İklim değişiminin buğday verimine etkilerinin incelenmesi: Kırklareli örneği.

The Thrace Region is located in The European part of the northwestern Turkey, which is responsible for a considerable portion of rice, sunflower and winter wheat productions of Turkey. Winter wheat (Triticum Aestivum L.) production rate in Thrace is generally higher than that in the rest of Turkey. However, there is not enough qualified study on the effects of drought and possible climate change on the region's agriculture. The dynamics and complex structure of the plant growth are under the influence of many factors which are out of human control. Meteorological parameters are the dominant ones among these factors. Hence, reliable research studies are to be conducted from the agricultural mete-orological viewpoint. Using crop growth simulation models, possible effects of the variations in plant, soil and atmosphere on wheat development and yield can be estimated. In this study, the CERES-Wheat crop growth simulation model have been used for wheat during the years between 1975 and 2005 to take a first step for filling the gaps in the desired drought related research in the Kırklareli City, in which the winter wheat production is popular. At first, the average growing season have been deter-mined by considering long term phenological data (3 November-22 July). Afterwards, the associated soil-and crop parameters have been supplied into the model database as well as the calculated daily average meteorological data (daily totals for precipitation). Average wheat grain yield was calculated as 2827 kg/ha by the model with an relative error of 3.1%. Model sensitivities to temperature, precipitation, global radiation and CO2 have been done by considering single and combined variations. The (Rg30%; CO2x4) combined application gave the highest yield increase response as 67.1%, where the opposite output was a 63.1% decrease for the (T+5;P-40%) application. Another step was the investigation of the development and yield reactions of the crop to the IPCC A2 scenarios from 2071 to 2100. To achieve this, model database used daily output from the RegCm3 Regional Climate Model by considering the associated field coordinates. The model showed a 9% increase in winter wheat grain yields during the mentioned time interval. It has been determined that the winter wheat yield was sensitive especially to variations in T and CO2.This study represents a part of the PhD named "Determination of the Agrometeorological Properties of the Thrace Region", in which additional cities (Edirne and Tekirdağ) have been investigated from the point of wheat production.30% yield increase has been simulated for Edirne, while a 13% decrease was expected for Tekirdağ during the aver-aged 2071-2100 growing period. Rice production has been also estimated for Edirne with a decrease of 53%. RegCm3 data were also arranged and adapted into CERES meteorology files to enable model runs for the hottest and driest growing seasons for the 2071-2100 interval. Moreover; three drought indices, namely the percentage of normal, z score and standardized precipitation indices have been calculated and interpreted for seasonal field crops (wheat, rice and sunflower) and forests (pine, oak and beech) during the years between 1975 and 2005 to take a first step for filling the gaps in the desired drought related research in the region. In general, dry periods were determined with low severities and continuities. Growing Degree Day (GDD) and vernalization calculations for field crops have been also applied. Among the average growing seasons in Kırklareli, the winter wheat collected 2893 GDD's for 0°C base temperature. Same parameter was calculated as 2719 and 2639 for Tekirdağ and Edirne, successively. It has been calculated that the wheat needs 70 and more vernalization days to complete it's average growth. Variations in the Normalized Difference Vegetation Index (NDVI) on the agricultural and forest areas were also estimated using the satellite data. Furthermore, correlations of the drought indices with the GDD's and NDVI's were investigated. Highest average NDVI values were calculated for the flowering periods of wheat (NDVI=0.45) and rice (NDVI=0.50), consequently. June showed the highest mean NDVI's for pine (NDVI=0.4), oak and beech (NDVI=0.65) when the forest areas were considered. Some growing periods showed nearly linear correlations between SPI and NDVI, while this was not the case when the whole growing periods were considered. On the other hand, high nonlinear relationships between NDVI and BDG were estimated for both field crops and trees. [ABSTRACT FROM AUTHOR]