated using the improved rainfall forcing with the merging observations. The performance of the improved WRF-Hydro structure is explored in both one-way and the fully coupled modes with the WRF model. Typical rainstorm events are selected from semi-humid and semi-arid catchments of Northern China as case studies. Results have shown that the improved WRF-Hydro is more effective in simulating floods with high peaks and steep rises-and-falls, but a problem is also shown of receding more quickly for low peaks. The improved model structure has changed the proportion of the mixed runoff generation. In the long term, there is an increase in the infiltration-excess runoff generation and decrease in the saturation-excess amount. Both the one-way and the fully coupled WRF/WRF-Hydro systems have demonstrated the suitability of the improved structure for rainfall-runoff processes dominated by the infiltration-excess runoff generation.

Uneven wrapped phase pattern denoising using a deep neural network.

The wrapped phase patterns obtained from an object composed of different materials have uneven gray values. In this paper, we improve the dilated-blocks-based deep convolution neural network (DBDNet) and build a new dataset for restoring the uneven gray values of uneven wrapped phase patterns as well as eliminating the speckle noise. In our method, we improve the structure of dilated blocks in DBDNet to enhance the ability of obtaining full scales of gray values and speckle noise information in the uneven phase patterns. We use the combined MS_SSIM+L1 loss function to improve the denoising and restoration performance of our method. We compare three representative networks ResNet-based, ADNet, and BRDNet in denoising with our proposed method. We test the three compared methods and our method on one group of computer-simulated and one group of experimentally obtained uneven noisy wrapped phase patterns from a dynamic measurement. We also conduct the ablation experiments on the improved model structure and the combined loss function used in our method. The denoising performance has been evaluated quantitatively and qualitatively. The denoising results demonstrate that our proposed method can reduce high speckle noise, restore the uneven gray values of wrapped phase patterns, and get better results than the compared methods.

Flow Characteristic Analysis of CFD-based Air Filter Forward Intake Pipe

Based on one type of heavy vehicles with air filter forward intake pipe as the research object, the flow field characteristics of air filter forward intake pipe is simulated. The original model of the flow characteristics such as pressure and velocity fields are analyzed based on the simulation results, and an improved model is proposed. Through the numerical simulation analysis of the improved model, it is proved that the internal flow characteristics have been significantly improved, and the intake resistance has been reduced, and the improved model structure is in line with the flow characteristics of the gas better.

Utility of Remotely Sensed Evapotranspiration Products to Assess an Improved Model Structure

There is a certain level of predictive uncertainty when hydrologic models are applied for operational purposes. Whether structural improvements address uncertainty has not well been evaluated due to the lack of observational data. This study investigated the utility of remotely sensed evapotranspiration (RS-ET) products to quantitatively represent improvements in model predictions owing to structural improvements. Two versions of the Soil and Water Assessment Tool (SWAT), representative of original and improved versions, were calibrated against streamflow and RS-ET. The latter version contains a new soil moisture module, referred to as RSWAT. We compared outputs from these two versions with the best performance metrics (Kling–Gupta Efficiency [KGE], Nash-Sutcliffe Efficiency [NSE] and Percent-bias [P-bias]). Comparisons were conducted at two spatial scales by partitioning the RS-ET into two scales, while streamflow comparisons were only conducted at one scale. At the watershed level, SWAT and RSWAT produced similar metrics for daily streamflow (NSE of 0.29 and 0.37, P-bias of 1.7 and 15.9, and KGE of 0.47 and 0.49, respectively) and ET (KGE of 0.48 and 0.52, respectively). At the subwatershed level, the KGE of RSWAT (0.53) for daily ET was greater than that of SWAT (0.47). These findings demonstrated that RS-ET has the potential to increase prediction accuracy from model structural improvements and highlighted the utility of remotely sensed data in hydrologic modeling.

