agragent logo
Precision & Digital Agriculture
Dashboard
Season 2025–2026 Upload a polygon to start
📍
Welcome to agragent
To get started, go to Satellite Maps and upload a KML/KMZ file or draw a polygon over your field. The dashboard will update automatically with real data from the location.

Satellite Maps

Google Earth Engine · Sentinel-2 · Real Imagery
🌍
Google Earth Engine
Conectando automáticamente a Sentinel-2...
💡 Sentinel-2 SR Harmonized · 10m resolution · Auto cloud masking (QA60) · Polygon area only
🌿 Vigor
🧪 Salud
💧 Agua
🟫 Suelo
🎨 Visual
GEE Activo
Loading GEE imagery...
🗺️

Drop your KMZ/KML file here

NDVI — Índice de Vegetación
0.0 (Sin veg.)0.30.60.80.9 (Alta)
Fórmula: (NIR − RED) / (NIR + RED) · Bandas B8, B4
Active layers 0 capas
Upload a KMZ/KML or draw to add layers.
🛰️ Sentinel-2 · OpenStreetMap base · GEE connected

Climate Data

Upload a polygon to see climate data
Upload a KML/KMZ or draw a polygon to load climate data for the location.
Annual GDD
Growing Degree Days (base 10°C)
Chill Hours
Hours 0–7.2°C (Apr–Sep)
Frost Days
Days with Tmin ≤ 0°C
Heat Waves
≥3 consecutive days >35°C
Dry Months
<30mm precipitation
Avg Humidity
Avg Humidity (%)
Total Solar Radiation
MJ/m²
Total ET₀
Evapotranspiration (mm)
Monthly Temperature (Min/Max) — 12-Month Season
Monthly Precipitation (mm)
Monthly Relative Humidity (%)
Monthly Solar Radiation (MJ/m²)
Monthly Evapotranspiration ET₀ (mm)
Growing Degree Days (GDD) Accumulation

Historical Analysis (2015–2026)

Multi-season comparison
Annual GDD Comparison by Season
Annual Precipitation by Season (mm)
Average Temperature by Season (°C)
Frost Days per Season
Heat Stress Days per Season (>35°C)
Season Comparison Table
Upload a polygon to load historical data.

Genomic Analysis

RNA-seq Differential Expression — Vitis vinifera phenological stages (Altimiras et al.)
Total Reads
1,336M
68 RNA-seq samples
Data Processed
87.0 GB
FASTQ files
Phenological Stages
8
E-L 3 → E-L 41
Unique DEGs
3,603
FDR < 0.05 (edgeR)
Phenological Stage Progress
68%
Season
Progress
Veraison
Stage 35 — Modified E-L Scale
EL 3
Bud
EL 15
Inflor.
EL 27
Berry
EL 35
Veraison
EL 38
Ripen
EL 41
Senesc.
DEGs per Developmental Stage (vs. E-L 3 baseline)
Sample Composition by Variety
Muscat Blanc a Petits Grains 24 samples
Corvina 9 samples
Cabernet Sauvignon 6 samples
Sangiovese 3 samples
Sangiovese somatic variant 3 samples
V. vinifera sylvestris (wild) 23 samples
DEG Accumulation Across Stages
Top Differentially Expressed Genes by Stage
Gene IDLog2FCFDRStageRegulation
VIT_03s0063g01440+7.711.89e-16Shoot & Inflor.UP
VIT_01s0011g00140+6.761.37e-14Shoot & Inflor.UP
VIT_19s0014g03940+5.625.15e-15Shoot & Inflor.UP
VIT_10s0003g02070+4.931.74e-24Shoot & Inflor.UP
VIT_02s0025g00200+4.001.70e-14Shoot & Inflor.UP
VIT_12s0142g00360+3.991.68e-21Shoot & Inflor.UP
Differential Expression Summary — All Comparisons vs. E-L 3 (Baseline)
ComparisonMajor StageTissueReplicatesDown-regulatedUp-regulatedTotal DEGs
EL 3 → 5Shoot & Inflor.Flower bud6000
EL 3 → 15Shoot & Inflor.Flower bud8118495
EL 3 → 17Shoot & Inflor.Flower bud59180171
EL 3 → 27Berry dev.Flower2304214518
EL 3 → 31Berry dev.Grape berry64555901,045
EL 3 → 35RipeningGrape berry131,0525961,648
EL 3 → 36RipeningGrape berry42,2596342,893
EL 3 → 38RipeningGrape berry151,6756612,336
EL 3 → 41SenescenceFruit32,5638403,403

