Try:droughtFusariumnitrogen fertilizerArabidopsisseed germination

“yield decrease”

151 matches

Claim types

Direction of the findings

plus 39 non-evaluative

Over time

1950 ’60 ’70 ’80 ’90 2000 ’10 ’20

Refine ·

More filters

Clear filters

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Forecast RMSE began decreasing in weeks 24 and 25.

Outcome: RMSE of the forecastse0
Direction: decreasee0
Value: weeks 24 and 25e0

Verbatim

e0
“The RMSE of the forecasts started to decrease in weeks 24 and 25.”
✓ located verbatim in the source · abstract, characters 1565–1630

Title

Improving operational maize yield forecasting in Hungary

Attila Bussay; Marijn van der Velde; Davide Fumagalli; et al. Agricultural Systems 2015. doi:10.1016/j.agsy.2015.10.001

Other claims from this paper (7)

Method / resource A LOESS trend had the lowest yield time-series RMSE versus quadratic and linear trends.
Method / resource WLLAI and WLB predictors with the LOESS trend produced the lowest RMSEs.
Method / resource Spatial aggregation levels showed only marginal forecast differences in the first half of the maize crop cycle.
Method / resource National NUTS0 forecasts were slightly less accurate from flowering onward.
Method / resource The best yield forecasts used WLLAI and WLB predictors about 5–8 weeks before harvest.
Method / resource The best forecasts aligned with flowering in weeks 27 to 30 and grain filling in weeks 31 to 35.
Method / resource The best national forecast from NUTS1 forecasts using WLLAI had r2 of 0.8565 and RMSE of 425.9 kg ha− 1.

Method / resourceobservedoriginal resultpositivesimulated

Claim · model-written summary

The 2D model decreased nRMSE for grain yield by 1.5–2.5 % points compared with the 1D model.

Outcome: nRMSE for grain yielde0
Direction: decreasee0
Value: 1.5–2.5 %e0
Comparator: 1D modele0

Verbatim

e0
“a decrease of 1.5–2.5 % and 2.5–4.8 % points in nRMSE for grain yield and AB, respectively, compared with those of 1D model”
✓ located verbatim in the source · abstract, characters 1525–1648

Title

Establishing and assessing a new two-dimensional module in CERES-maize for maize production under mulched drip irrigation

Dan Wang; Yan Mo; Juan Xiao; et al. Agricultural Water Management 2025. doi:10.1016/j.agwat.2025.109824

Other claims from this paper (8)

Method / resource The modified 2D CERES-Maize model decreased the Absolute Relative Error for maize emergence date by 11.2 % points.
Method / resource The modified model decreased the Absolute Relative Error for maize anthesis date by 1.8 % points.
Method / resource The modified model decreased the Absolute Relative Error for maize maturity date by 2.1 % points.
Method / resource The modified model reduced the normalized RMSE for grain number per ear by 1.5 % points.
Method / resource The modified model reduced the normalized RMSE for aboveground biomass by 5.9 % points.
Method / resource The 2D model decreased nRMSE for aboveground biomass by 2.5–4.8 % points compared with the 1D model.
Method / resource The 2D model simulated the differences of maize actual evapotranspiration under different irrigation quotas very well (nRMSE below 10 %).
Method / resource The 2D model decreased nRMSE for maize actual evapotranspiration by 2.2–5.9 % points compared with the 1D model.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Adding APSIM variables reduced corn yield prediction RMSE by 7 to 20%.

Taxon: cornZea mays subsp. mayst0
Treatment: adding simulation crop model variables (APSIM) as input features to ML modelse0
Outcome: yield prediction root mean squared error (RMSE)e0
Direction: decreasee0
Value: 7 to 20%e0

Verbatim

e0
“adding simulation crop model variables (APSIM) as input features to ML models can decrease yield prediction root mean squared error (RMSE) from 7 to 20%”
✓ located verbatim in the source · abstract, characters 647–799

t0
“Coupling machine learning and crop modeling improves crop yield prediction in the US Corn Belt”
✓ located verbatim in the source · title, characters 0–94

Title

Coupling machine learning and crop modeling improves crop yield prediction in the US Corn Belt

Mohsen Shahhosseini; Guiping Hu; Isaiah Huber; et al. Scientific Reports 2021. doi:10.1038/s41598-020-80820-1

Other claims from this paper (3)

Method / resource Soil moisture APSIM variables were most influential, followed by crop and phenology variables.
Method / resource Simulated average drought stress and average water table depth were the most important APSIM inputs.
Recommendation The authors conclude that weather information alone is insufficient for improved yield predictions.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The system reduced fertilizer consumption by 58.82%.

