| Ecosystem characteristics are continuously changing at a variety of spatial and temporal scales. Since ecosystems provide us essential services that are associated with ecosystem characteristics, it is critical to understand the spatial and temporal variations of ecosystem characteristics at different scales as well as their corresponding consequences and underlying drivers. In this dissertation, I present three studies that investigate the temporal variations of NEE (net ecosystem exchange of CO2) in a terrestrial ecosystem (i.e., Oak Openings in Ohio), of the temporal changes of water surface area (WSA) in a wetland ecosystem's (i.e., Prairie Pothole Region), and the spatial-temporal change of NEE and in an aquatic ecosystem (i.e., Western Lake Erie). I found that NEE in Oak Openings mostly varied at daily, seasonal, and annual time scales. At the daily scale, PAR is a more important meteorological driver than Ta (air temperature), while at the seasonal and annual scale, Ta is a more important meteorological driver than PAR. Additionally, vegetation phenology is another major driver of NEE variation at the seasonal and annual scales. The study in the Prairie Pothole Region (PPR) suggests that WSA is a linear function of the Palmer Drought Severity Index (PDSI). The PDSI of the year before last and the current year explain 79% of the annual variance of WSA. Also, past drought conditions were more important than current drought conditions in determining the current WSA of the potholes. Based on this model and future climate projection, WSA in PPR is predicted to lose over half of its current value by 2050s and retain less than one-third of itself by the 2080s. In western Lake Erie, neither strong diurnal course nor strong seasonal/annual course of NEE were found. However, monthly mean NEE and monthly mean chlorophyll-a were significantly correlated with 1--2 month lags, depending on the location. Chlorophyll-a was also found to affect the relationship between meteorological variables and NEE. During algal bloom seasons, daily PAR, Ta, wind speed, and VPD were significantly correlated with daily NEE; while during non-algal bloom seasons, no relationship was observed between them. Overall, studies in this dissertation suggest that NEE varied uniquely among different time periods, different frequencies, different ecosystems and locations, and WSA in PPR varied intra-annually and inter-annually. These variations were driven by a few key climatic and biological factors with long lasting or lagged effects. |
