scholarly journals The impacts of data constraints on the predictive performance of a general process-based crop model (PeakN-crop v1.0)

2016 ◽  
Author(s):  
Silvia Caldararu ◽  
Drew W. Purves ◽  
Matthew J. Smith
Keyword(s):  
Environmental Changes ◽  
Crop Yields ◽  
Vegetation Indices ◽  
Model Fitting ◽  
Model Performance ◽  
Predictive Performance ◽  
Crop Model ◽  
Model Fit ◽  
General Process ◽  
Fitting Algorithm

Abstract. Improving international food security under a changing climate and increasing human population will be greatly aided by improving our ability to modify, understand and predict crop growth. What we predominantly have at our disposal are either process based models of crop physiology or statistical analyses of yield datasets, both of which suffer from various sources of error. In the current paper we present a generic process based crop model which we parametrise using a Bayesian model fitting algorithm to three different sources of data – space based vegetation indices, eddy covariance productivity measurements and regional crop yields. We show that the model parametrised without data, based on prior knowledge of the parameters can largely capture the observed behaviour but the data constrained model greatly improves both the model fit and reduces prediction uncertainty. We investigate the extent to which each dataset contributes to the model performance and show that while all data improves on the prior model fit, the satellite based data and crop yield estimates are particularly important for reducing model error and uncertainty. Despite these improvements, we conclude that there are still significant knowledge10 gaps, in terms of available data for model parametrisation, but our study can help indicate the necessary data collection steps for improvement in our predictions of crop yields and crop responses to environmental changes.

scholarly journals The impacts of data constraints on the predictive performance of a general process-based crop model (PeakN-crop v1.0)

2017 ◽  
Vol 10 (4) ◽  
pp. 1679-1701 ◽  
Author(s):  
Silvia Caldararu ◽  
Drew W. Purves ◽  
Matthew J. Smith
Keyword(s):  
Environmental Changes ◽  
Crop Yields ◽  
Vegetation Indices ◽  
Model Fitting ◽  
Model Performance ◽  
Predictive Performance ◽  
Crop Model ◽  
Model Fit ◽  
General Process ◽  
Fitting Algorithm

Abstract. Improving international food security under a changing climate and increasing human population will be greatly aided by improving our ability to modify, understand and predict crop growth. What we predominantly have at our disposal are either process-based models of crop physiology or statistical analyses of yield datasets, both of which suffer from various sources of error. In this paper, we present a generic process-based crop model (PeakN-crop v1.0) which we parametrise using a Bayesian model-fitting algorithm to three different sources: data–space-based vegetation indices, eddy covariance productivity measurements and regional crop yields. We show that the model parametrised without data, based on prior knowledge of the parameters, can largely capture the observed behaviour but the data-constrained model greatly improves both the model fit and reduces prediction uncertainty. We investigate the extent to which each dataset contributes to the model performance and show that while all data improve on the prior model fit, the satellite-based data and crop yield estimates are particularly important for reducing model error and uncertainty. Despite these improvements, we conclude that there are still significant knowledge gaps, in terms of available data for model parametrisation, but our study can help indicate the necessary data collection to improve our predictions of crop yields and crop responses to environmental changes.

scholarly journals Evaluating the Sensitivity of Agricultural Model Performance to Different Climate Inputs

2016 ◽  
Vol 55 (3) ◽  
pp. 579-594 ◽  
Author(s):  
Michael J. Glotter ◽  
Elisabeth J. Moyer ◽  
Alex C. Ruane ◽  
Joshua W. Elliott
Keyword(s):  
Arid Regions ◽  
Crop Yields ◽  
Heat Waves ◽  
Model Performance ◽  
Ground Truth ◽  
Crop Model ◽  
Systematic Evaluation ◽  
Multiple Sources ◽  
Crop Models ◽  
Maize Yields

