Eddy Kovaryans Yöntemiyle Buğday Bitkisinin Karbondioksit Akılarının Belirlenmesi

Abstract

Farklı ekosistemlerin karbon bütçelerini belirlemek küresel karbon döngüsünü anlayabilmek açısından oldukça önemli bir konudur. Özellikle küresel ısınmanın en büyük sebebinin insan kaynaklı sera gazlarının (karbondioksit ve metan başta olmak üzere) atmosferik konsantrasyonlarının artışı olduğu düşünülünce, bu konuda yapılan bilimsel araştırmaların artış göstermesi yadsınamaz. FLUXNET (Akı Ağı) projesi kapsamında dünya üzerindeki farklı ekosistemlerin karbon bütçesini belirleyebilmek için altı kıtada karbondioksit akıları ölçülmektedir. Akı Ağında ölçüm alanları başta ormanlar olmak üzere, tarım alanları ve çayırlardır. Türkiye ise, bu akı ağı projesine hala dahil değildir. FAO’ya göre Avrupa’nın 1/3’ünü tarım alanları oluşturmaktadır. Bu sebeple toplam karbon döngüsündeki etkileri küçük de olsa tarım alanları üzerinde yapılan ölçümler, farklı tarım ürünleri ekilmesinin ve farklı tarımsal faaliyetler yürütülmesinin karbon bütçesi üzerindeki etkilerini belirleyebilmek için oldukça gereklidir. Bu ölçümlerin uzun dönemler devam ettirilmesi elde edilecek sonuçların güvenilirliği açısından oldukça önemli bir husustur. Bu çalışmada, Kırklareli’nde bulunan Atatürk Toprak Su ve Tarımsal Meteoroloji Araştırma İstasyonu Müdürlüğü’ne ait deneme alanında eddy kovaryans ölçüm sistemi kurularak 2009-2010 ve 2010-2011 gelişme dönemleri için buğday bitkisinin karbondioksit akıları ölçülmüştür. Ölçülen verilere frekans tepki düzeltmeleri, Webb-Pearman-Leuning düzeltmesi ve koordinat düzeltmesi yapılmıştır. Yapılan düzeltmelerden sonra elde edilen verilerden yağış, rüzgar yönü, yetersiz türbülans sebebiyle hatalı kabul edilen veriler atılmış ve eksik veriler uluslararası çalışmalarda kabul görmüş tamamlama yöntemleriyle doldurulmuştur. Bu ölçüm sonuçlarından bitkinin ne kadar karbonu atmosferden aldığını ve ne kadarını atmosfere tekrar geri verdiğini gösteren NEE (Net Ekosistem Değişimi), GPP (Brüt Fotosentez) ve Reco (Ekosistem Solunumu) değerleri hesaplanmıştır. Yapılan çalışma sonucunda kışlık buğdayın toplam NEE, GPP ve Reco değerleri 2009-2010 gelişme döneminde sırasıyla -354.9, 1142.2 ve 788.6 g C m-2 olarak bulunmuştur. Bu dönem için ortalama günlük GPP 4.21 g C m-2, Reco 2.91 g C m-2 ve NEE -1.31 g C m-2 olmuştur. 2010-2011 gelişme döneminde toplam NEE, GPP ve Reco değerleri sırasıyla -441.3 g C m-2, 1046.8 g C m-2 ve 605.5 g C m-2 olmuştur. Bu dönem için ortalama günlük NEE, GPP ve Reco değerleri ise sırasıyla -1.72 g C m-2, 4.09 g C m-2 ve 2.37 g C m-2’dir. Bu çalışmada ayrıca karbon akıları ile meteorolojik faktörler ve yine karbon akıları ile vejetasyon dinamikleri (NDVI, LAI, sPRI, biyokütle) arasında ilişki olup olmadığı incelenmiştir. Karbon akıları ile meteorolojik faktörler arasındaki ilişkiler incelendiğinde, özellikle bitkinin fotosentez aktivitesinin arttığı zaman aralığında PPFD ile GPP arasında doğrusal olmayan iyi bir ilişki olduğu görülmektedir. Reco daha çok toprak sıcaklığıyla bağlantılı çıkmıştır. Ayrıca vejetasyon dinamikleri ile karbon akıları arasında da oldukça yüksek belirlilik katsayısına sahip ilişkiler tespit edilmiştir.