Multi-model evaluation of phenology prediction for wheat in Australia

Predicting wheat phenology is important for cultivar selection, for effective crop management and provides a baseline for evaluating the effects of global change. Evaluating how well crop phenology can be predicted is therefore of major interest. Twenty-eight wheat modeling groups participated in this evaluation. Our target population was wheat fields in the major wheat growing regions of Australia under current climatic conditions and with current local management practices. The environments used for calibration and for evaluation were both sampled from this same target population. The calibration and evaluation environments had neither sites nor years in common, so this is a rigorous evaluation of the ability of modeling groups to predict phenology for new sites and weather conditions. Mean absolute error (MAE) for the evaluation environments, averaged over predictions of three phenological stages and over modeling groups, was 9 days, with a range from 6 to 20 days. Predictions using the multi-modeling group mean and median had prediction errors nearly as small as the best modeling group. About two thirds of the modeling groups performed better than a simple but relevant benchmark, which predicts phenology by assuming a constant temperature sum for each development stage. The added complexity of crop models beyond just the effect of temperature was thus justified in most cases. There was substantial variability between modeling groups using the same model structure, which implies that model improvement could be achieved not only by improving model structure, but also by improving parameter values, and in particular by improving calibration techniques.

Comparison of global sensitivity analysis techniques based on a process-based model CROBAS.

Constructions of process or mechanistic models are limited by physiological parameters, due to difficulty in direct and precise measurement. Global sensitivity analysis could evaluate the response of model outputs to changes in physiological parameters, and provide information for improving model structure, data collection, and parameter calibration. Based on a process model CROBAS, 10 parameters related to tree structure of Pinus armandii were selected to compare three widely used global sensitivity analysis methods (the Morris screening method, the variance-based Sobol indices, and the Extended Fourier Amplitude Sensitivity Test (EFAST)), with the objective function formulated by the Nash-Sutcliffe Efficiency (NSE) of tree height and biomass. The results showed that the sensitivity order of parameters slightly varied across different methods, which considerably changed with different objective functions. Both the Morris method and the EFAST method outperformed the Sobol method in terms of time consuming and convergence efficiency. All outputs were sensitive to the maximum rate of canopy photosynthesis per unit area, the specific leaf area, and the extinction coefficient. The light interception of tree canopy played a key role in the simulation of tree growth with CROBAS, suggesting that the module of photosynthetic carbon fixation took priority over any other modules for data collection and model validation during module calibration and tree growth simulation for CROBAS. The calculation and validation of foliage biomass module were crucial when applying carbon balance theory to biomass simulations. In conclusion, for the sensitivity analysis of a complex process-based model, the Morris method was suitable for qualitative studies, while the EFAST method was recommended for quantitative studies.

Multiple order model migration and optimal model selection for online glucose prediction in Type 1 diabetes

In order to address the problem of insufficient available modelling data for glucose prediction, as well as modelling burden, a model migration method was developed in a previous work to quickly transfer an old model to a new subject by using a simple parameter adjustment. However, this method, which is referred to as first order model migration (FOMM), only considers a single order for each exogenous input, and may not produce an optimal model structure for accurate prediction. To overcome this problem, a multiple order model migration (MOMM) algorithm is proposed in this study. For different numbers of modelling samples, including glucose and two exogenous inputs (meal and insulin), the optimal modelling method may be different, and therefore must be properly determined for each modelling scenario. First, the optimal model order is determined for each input and a multiple order prediction model is used. Then, a MOMM algorithm is developed based on particle swarm optimization (PSO) to simultaneously revise multiple parameters. The multiple order parameters of each input in the old model are quickly customized so that the revised model can be used for new subjects with desirable prediction accuracy. In particular, the influence of the number of modelling samples is analysed to check the applicability of different methods; this analysis determines the appropriate selection guidelines for the optimal model in response to different data sizes. The proposed method was evaluated using thirty in silico subjects and clinical data from seven individuals with type 1 diabetes mellitus (DM). Overall, the MOMM algorithm presented superior results when the time period for collecting the samples was larger than 10 h (50 samples). In particular, the size of the modelling samples was separated into three different regions by evaluating the glucose prediction performance and the comparison between different algorithms for both in silico and clinical subjects. In Region I, the FOMM method achieves the best performance. In Region II, the MOMM method should be used and the prediction accuracy is superior in Region II in general. With enough samples (Region III), the subject-dependent model (SM) algorithm can be chosen. The MOMM algorithm is demonstrated to be able to transfer models for new subjects with improved model structure. This article is protected by copyright. All rights reserved.