AbstractProjections of future food production necessarily rely on models, which must themselves be validated through historical assessments comparing modeled and observed yields. Reliable historical validation requires both accurate agricultural models and accurate climate inputs. Problems with either may compromise the validation exercise. Previous studies have compared the effects of different climate inputs on agricultural projections but either incompletely or without a ground truth of observed yields that would allow distinguishing errors due to climate inputs from those intrinsic to the crop model. This study is a systematic evaluation of the reliability of a widely used crop model for simulating U.S. maize yields when driven by multiple observational data products. The parallelized Decision Support System for Agrotechnology Transfer (pDSSAT) is driven with climate inputs from multiple sources—reanalysis, reanalysis that is bias corrected with observed climate, and a control dataset—and compared with observed historical yields. The simulations show that model output is more accurate when driven by any observation-based precipitation product than when driven by non-bias-corrected reanalysis. The simulations also suggest, in contrast to previous studies, that biased precipitation distribution is significant for yields only in arid regions. Some issues persist for all choices of climate inputs: crop yields appear to be oversensitive to precipitation fluctuations but undersensitive to floods and heat waves. These results suggest that the most important issue for agricultural projections may be not climate inputs but structural limitations in the crop models themselves.

scholarly journals A summary of WIND magnetic clouds for years 1995-2003: model-fitted parameters, associated errors and classifications

2006 ◽  
Vol 24 (1) ◽  
pp. 215-245 ◽  
Author(s):  
R. P. Lepping ◽  
D. B. Berdichevsky ◽  
C.-C. Wu ◽  
A. Szabo ◽  
T. Narock ◽  
...  
Keyword(s):  
Temperature Drop ◽  
Model Fitting ◽  
Sunspot Cycle ◽  
Flux Rope ◽  
Axial Current ◽  
Model Fit ◽  
Magnetic Clouds ◽  
Symmetric Model ◽  
Large Set ◽  
Interplanetary Shocks

Abstract. Interplanetary magnetic clouds (MCs) have been identified for the first 8.6 years of the WIND mission, and their magnetic field structures have been parameter-fitted by a static, force free, cylindrically-symmetric model (Lepping et al., 1990) with various levels of success. This paper summarizes various aspects of the results of the model fitting by providing: seven estimated model fit-parameter values for each of the 82 MCs found, their objectively determined quality estimates, closest approach vectors (in two coordinate frames), fit-parameter errors for the cases of acceptable quality (50 cases, or 61%), axial magnetic fluxes, axial current densities, and total axial current - as well as some examples of MC profiles for various conditions and "categories" for each case (e.g. Bz: N→S or S→N, etc.). MC quality is estimated from a quantitative consideration of a large set of parameters, such as the chi-squared of the model fit, degree of asymmetry of the B profile, and a comparison of two means of estimating radius. This set of MCs was initially identified by visual inspection of relevant field and plasma data. Each resulting MC candidate is then tested through the use of the MC parameter model, for various adjusted durations to determine the best fit, which helps to refine the boundary-times. The resulting MC set is called Set 1. Another, larger, set (Set 2) of MCs is identified through an automated program whose criteria are based on general MC plasma and field characteristics at 1AU determined through past experience. Set 1 is almost fully contained within Set 2, whose frequency of occurrence better matches that of the sunspot cycle than Set 1. The difference-set (Set 2-Set 1) is referred to as the magnetic cloud-like (MCL) set, whose members do not very well represent good flux ropes through modeling. We present a discussion of how a MC's front boundary is specifically identified in terms of multi-parameter considerations (i.e. any one or more of: increase in B, directional discontinuity, magnetic hole in B, drop in proton plasma beta, B-fluctuation level change, proton temperature drop, etc.), as well as through the application of the flux rope model. Also presented are examples of unusual MCs, as well as some commonly occurring relationships, such as the existence and frequency (approx. 1/2 the time) of upstream interplanetary shocks, and less frequent internal shocks.