One of the most important research subjects of agricultural meteorology is evaluating the possible effects of agriculture on climate change. Increasing concentrations of greenhouse gases, especially anthropogenic carbon dioxide, is seen as the most significant reason of global warming. Therefore, determination of the carbon dioxide released to and captured from the atmosphere within the components of global carbon cycle is still being searched by scientists. Terrestrial ecosystems are among the most important components of global carbon cycle on earth because they capture carbon dioxide from the atmosphere and release it to the atmosphere again. Hence, evaluation of the effect of plants in these terrestrial ecosystems on the carbon cycle must be done carefully. Additionally, searching the agricultural activities that affect the amount of captured and released carbon (dioxide) is a scientifically supported issue and long-term flux measurements of different ecosystems are required to be able to do all related researches. For this reason, flux networks are established to measure carbon dioxide, water vapor and energy fluxes between different terrestrial ecosystems (forests, croplands, grasslands, peat lands etc.) and the atmosphere across five continents. Most of the micrometeorological systems are installed first in forests then in croplands. The amount of micrometeorological tower sites is increasing in order to measure mass and energy fluxes above canopies. In Turkey, there are limited studies about these issues. To fulfill this lack,, changes in carbon dioxide fluxes during the two growing periods of winter wheat were measured and determined in the Thrace Region using a micrometeorological method called Eddy Covariance for the first time in Turkey in the frame of the TÜBİTAK project named “Determination of Carbon Dioxide, Water Vapor and Energy Fluxes for Winter Wheat”. The research field was located at Atatürk Soil Water and Agricultural Meteorology Research Station, Kırklareli. An eddy covariance measurement system that consists of a 3-dimensional sonic anemometer, open path infrared CO2/H2O gas analyzer, temperature, humidity sensors and data logger was used to calculate fluxes. Additionally, an agrometeorological station measuring air temperature, soil temperature, relative humidity, wind speed and direction (at different heights), precipitation, global radiation and photosynthetically active radiation etc. was installed in the research field. Furthermore, spectral measurements have been done periodically. Normalized difference vegetation index (NDVI) and scaled photochemical reflectance index (sPRI) have also been calculated from spectral measurements. Additional Leaf area index (LAI) measurements have been done once in a two-week period. General principle of eddy covariance measurements is the covariance between concentration of interest and vertical wind speed in eddies. Eddy covariance is a widely used micrometeorological method to calculate turbulent fluxes within the atmospheric boundary layer. Complexity of the method, high cost of the system, correction issues etc. are some disadvantages of this technique. Moreover; experimental knowledge, micrometeorological and agrometeorological background are needed to apply the method correctly. Being a direct measuring method and having no need to empirical constants are the main advantages of this method. In this study; frequency response corrections, WPL (Webb, Pearman, Leuning) correction and coordinate rotation (tilt correction) have been applied to 10 Hz time series of eddy covariance data after de-spiking. Heavy precipitation and low turbulent exchange conditions affect the eddy covariance measurements so that the system produces bad data. Hence, the data gaps caused by the above conditions must be filled. Here, international standard procedures have been followed. Finally, flux partitioning was done and the net ecosystem exchange (NEE), gross primary production (GPP) and ecosystem respiration (Reco) values were calculated during 2009-2010 and 2010-2011 growth period of winter wheat. NEE is positive (+) when carbon dioxide flux is from winter wheat ecosystem to atmosphere and negative when carbon dioxide flux is from atmosphere to winter wheat ecosystem. In the 2009-2010 growth period, winter wheat was planted on 9 October 2009 and harvested on 6 July 2010. The eddy covariance flux measurements showed that cumulative NEE, GPP and Reco were -354.9, 1142.2 and 788.6 g C m-2 respectively for the whole growing period. Daily means of NEE, GPP and Reco for the whole growth period are -1.31, 4.21 and 2.91 g C m-2, succesively. During the 2010-2011 growth period, the crop was sown on 25 October 2010 and harvested on 8 July 2011. For this growth period, cumulative NEE, GPP and Reco have been determined as -441.3, 1046.8 and 605.5 g C m-2, respectively. Similarly, succesive daily means of NEE, GPP and Reco for 2010-2011 of winter wheat were obtained as -1.72, 4.09 and 2.37 g C m-2. A further step was taken by the analysis of possible relationships between