Image Analysis

Embrapa WGISD · Wine Grape Instance Segmentation Dataset
Embrapa Wine Grape Instance Segmentation Dataset (WGISD)
Thiago T. Santos et al. · Embrapa · CC BY-NC 4.0 · github.com/thsant/wgisd
300Imágenes
4,431Racimos
187KBayas anotadas
5Variedades
65 imágenes · 840 racimos anotados · 308 con máscara de instancia · Canon EOS REBEL T3i · 2048×1365px Clic en imagen para ver anotaciones
📤 Analizar imagen propia
🖼️
Arrastra o haz clic para subir imagen de viñedo
Acepta .jpg · .png · .tiff — Máx 50MB

Yield Prediction

Extra Trees Regressor — Season 2025/2026
Predicted Yield
9.8
tonnes per hectare
Confidence Interval: ± 0.8 t/ha (9.0 – 10.6)
Extra Trees Regressor R² = 0.972 MAPE = 3.8%
Feature Importance
Precipitation 0.92
GDD Accumulated 0.87
Max Temperature 0.81
NDVI 0.74
NDRE 0.65
MSAVI 0.58
TCARI 0.45
Parcel Predictions
ParcelAreaPredicted±CITrend
Parcel A4.2 ha10.3 t/ha±0.7▲ +0.4
Parcel B3.8 ha8.9 t/ha±1.0▼ −0.6
Parcel C5.1 ha10.1 t/ha±0.8▲ +0.2
Parcel D2.9 ha9.2 t/ha±0.9— 0.0
Parcel E6.3 ha10.6 t/ha±0.6▲ +0.5
Historical Yield: Predicted vs Actual (2020–2025)
Model Comparison
ModelMAPE (%)RMSETrain TimeBest
Extra Trees 0.972 3.8 0.41 2.3s
CatBoost 0.9694.10.44 18.7s
Random Forest 0.9644.50.47 3.1s
XGBoost 0.9585.00.52 6.4s
SVR 0.9316.80.66 0.9s

References & Credits

Author, publications, and data sources
Author
FA
Dr. Francisco Altimiras
PhD in Computer Science
Associated Publications
Transcriptome Data Analysis Applied to Grapevine Growth Stage Identification
Altimiras, F. et al. — Agronomy, 14(3), 613, 2024
Source of genomic data: RNA-seq pipeline, DEGs, phenological stage comparisons (Tables S1–S5)
A Computational Framework for Crop Yield Estimation and Phenological Monitoring
Altimiras, F. et al. — Progress in Artificial Intelligence, EPIA 2024. LNCS vol. 15400, Springer, 2025
Datasets & Data Sources
ResourceDescriptionUsage in agragent
WGISD Wine Grape Instance Segmentation Dataset (Embrapa). 300+ images, 5 varieties, YOLO bounding boxes.
Santos, T.T. et al. Computers and Electronics in Agriculture, 2020
Image Analysis section
Sentinel-2 SR COPERNICUS/S2_SR_HARMONIZED — 10m multispectral satellite imagery via Google Earth Engine Satellite Maps (NDVI, EVI, SAVI, MSAVI, NDRE, GNDVI, TCARI, CIre, NDMI, NDWI, BSI)
Open-Meteo Free open-source weather API — historical daily temperature and precipitation Climate Data section
RNA-seq (ArrayExpress) 68 Vitis vinifera samples across 8 E-L stages. Processed with Salmon + edgeR pipeline Genomic Analysis section
License & Source Code

This project is open source under the MIT License.

White Paper: Altimiras et al. (2026)

Conversations

AgrAgent

Ask me about climate data, satellite imagery, soil analysis, irrigation planning, or any agronomic question about your fields.

AgrAgent

AgrAgent

Ask me about climate data, satellite imagery, soil analysis, irrigation planning, or any agronomic question about your fields.

✏️ Propiedades de la Capa