Cardiovascular risk model performance in women with and without hypertensive disorders of pregnancy

Heart ◽  
2018 ◽  
Vol 105 (4) ◽  
pp. 330-336 ◽  
Author(s):  
Veerle Dam ◽  
N Charlotte Onland-Moret ◽  
W M Monique Verschuren ◽  
Jolanda M A Boer ◽  
Laura Benschop ◽  
...  
Keyword(s):  
Risk Model ◽  
Model Performance ◽  
Predictive Performance ◽  
Baseline Risk ◽  
Hypertensive Disorders Of Pregnancy ◽  
Cvd Risk ◽  
Hypertensive Disorders ◽  
Framingham Risk ◽  
History Of ◽  
Score Model

ObjectivesCompare the predictive performance of Framingham Risk Score (FRS), Pooled Cohort Equations (PCEs) and Systematic COronary Risk Evaluation (SCORE) model between women with and without a history of hypertensive disorders of pregnancy (hHDP) and determine the effects of recalibration and refitting on predictive performance.MethodsWe included 29 751 women, 6302 with hHDP and 17 369 without. We assessed whether models accurately predicted observed 10-year cardiovascular disease (CVD) risk (calibration) and whether they accurately distinguished between women developing CVD during follow-up and not (discrimination), separately for women with and without hHDP. We also recalibrated (updating intercept and slope) and refitted (recalculating coefficients) the models.ResultsOriginal FRS and PCEs overpredicted 10-year CVD risks, with expected:observed (E:O) ratios ranging from 1.51 (for FRS in women with hHDP) to 2.29 (for PCEs in women without hHDP), while E:O ratios were close to 1 for SCORE. Overprediction attenuated slightly after recalibration for FRS and PCEs in both hHDP groups. Discrimination was reasonable for all models, with C-statistics ranging from 0.70-0.81 (women with hHDP) and 0.72–0.74 (women without hHDP). C-statistics improved slightly after refitting 0.71–0.83 (with hHDP) and 0.73–0.80 (without hHDP). The E:O ratio of the original PCE model was statistically significantly better in women with hHDP compared with women without hHDP.ConclusionsSCORE performed best in terms of both calibration and discrimination, while FRS and PCEs overpredicted risk in women with and without hHDP, but improved after recalibrating and refitting the models. No separate model for women with hHDP seems necessary, despite their higher baseline risk.

scholarly journals Grain Yield Estimation in Rice Breeding Using Phenological Data and Vegetation Indices Derived from UAV Images

Agronomy ◽  
2021 ◽  
Vol 11 (12) ◽  
pp. 2439
Author(s):  
Haixiao Ge ◽  
Fei Ma ◽  
Zhenwang Li ◽  
Changwen Du
Keyword(s):  
Grain Yield ◽  
Low Cost ◽  
Vegetation Indices ◽  
Model Performance ◽  
Estimation Accuracy ◽  
Accurate Estimation ◽  
Rice Breeding ◽  
Yield Estimation ◽  
Phenological Data ◽  
Significant Difference

The accurate estimation of grain yield in rice breeding is crucial for breeders to screen and select qualified cultivars. In this study, a low-cost unmanned aerial vehicle (UAV) platform mounted with an RGB camera was carried out to capture high-spatial resolution images of rice canopy in rice breeding. The random forest (RF) regression techniques were used to establish yield models by using (1) only color vegetation indices (VIs), (2) only phenological data, and (3) fusion of VIs and phenological data as inputs, respectively. Then, the performances of RF models were compared with the manual observation and CERES-Rice model. The results indicated that the RF model using VIs only performed poorly for estimating yield; the optimized RF model that combined the use of phenological data and color VIs performed much better, which demonstrated that the phenological data significantly improved the model performance. Furthermore, the yield estimation accuracy of 21 rice cultivars that were continuously planted over three years in the optimal RF model had no significant difference (p > 0.05) with that of the CERES-Rice model. These findings demonstrate that the RF model, by combining phenological data and color Vis, is a potential and cost-effective way to estimate yield in rice breeding.

scholarly journals Mind the gap: preventing circularity in missense variant prediction

2020 ◽  
Author(s):  
Stephan Heijl ◽  
Bas Vroling ◽  
Tom van den Bergh ◽  
Henk-Jan Joosten
Keyword(s):  
Real World ◽  
Performance Metrics ◽  
Model Performance ◽  
Predictive Performance ◽  
Missense Variant ◽  
Training Methods ◽  
The Real ◽  
Evaluation Scores ◽  
Evaluation Strategies ◽  
Generic Strategy