winter wheat NEE, GPP and Reco and some meteorological factors such as air temperature, soil temperature and photosynthetically active radiation (PPFD). These relationships were examined by taking different winter wheat growing seasons into account. The first period was between sowing and tillering, the second period was between tillering and bolting, whereas the third period was between bolting and flowering and finally the fourth one was between flowering and harvest. In 2009-2010 growth period, the highest value of the determination coefficient (r2=0.7) between GPP and air temperature has been determined for the third period. Determination coefficient between GPP and soil temperature for the third period was also 0.7. In the other three periods, GPP was less or even not related to air or soil temperature. There were nonlinear relationships between GPP and PPFD during the first (r2=0.72) and the third periods (r2=0.6). For the second and the last periods, no significant relationship could be obtained. The results have shown that NEE was not related to air and soil temperature at all. The relationship between NEE and PPFD was the best (r2=0.82) during the first period. There were strong nonlinear correlations (determination of coefficients are between 0.89 and 0.99) between Reco and both air and soil temperatures during the first and the third periods. Similar relationships between carbon fluxes and meteorological factors have also been determined for the 2010-2011 winter wheat growth period. However, the relationships are not appreciable when compared with these during the previous growth period. Another goal of this study was to evaluate the possible interactions between cumulative carbon fluxes and vegetation indices (NDVI, sPRI, LAI and biomass). To achieve this, the relationships between cumulative NEE, GPP, Reco and NDVI have been determined for the whole 2009-2010 growth period. Determination coefficients were 0.86 for NEE and Reco and 0.95 for GPP. Same relationships have been investigated for two separated periods in case of the 2010-2011 growth season. The first period was the one during which NDVI is continuously rising whereas NDVI was continuously decreasing in the second period. There were strong nonlinear relationships between cumulative carbon fluxes (NEE, GPP and Reco) and NDVI. The determination coefficients between cumulative GPP and NDVI were 0.74 when NDVI was continuously rising and 0.91 when NDVI was continuously decreasing. r2 values for the correlation between cumulative NEE and NDVI were 0.7 while NDVI was continuously rising and 0.99 while NDVI was continuously decreasing. Strong nonlinear relationships have also been found (r2 values are 0.79 while NDVI was rising and 0.997 while NDVI was decreasing) between cumulative Reco and NDVI. Furthermore, it has been found that cumulative carbon fluxes were strongly correlated with LAI both for 2009-2010 and 2010-2011 growing seasons. During the 2009-2010 growth season, the coefficients of determination (r2) between cumulative GPP, NEE, Reco and LAI were 0.97 for the period in which the LAI values were continuously rising (0<LAI<3.7). The determination coefficients between carbon fluxes and LAI varied from 0.0.87 to 0.98 while LAI values were decreasing. In addition, strong linear correlations have also been specified between all cumulative carbon fluxes and LAI values in 2010-2011 growth season. Associations between cumulative carbon fluxes and biomass values have been investigated for the whole growth period. In this context, determination coefficients between cumulative NEE, GPP, Reco and biomass were obtained as 0.97, 0.99 and 0.97, for the 2009-2010 growth period, respectively. Also in 2010-2011 growth period, strong relationships (0.93 < r2< 0.97) between cumulative carbon fluxes and biomass values have been determined. The relationships between carbon exchange parameters and sPRI were evaluated by using all the data within the growing period. Consequently, good nonlinear relationships (0.86 &#8804; r2 &#8804; 0.87) have been obtained between these parameters. For the whole growth period, a good nonlinear relationship (r2= 0.87) has been obtained between cumulative GPP and sPRI. Consequently, the results of this study showed that carbon exchange parameters of winter wheat plant are affected by many meteorological and environmental factors. Clarification of the effects of these factors by making related studies would enable the researchers to define, explain and model the carbon exchange of winter wheat. Onthe other hand, evaluation of long term eddy covariance measurements is necessary for the reliability of the model results. After that, carbon budget of winter wheat can be estimated for wide areas. In addition to these, it is a clear need to measure and observe fluxes, meteorological factors, vegetation dynamics such as NDVI, sPRI, LAI, biomass for different plant varieties. Modeling by associating the carbon flux measurements with meteorological factors and vegetation dynamics gives the opportunity to use the results for wider areas.