Adresse
Infodoc : Réseau des bibliothèques et centres de documentation d'AgroParisTechFrance
contact
Array ( [TITRE] => <b>Type de document : </b> [TITRE_CLEAN] => Type de document [OPAC_SHOW] => 1 [TYPE] => list [AFF] => Mémoire [ID] => 4 [NAME] => cp_typdoc [DATATYPE] => integer [VALUES] => Array ( [0] => 4 ) )

Titre : |
Bayesian modeling to predict pistachio masting
|
Auteur(s) : |
Lolita MULLER, Auteur (et co-auteur)
Maguy Dulormne, Responsable de stage Université des Antilles, Auteur (et co-auteur) |
Type de document : | Mémoire |
Filière : | M. : BEE -- Biodiversité Ecologie Evolution (comprend les parcours BIOGET et EFT ) |
Sujets : | Pistache ; Changement climatique ; Simulation |
Résumé : |
Masting, the synchronous and highly variable production of fruits and seeds by perennial plants, is a phenomenon observed in various fruit and nut trees worldwide, influencing agricultural practices and food markets. Crop yield prediction models are essential decision support tools for farmers, helping in decisions regarding crop management and agricultural practices. With climate change becoming an increasing concern, the development of models to predict yields of masting trees has gained significance, necessitating high-quality data and appropriate analytical tools. In this study, we created Bayesian models to predict pistachio yield and investigate the potential impacts of climate change on pistachio production in California. Pistachio makes an interesting model species for this in[...] Masting, the synchronous and highly variable production of fruits and seeds by perennial plants, is a phenomenon observed in various fruit and nut trees worldwide, influencing agricultural practices and food markets. Crop yield prediction models are essential decision support tools for farmers, helping in decisions regarding crop management and agricultural practices. With climate change becoming an increasing concern, the development of models to predict yields of masting trees has gained significance, necessitating high-quality data and appropriate analytical tools. In this study, we created Bayesian models to predict pistachio yield and investigate the potential impacts of climate change on pistachio production in California. Pistachio makes an interesting model species for this investigation because it is, a high-value tree species showing masting behavior. Despite their adaptability to dry conditions, pistachios are susceptible to the effects of climate change. Achieving satisfactory yields requires an average of 850 chilling hours annually.To forecast future pistachio yields, a Bayesian model was developed utilizing a unique data set that tracked individual tree growth and reproduction over seven years in Kings County, California. Additionally, chilling hour projections were calculated for all counties in California under two climate scenarios: RCP4.5 and RCP8.5, to assess future pistachio cultivation opportunities. We successfully built a model that is able to predict pistachio yield demonstrating the power of Bayesian modeling for yield prediction. The model had a mean absolute error of 2.71 representing a ±5% error. The projections indicate a significant reduction in chilling hours under both RCP4.5and RCP8.5 scenarios, resulting in a severe decline in pistachio yields by 2040. Yield projections decreased by 50% and 25% under RCP8.5 and RCP4.5 scenarios respectively. Additionnaly pistachio production will be threatened by the impact of climate change on masting cycles which still remains unclear.The majority of counties currently cultivating pistachios are projected to face challenges in meeting the future chilling requirements, posing a significant threat to pistachio production in Cali-fornia. However, counties in the northern regions are expected to experience fewer frost events and more favorable chilling hours, offering potential prospects for future pistachio cultivation. In order to ensure a smooth transition and timely availability of mature pistachio trees in the future, it is imperative to implement adaptive measures without delay. |
Date de publication : | 2023 |
Format : | 39 p. / graph.coul., carte coul. |
Langue(s) : | Anglais |
Lien vers la notice : | https://infodoc.agroparistech.fr/index.php?lvl=notice_display&id=222628 |
Exemplaires (1)
Localisation | Emplacement | Pôle | Section | Cote | Support | Disponibilité |
---|---|---|---|---|---|---|
Kourou | Bibliothèque | AgroParisTech-Kourou | - | Numérique | Consultable sous conditions Disponible |