AbstractDespite advances in the field of missense variant effect prediction, the real clinical utility of current computational approaches remains rather limited. There is a large difference in performance metrics reported by developers and those observed in the real world. Most currently available predictors suffer from one or more types of circularity in their training and evaluation strategies that lead to overestimation of predictive performance. We present a generic strategy that is independent of dataset properties and algorithms used, to deal with circularity in the training phase. This results in more robust predictors and evaluation scores that accurately reflect the real-world performance of predictive models. Additionally, we show that commonly used training methods can have an adverse impact on model performance and lead to gross overestimation of true predictive performance.

scholarly journals Predictive Performance of Hargreaves Model in Estimating Solar Radiation in Yola, Adamawa State Nigeria Using Minimum Climatological Data

2018 ◽  
pp. 1-8
Author(s):  
Ojo Samuel ◽  
Alimi Taofeek Ayodele ◽  
Amos Anna Solomon
Keyword(s):  
Solar Radiation ◽  
Model Performance ◽  
Predictive Performance ◽  
Global Solar Radiation ◽  
Climatic Data ◽  
Climatological Data ◽  
University Of Technology ◽  
Hargreaves Model ◽  
The Many

Mathematical models have been very useful in reducing challenges encountered by researchers due to the inability of having solar radiation data or lack of instrumental sites at every point on the Earth.  This work aimed at investigating the prediction performance of Hargreaves-Samani’s model in estimating global solar radiation (GSR) out of the many other empirical models so far formulated for this purpose. This model basically uses maximum and minimum temperature data and basically used in mid-latitudes. The paper attempts to assess the predictive performance of Hargreaves-Samani’s model in the Savanna region using Yola as a case study. Estimated values of GSR from one month data adopted from the Meteorological station of the Department of Geography, Federal University of Technology, Yola, Nigeria was used for this purpose. Using this model shows a 95% index of agreement (IA) with the observed values; which suggests a good model performance and can also be used in estimating global solar radiation in the Savanna region particularly in areas with little or no such climatic data.

scholarly journals MONITORAMENTO DA COBERTURA VEGETAL POR SENSORIAMENTO REMOTO NO SEMIÁRIDO BRASILEIRO ATRAVÉS DE ÍNDICES DE VEGETAÇÃO

Nativa ◽  
2019 ◽  
Vol 7 (6) ◽  
pp. 708
Author(s):  
Caio Victor Santos Silva ◽  
Jhon Lennon Bezerra da Silva ◽  
Geber Barbosa De Albuquerque Moura ◽  
Pabrício Marcos Oliveira Lopes ◽  
Cristina Rodrigues Nascimento ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Vegetation Cover ◽  
Environmental Changes ◽  
Spectral Response ◽  
Vegetation Indices ◽  
Digital Processing ◽  
Semiarid Region ◽  
Dry Period ◽  
Biophysical Parameters ◽  
Brazilian Semiarid

São necessárias medidas que visem à proteção e conservação dos recursos hídricos e naturais de forma rápida e eficiente. As técnicas de sensoriamento remoto são essenciais para o monitoramento ambiental dos recursos no semiárido no espaço e no tempo. Objetivou-se monitorar e analisar à dinâmica da cobertura vegetal através da variabilidade espaço-temporal do albedo da superfície e índices de vegetação em região de Caatinga do semiárido brasileiro por sensoriamento remoto. A área de estudo é o município de Arcoverde, localizado no semiárido de Pernambuco. O estudo foi desenvolvido através de seis imagens orbitais do Landsat-5 do sensor TM. O processamento digital dos parâmetros biofísicos foi realizado pelo algoritmo SEBAL. Os resultados foram analisados através da estatística descritiva e quanto a sua variabilidade. Áreas possivelmente degradadas foram identificadas pelos altos valores de albedo e índices de vegetação significativamente menores, localizadas à sudoeste e noroeste da região. Os índices apresentaram comportamento similares, principalmente no período seco, com baixos valores sendo próximos de zero, áreas afetadas pelo período de seca no semiárido. O SAVI apresentou maior precisão, destacando melhor resposta espectral da vegetação. O sensoriamento remoto promoveu monitoramento espaço-temporal adequado, destacando principalmente o período classificado como climaticamente seco através do albedo e índices de vegetação.Palavras-chave: Caatinga; NDVI; SAVI; mudanças ambientais; SEBAL. MONITORING OF VEGETATION COVER BY REMOTE SENSING IN BRAZILIAN SEMIARID THROUGH VEGETATION INDICES ABSTRACT: Measures are needed aimed at the protection and conservation of water and natural resources quickly and efficiently. Remote sensing techniques are essential for the environmental monitoring of resources in the semiarid region in space and time. Aimed to monitor and analyze the dynamics of vegetation cover through the spatial-temporal variability of the surface albedo and indices of vegetation in the Caatinga region of the Brazilian semiarid by remote sensing. The study area is the municipality of Arcoverde, located in the semiarid of Pernambuco. The study was developed through six orbital images of Landsat-5 of the TM sensor. The digital processing of the biophysical parameters was performed by the SEBAL algorithm. The results were analyzed through descriptive statistics and their variability. Possibly degraded areas were identified by high albedo values and significantly lower vegetation indices, located in the southwest and northwest of the region. The indexes showed similar behavior, mainly in the dry period, with low values being close to zero, areas affected by the dry period in the semiarid. The SAVI presented higher accuracy, highlighting better spectral response of the vegetation. Remote sensing promoted adequate space-time monitoring, highlighting mainly the period classified as climatically dry through the albedo and vegetation indexes.Keywords: Caatinga; NDVI; SAVI; environmental changes; SEBAL.