Taxon: tomatoSolanum lycopersicume1
Treatment: fuzzy logic-based algorithme2
Outcome: fertilizer consumptione0
Direction: decreasee0
Value: 58.82%e0

Verbatim

e0
“58.82% decrease in fertilizer consumption”
✓ located verbatim in the source · abstract, characters 534–575

e1
“tomato yield”
✓ located verbatim in the source · abstract, characters 601–613

e2
“fuzzy logic-based algorithm to optimize bio-fertigation”
✓ located verbatim in the source · abstract, characters 173–228

Title

Development of a fuzzy logic-based greenhouse system for optimizing bio-fertigation

Achouak Touhami; Amina Bourouis; Amel Mahammedi; et al. International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering 2025. doi:10.11591/ijece.v15i5.pp4555-4568

Other claims from this paper (4)

Method / resource The system reduced water use by 27.58%.
Method / resource The system increased tomato yield by 47.5%.
Method / resource MAE, MSE and RMSE were reduced to zero.
Method / resource MAPE was reduced to zero.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

SymbioMamba has a 50% decrease in parameter count relative to monolithic state-space baselines.

Taxon: maizeZea mayse0
Outcome: parameter counte1
Direction: decreasee1
Value: 50%e1
Comparator: monolithic state-space baselinese1

Verbatim

e0
“Experimental results from a maize UAV dataset comprising 12,074 annotated image patches demonstrate that SymbioMamba achieves 89.4% mAP@0.5 and an R2 of 0.915”
✓ located verbatim in the source · abstract, characters 1151–1309

e1
“Compared to the industry-standard YOLOv11, the framework improves mAP@0.5:0.95 by 2.4% while reducing the parameter count to 6.2 M—a 50% decrease relative to monolithic state-space baselines”
✓ located verbatim in the source · abstract, characters 1311–1501

Title

SymbioMamba: An Efficient Dual-Stream State-Space Framework for Real-Time Maize Disease and Yield Analysis on UAV Platforms

Zihuan Wang; Yuru Wang; Bocheng Zhou; et al. Agriculture 2026. doi:10.3390/agriculture16070801

Other claims from this paper (7)

Method / resource SymbioMamba achieves 89.4% mAP@0.5 on the maize UAV dataset.
Method / resource SymbioMamba achieves an R2 of 0.915 on the maize UAV dataset.
Method / resource SymbioMamba improves mAP@0.5:0.95 by 2.4% over the industry-standard YOLOv11.
Method / resource SymbioMamba reduces the parameter count to 6.2 M relative to monolithic state-space baselines.
Method / resource Yield prediction error is significantly reduced to an RMSE of 485.6 kg/ha.
Method / resource SymbioMamba runs at 38.2 FPS on the NVIDIA Jetson AGX Orin platform.
Method / resource SymbioMamba has a compact model size of 2.4 G FLOPs.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

RF feature selection reduced bias by 67%.

Outcome: biase0
Direction: decreasee0
Value: 67%e0

Verbatim

e0
“resulting in a 29.41% reduction in Root Mean Square Error (RMSE) and a 67% decrease in bias”
✓ located verbatim in the source · abstract, characters 1626–1717

e1
“The RF model explained 44% of the variance in the data”
✓ located verbatim in the source · abstract, characters 1570–1624

Title

Estimating paddy rice yield using PlanetScope imagery and machine learning

Sa’ad Ibrahim; Heiko Balzter; Mohammed S. Ozigis Smart Agricultural Technology 2025. doi:10.1016/j.atech.2025.101447

Other claims from this paper (14)

Method / resource All tested machine-learning algorithms performed best with RFE and MIR feature selection.
Method / resource k-NN was the best yield model, with R² of 0.61.
Method / resource k-NN yielded RMSE of 578.43 kg/ha.
Method / resource ANN yield prediction had R² of 0.58.
Method / resource ANN yield prediction had RMSE of 601.29 kg/ha.
Method / resource RF yield prediction had R² of 0.44.
Method / resource RF yield prediction had RMSE of 694.48 kg/ha.
Method / resource XGBoost yield prediction had R² of 0.34.
Method / resource XGBoost yield prediction had RMSE of 756.48 kg/ha.
Method / resource ANN performed optimally with only a few spectral features.
Method / resource RF explained 44% of yield variance.
Method / resource RF feature selection reduced RMSE by 29.41%.
Method / resource XGBoost explained 13.33% more yield variance than the previous model.
Method / resource XGBoost reduced RMSE by 3.25% against the previous model.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Adding APSIM simulated variables as ML input features decreased yield prediction RMSE by 7 to 20%.