Reducing the uncertainty of parameters controlling seasonal carbon and water fluxes in Chinese forests and its implication for simulated climate sensitivities

AbstractReducing parameter uncertainty of process‐based terrestrial ecosystem models (TEMs) is one of the primary targets for accurately estimating carbon budgets and predicting ecosystem responses to climate change. However, parameters in TEMs are rarely constrained by observations from Chinese forest ecosystems, which are important carbon sink over the northern hemispheric land. In this study, eddy covariance data from six forest sites in China are used to optimize parameters of the ORganizing Carbon and Hydrology In Dynamics EcosystEms TEM. The model‐data assimilation through parameter optimization largely reduces the prior model errors and improves the simulated seasonal cycle and summer diurnal cycle of net ecosystem exchange, latent heat fluxes, and gross primary production and ecosystem respiration. Climate change experiments based on the optimized model are deployed to indicate that forest net primary production (NPP) is suppressed in response to warming in the southern China but stimulated in the northeastern China. Altered precipitation has an asymmetric impact on forest NPP at sites in water‐limited regions, with the optimization‐induced reduction in response of NPP to precipitation decline being as large as 61% at a deciduous broadleaf forest site. We find that seasonal optimization alters forest carbon cycle responses to environmental change, with the parameter optimization consistently reducing the simulated positive response of heterotrophic respiration to warming. Evaluations from independent observations suggest that improving model structure still matters most for long‐term carbon stock and its changes, in particular, nutrient‐ and age‐related changes of photosynthetic rates, carbon allocation, and tree mortality.

A Vision for Incorporating Environmental Effects into Nitrogen Management Decision Support Tools for U.S. Maize Production.

Meeting crop nitrogen (N) demand while minimizing N losses to the environment has proven difficult despite significant field research and modeling efforts. To improve N management, several real-time N management tools have been developed with a primary focus on enhancing crop production. However, no coordinated effort exists to simultaneously address sustainability concerns related to N losses at field- and regional-scales. In this perspective, we highlight the opportunity for incorporating environmental effects into N management decision support tools for United States maize production systems by integrating publicly available crop models with grower-entered management information and gridded soil and climate data in a geospatial framework specifically designed to quantify environmental and crop production tradeoffs. To facilitate advances in this area, we assess the capability of existing crop models to provide in-season N recommendations while estimating N leaching and nitrous oxide emissions, discuss several considerations for initial framework development, and highlight important challenges related to improving the accuracy of crop model predictions. Such a framework would benefit the development of regional sustainable intensification strategies by enabling the identification of N loss hotspots which could be used to implement spatially explicit mitigation efforts in relation to current environmental quality goals and real-time weather conditions. Nevertheless, we argue that this long-term vision can only be realized by leveraging a variety of existing research efforts to overcome challenges related to improving model structure, accessing field data to enhance model performance, and addressing the numerous social difficulties in delivery and adoption of such tool by stakeholders.

Load Identification Modeling with Improved Model Structure

The running state of different induction motor in the load group will be moved toward two directions when it was occurred the large disturbance in the power system, one is to keep on running, another is out of step or stall. If the quantity of stall induction motor in the group is a large number, the dynamic of stall motors will effect on the stability of power system obviously. So it is necessary to develop the detailed load model to simulate the complicated dynamic characteristic of the load group. An improved synthesis load model which combines two kinds of induction motor is proposed, and the improved genetic algorithm is used to identify the parameter of load model. The case is studied to illustrate the effective and comprehensive adaptability of the proposed identification model.

Resolving structural errors in a spatially distributed hydrologic model using ensemble Kalman filter state updates

Abstract. In hydrological modeling, model structures are developed in an iterative cycle as more and different types of measurements become available and our understanding of the hillslope or watershed improves. However, with increasing complexity of the model, it becomes more and more difficult to detect which parts of the model are deficient, or which processes should also be incorporated into the model during the next development step. In this study, we first compare two methods (the Shuffled Complex Evolution Metropolis algorithm (SCEM-UA) and the Simultaneous parameter Optimization and Data Assimilation algorithm (SODA)) to calibrate a purposely deficient 3-D hillslope-scale model to error-free, artificially generated measurements. We use a multi-objective approach based on distributed pressure head at the soil–bedrock interface and hillslope-scale discharge and water balance. For these idealized circumstances, SODA's usefulness as a diagnostic methodology is demonstrated by its ability to identify the timing and location of processes that are missing in the model. We show that SODA's state updates provide information that could readily be incorporated into an improved model structure, and that this type of information cannot be gained from parameter estimation methods such as SCEM-UA. We then expand on the SODA result by performing yet another calibration, in which we investigate whether SODA's state updating patterns are still capable of providing insight into model structure deficiencies when there are fewer measurements, which are moreover subject to measurement noise. We conclude that SODA can help guide the discussion between experimentalists and modelers by providing accurate and detailed information on how to improve spatially distributed hydrologic models.