scholarly journals Comparison of Empirical and Physical Modelling for Estimation of Biochemical and Biophysical Vegetation Properties: Field Scale Analysis across an Arctic Bioclimatic Gradient

2020 ◽  
Vol 12 (18) ◽  
pp. 3073
Author(s):  
Blair E. Kennedy ◽  
Douglas J. King ◽  
Jason Duffe
Keyword(s):  
Numerical Optimization ◽  
Vegetation Indices ◽  
Physical Modelling ◽  
Predictive Performance ◽  
High Arctic ◽  
Canopy Reflectance ◽  
Area Index ◽  
Empirical Modelling ◽  
Plant Area Index ◽  
Reflectance Data

To evaluate the potential of multi-angle hyperspectral sensors for monitoring vegetation variables in Arctic environments, empirical and physical modelling using field data was implemented for the retrieval of leaf and canopy chlorophyll content (LCC, CCC) and plant area index (PAI) measured at four sites situated across a bioclimatic gradient in the Western Canadian Arctic. Field reflectance data were acquired with an ASD FieldSpec (305–1075 nm) and used to simulate CHRIS Mode1 spectra (411–997 nm). Multi-angle measurements were taken corresponding to CHRIS view zenith angles (VZA) (−55°, −36°, 0°, +36°, +55°). Empirical modelling compared parametric regression based on vegetation indices (VIs) to non-parametric Gaussian Processes Regression (GPR). In physical modelling, PROSAIL was inverted using numerical optimization and look-up table (LUT) approaches. Cross-validation of the empirical models ranked GPR as best, followed by simple ratio (SR) with optimally selected NIR and red wavelengths, and then ROSAVI using its published wavelengths (mean r2cv = 0.62, 0.58, and 0.54, respectively across all sites, variables, and VZAs). However, the best predictive performance was achieved by SR followed by GPR and ROSAVI (NRMSEcv = 0.12, 0.16, 0.16, respectively). PROSAIL simulated the multi-angle top-of-canopy reflectance well with numerical optimization (r2 = ~0.99, RMSE = 0.004 ± 0.002), but best performing LUT models of LCC, CCC and PAI were poorer than the empirical approaches (mean r2 = 0.48, mean NRMSE = 0.22). PROSAIL performed best at the high Arctic sparsely vegetated site (r2 = 0.57–0.86 for all parameters). Overall, the best performing VZA was −55° for empirical modelling and 0° and ±55° for physical modelling; however, these were not significantly better than the other VZAs. Overall, this study demonstrates that, for Arctic vegetation, nadir narrowband reflectance data used to derive simple empirical VIs with optimally selected bands is a more efficient approach for modelling chlorophyll and PAI than more complex empirical and physical approaches.

Sign in / Sign up

Export Citation Format

Share Document