Taxon: cornZea mays subsp. mayse0
Treatment: adding simulation crop model variables (APSIM) as input features to ML modelse1
Outcome: yield prediction root mean squared error (RMSE)e1
Direction: decreasee1
Value: from 7 to 20%e1

Verbatim

e0
“investigates whether coupling crop modeling and machine learning (ML) improves corn yield predictions in the US Corn Belt”
✓ located verbatim in the source · abstract, characters 11–132

e1
“adding simulation crop model variables (APSIM) as input features to ML models can decrease yield prediction root mean squared error (RMSE) from 7 to 20%”
✓ located verbatim in the source · abstract, characters 647–799

Title

Coupling Machine Learning and Crop Modeling Improves Crop Yield Prediction in the US Corn Belt

Shahhosseini, Mohsen; Hu, Guiping; Huber, Isaiah; et al. Iowa State University Digital Repository (Iowa State University) 2021

Other claims from this paper (3)

Method / resourceobservedoriginal resultbasis n/a

Claim · model-written summary

Equations were formulated for yield and relative evapotranspiration responses to saline and pure irrigation water.

Taxon: sorghumSorghumt0
Outcome: real Yield and relative evaporation transpiratione0
Direction: increase or decreasee0
Comparator: salt water and pure water additivese0

Verbatim

e0
“Mathematical equations have been formulation to find out the increase or decrease in the real Yield and relative evaporation transpiration as a result of the influence of the amount of salt water and pure water additives under the conditions and limitations of the study area”
✓ located verbatim in the source · abstract, characters 1102–1377

t0
“sorghum”
✓ located verbatim in the source · title, characters 131–138

Title

Modeling Effect of Irrigation Water quality Frequency with difference Leaching requirement on the salts accumulation and yield for sorghum

Omar Mkdad Abdul Guny Mahmood Maǧallaẗ al-handasaẗ al-rāfidayn 2012. doi:10.33899/rengj.2012.54145

Other claims from this paper (5)

Observation Modeled soil salt accumulation varied by irrigation style.
Intervention effect Continuous irrigation increased modeled soil salt accumulation.
Intervention effect Alternant irrigation reduced modeled soil salt accumulation.
Intervention effect Alternant irrigation increased modeled sorghum production.
Association Reduced salt accumulation depended on alternant irrigation method and leaching requirements.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

An 11-year moving average reduced forecast deviation to 5-6%.

Taxon: Spring wheatTriticum aestivumt0
Outcome: deviatione0
Direction: decreasee0
Value: 5-6 %e0
Qualifier: step length of 11 yearse0
Comparator: moving average of crop yielde0

Verbatim

e0
“the moving average of crop yield for the step length of 11 years gives a marked decrease in the deviation of 5-6 %”
✓ located verbatim in the source · abstract, characters 1498–1612

t0
“Spring wheat”
✓ located verbatim in the source · title, characters 0–12

Title

Spring wheat productivity forecasting in the forest steppe zone of Volga by least squares method

Аввакумов; Oleg Avvakumov Vestnik of Kazan state agrarin university 2014. doi:10.12737/2167

Other claims from this paper (6)

Association Actual yield correlated 0.52 with phosphorus.
Association Actual yield correlated 0.61 with potassium.
Association Moving-average yield correlated 0.94 with phosphorus.
Association Moving-average yield correlated 0.95 with potassium.
Method / resource Forecasts of actual spring wheat yield had 30% average deviation.
Recommendation Authors recommend including precipitation and temperature in yield forecasts.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Combining S1 with S2 reduced RMSE by 30% relative to S2 alone.

Taxon: wheatTriticum aestivume1
Outcome: RMSEe0
Direction: decreasee0
Value: 30%e0
Comparator: S2 alonee0

Verbatim

e0
“a decrease of 30% in RMSE when compared to using S2 alone”
✓ located verbatim in the source · abstract, characters 1588–1645

e1
“wheat (Triticum spp. L.) fields”
✓ located verbatim in the source · abstract, characters 752–783

Title

Optimizing Wheat Yield Prediction Integrating Data from Sentinel-1 and Sentinel-2 with CatBoost Algorithm

Asier Uribeetxebarria; Ander Castellón; Ana Aizpurua Remote Sensing 2023. doi:10.3390/rs15061640

Other claims from this paper (5)

Method / resource Combined S1 and S2 data with CatBoost predicted wheat yield with RMSE 0.24 t ha−1.
Method / resource Combined S1 and S2 data with CatBoost had relative RMSE 3.46%.
Method / resource Combined S1 and S2 data with CatBoost had R2 of 0.95.
Method / resource Whole-plot estimation using surrounding plots had MAE 0.31 t ha−1.
Method / resource Whole-plot estimation had mean error 4.38%.

Lexical search (BM25). All words must match; if nothing does, any word. Organism names also match every spelling of the same organism. No semantic search yet.

“yield decrease”

Claim types

Direction of the findings

Over time

Claim type

Research area

Polarity

Statistics