On the sources of hydrological prediction uncertainty in the Amazon

Abstract. Recent extreme events in the Amazon River basin and the vulnerability of local population motivate the development of hydrological forecast systems using process based models for this region. In this direction, the knowledge of the source of errors in hydrological forecast systems may guide the choice on improving model structure, model forcings or developing data assimilation systems for estimation of initial model states. We evaluate the relative importance of hydrologic initial conditions and model meteorological forcings errors (precipitation) as sources of stream flow forecast uncertainty in the Amazon River basin. We used a hindcast approach that compares Ensemble Streamflow Prediction (ESP) and a reverse Ensemble Streamflow Prediction (reverse-ESP). Simulations were performed using the physically-based and distributed hydrological model MGB-IPH, comprising surface energy and water balance, soil water, river and floodplain hydrodynamics processes. The model was forced using TRMM 3B42 precipitation estimates. Results show that uncertainty on initial conditions plays an important role for discharge predictability, even for large lead times (∼1 to 3 months) on main Amazonian Rivers. Initial conditions of surface waters state variables are the major source of hydrological forecast uncertainty, mainly in rivers with low slope and large floodplains. Initial conditions of groundwater state variables are important, mostly during low flow period and in the southeast part of the Amazon where lithology and the strong rainfall seasonality with a marked dry season may be the explaining factors. Analyses indicate that hydrological forecasts based on a hydrological model forced with historical meteorological data and optimal initial conditions may be feasible. Also, development of data assimilation methods is encouraged for this region.

Using discharge data to reduce structural deficits in a hydrological model with a Bayesian inference approach and the implications for the prediction of critical source areas

A distributed hydrological model was used to simulate the distribution of fast runoff formation as a proxy for critical source areas for herbicide pollution in a small agricultural catchment in Switzerland. We tested to what degree predictions based on prior knowledge without local measurements could be improved upon relying on observed discharge. This learning process consisted of five steps: For the prior prediction (step 1), knowledge of the model parameters was coarse and predictions were fairly uncertain. In the second step, discharge data were used to update the prior parameter distribution. Effects of uncertainty in input data and model structure were accounted for by an autoregressive error model. This step decreased the width of the marginal distributions of parameters describing the lower boundary (percolation rates) but hardly affected soil hydraulic parameters. Residual analysis (step 3) revealed model structure deficits. We modified the model, and in the subsequent Bayesian updating (step 4) the widths of the posterior marginal distributions were reduced for most parameters compared to those of the prior. This incremental procedure led to a strong reduction in the uncertainty of the spatial prediction. Thus, despite only using spatially integrated data (discharge), the spatially distributed effect of the improved model structure can be expected to improve the spatially distributed predictions also. The fifth step consisted of a test with independent spatial data on herbicide losses and revealed ambiguous results. The comparison depended critically on the ratio of event to preevent water that was discharged. This ratio cannot be estimated from hydrological data only. The results demonstrate that the value of local data is strongly dependent on a correct model structure. An iterative procedure of Bayesian updating, model testing, and model modification is suggested.

The limits to prediction in ecological systems

Predicting the future trajectories of ecological systems is increasingly important as the magnitude of anthropogenic perturbation of the earth systems grows. We distinguish between two types of predictability: the intrinsic or theoretical predictability of a system and the realized predictability that is achieved using available models and parameterizations. We contend that there are strong limits on the intrinsic predictability of ecological systems that arise from inherent characteristics of biological systems. While the realized predictability of ecological systems can be limited by process and parameter misspecification or uncertainty, we argue that the intrinsic predictability of ecological systems is widely and strongly limited by computational irreducibility. When realized predictability is low relative to intrinsic predictability, prediction can be improved through improved model structure or specification of parameters. Computational irreducibility, however, asserts that future states of the system cannot be derived except through computation of all of the intervening states, imposing a strong limit on the intrinsic or theoretical predictability. We argue that ecological systems are likely to be computationally irreducible because of the difficulty of pre-stating the relevant features of ecological niches, the complexity of ecological systems and because the biosphere can enable its own novel system states or adjacent possible. We argue that computational irreducibility is likely to be pervasive and to impose strong limits on the potential for prediction in ecology.