Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model
Seasonal climate variability and agronomic management profoundly influence both the productivity and nutritive value of temperate hay meadows. We analyzed five years of data (2019, 2020, 2022–2024) from 15 meadows in the central Spanish Pyrenees to quantify how environmental variables (January–June...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Plants |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2223-7747/14/14/2150 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849418356649623552 |
|---|---|
| author | Adrián Jarne Asunción Usón Ramón Reiné |
| author_facet | Adrián Jarne Asunción Usón Ramón Reiné |
| author_sort | Adrián Jarne |
| collection | DOAJ |
| description | Seasonal climate variability and agronomic management profoundly influence both the productivity and nutritive value of temperate hay meadows. We analyzed five years of data (2019, 2020, 2022–2024) from 15 meadows in the central Spanish Pyrenees to quantify how environmental variables (January–June minimum temperatures, rainfall), management variables (fertilization rates (N, P, K), livestock load, cutting date), and vegetation (plant biodiversity (Shannon index)) drive total biomass yield (kg ha<sup>−1</sup>), protein content (%), and Relative Feed Value (RFV). Using Random Forest regression with rigorous cross-validation, our yield model achieved an R<sup>2</sup> of 0.802 (RMSE = 983.8 kg ha<sup>−1</sup>), the protein model an R<sup>2</sup> of 0.786 (RMSE = 1.71%), and the RFV model an R<sup>2</sup> of 0.718 (RMSE = 13.86). Variable importance analyses revealed that March rainfall was the dominant predictor of yield (importance = 0.430), reflecting the critical role of early-spring moisture in tiller establishment and canopy development. In contrast, cutting date exerted the greatest influence on protein (importance = 0.366) and RFV (importance = 0.344), underscoring the sensitivity of forage quality to harvest timing. Lower minimum temperatures—particularly in March and May—and moderate livestock densities (up to 1 LU) were also positively associated with enhanced protein and RFV, whereas higher biodiversity (Shannon ≥ 3) produced modest gains in feed quality without substantial yield penalties. These findings suggest that adaptive management—prioritizing soil moisture conservation in early spring, timely harvesting, balanced grazing intensity, and maintenance of plant diversity—can optimize both the quantity and quality of hay meadow biomass under variable climatic conditions. |
| format | Article |
| id | doaj-art-5cb74e02f3cd46f984ca5e9560b82446 |
| institution | Kabale University |
| issn | 2223-7747 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Plants |
| spelling | doaj-art-5cb74e02f3cd46f984ca5e9560b824462025-08-20T03:32:27ZengMDPI AGPlants2223-77472025-07-011414215010.3390/plants14142150Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest ModelAdrián Jarne0Asunción Usón1Ramón Reiné2Departamento de Ciencias Agrarias y del Medio Natural, Escuela Politécnica Superior, Universidad de Zaragoza, Ctra Cuarte s/n, 22071 Huesca, SpainDepartamento de Ciencias Agrarias y del Medio Natural, Escuela Politécnica Superior, Universidad de Zaragoza, Ctra Cuarte s/n, 22071 Huesca, SpainDepartamento de Ciencias Agrarias y del Medio Natural, Escuela Politécnica Superior, Universidad de Zaragoza, Ctra Cuarte s/n, 22071 Huesca, SpainSeasonal climate variability and agronomic management profoundly influence both the productivity and nutritive value of temperate hay meadows. We analyzed five years of data (2019, 2020, 2022–2024) from 15 meadows in the central Spanish Pyrenees to quantify how environmental variables (January–June minimum temperatures, rainfall), management variables (fertilization rates (N, P, K), livestock load, cutting date), and vegetation (plant biodiversity (Shannon index)) drive total biomass yield (kg ha<sup>−1</sup>), protein content (%), and Relative Feed Value (RFV). Using Random Forest regression with rigorous cross-validation, our yield model achieved an R<sup>2</sup> of 0.802 (RMSE = 983.8 kg ha<sup>−1</sup>), the protein model an R<sup>2</sup> of 0.786 (RMSE = 1.71%), and the RFV model an R<sup>2</sup> of 0.718 (RMSE = 13.86). Variable importance analyses revealed that March rainfall was the dominant predictor of yield (importance = 0.430), reflecting the critical role of early-spring moisture in tiller establishment and canopy development. In contrast, cutting date exerted the greatest influence on protein (importance = 0.366) and RFV (importance = 0.344), underscoring the sensitivity of forage quality to harvest timing. Lower minimum temperatures—particularly in March and May—and moderate livestock densities (up to 1 LU) were also positively associated with enhanced protein and RFV, whereas higher biodiversity (Shannon ≥ 3) produced modest gains in feed quality without substantial yield penalties. These findings suggest that adaptive management—prioritizing soil moisture conservation in early spring, timely harvesting, balanced grazing intensity, and maintenance of plant diversity—can optimize both the quantity and quality of hay meadow biomass under variable climatic conditions.https://www.mdpi.com/2223-7747/14/14/2150grasslandprotein contentrelative feed valuecutting datemodeling |
| spellingShingle | Adrián Jarne Asunción Usón Ramón Reiné Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model Plants grassland protein content relative feed value cutting date modeling |
| title | Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model |
| title_full | Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model |
| title_fullStr | Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model |
| title_full_unstemmed | Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model |
| title_short | Assessing the Impact of Environmental and Management Variables on Mountain Meadow Yield and Feed Quality Using a Random Forest Model |
| title_sort | assessing the impact of environmental and management variables on mountain meadow yield and feed quality using a random forest model |
| topic | grassland protein content relative feed value cutting date modeling |
| url | https://www.mdpi.com/2223-7747/14/14/2150 |
| work_keys_str_mv | AT adrianjarne assessingtheimpactofenvironmentalandmanagementvariablesonmountainmeadowyieldandfeedqualityusingarandomforestmodel AT asuncionuson assessingtheimpactofenvironmentalandmanagementvariablesonmountainmeadowyieldandfeedqualityusingarandomforestmodel AT ramonreine assessingtheimpactofenvironmentalandmanagementvariablesonmountainmeadowyieldandfeedqualityusingarandomforestmodel |