Rahmani, Elham: The effect of climate variability on wheat in Iran. - Bonn, 2015. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5n-38793
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5n-38793,
author = {{Elham Rahmani}},
title = {The effect of climate variability on wheat in Iran},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2015,
month = jan,

volume = 67,
note = {In this study, we investigated the impact of temperature variability on wheat phenology in Iran. Temperature is the most appropriate climate variable affecting wheat production wheat in cultivation under irrigation in Iran.
To that aim, an effective and potentially scalable statistical downscaling method is developed for temperature and growing degree days (GDD) of wheat. Statistical downscaling quantitatively establishes statistical links between the large-scale reanalysis or climate model and regional climate data. GDD is the atmospheric energy that a plant utilizes to grow over the phenological phases until the harvesting stage. The GDD values are calculated during the growth period from the phenological dates and the daily mean temperature data of observations and reanalysis.
The underlying database in downscaling comprises the ERA-40 reanalysis for the global scale and observations of local daily temperature and annual GDD of wheat at 16 synoptic stations for the period 1961-2001 for the regional scale. For the quantitative analysis of the statistical downscaling, we used the linear regression model (LR) and multiple regression model (MR). The LR is implemented using the ERA-40 fingerprints (FP) of local variability by squared correlation coefficients between the variable at ERA-40 grid points and each station. The MR technique is performed to relate the large-scale information at the neighboring grid points to the stations data. Extending the usual downscaling, we implement a weather generator (WG) providing realizations of the local temperatures and GDD by adding Gaussian random noise with expectation zero and the variance between the downscaled values and the observations.
ERA-40 reanalysis well represents the local daily temperature and the annual GDD. From the analysis of 2m temperature, FPs are more localized in warm seasons than cold seasons. FP statistical downscaling seems to perform best for annual GDD and it is particularly beneficial for the annual GDD. Whereas, the MR calculated robust results for daily mean temperature time series. The quality of the WGs is assessed along with verification score such as the continuous ranked probability score, CRPS. The local temperature time series through WGs are more realistic and well represented than the deterministic downscaling.
As a next step, the probabilistic wheat model is developed. It represents the probabilistic relations between the phenological and climate parameters. The basic idea of the model is to interpret a survival function which is based on the normal distribution, on a time scale which is defined by lifetime or growth duration for wheat. The probabilistic phenological model is adjusted by the survival analysis (SA) considering the risk in interpreting the maturity time of wheat. SA is a statistical method to study the occurrence and timing of event which here is the ripening time of wheat from the random variable of ripening dates.
In summary, we believe that the probabilistic phenological model have the potential to reduce the vulnerability of agricultural production system and can increase the food security in the region.},

url = {https://hdl.handle.net/20.500.11811/6396}

Die folgenden Nutzungsbestimmungen sind mit dieser Ressource verbunden: