Developing a Grassland Biomass Monitoring Tool Using a Time Series of Dual Polarimetric SAR and Optical Data
dc.contributor.advisor | Guo, Xulin | en_US |
dc.contributor.committeeMember | Kamal, Mohammad | en_US |
dc.contributor.committeeMember | Patrick, Bob | en_US |
dc.contributor.committeeMember | Sofko, George | en_US |
dc.creator | Finnigan, Carmen | en_US |
dc.date.accessioned | 2013-10-05T12:00:25Z | |
dc.date.available | 2013-10-05T12:00:25Z | |
dc.date.created | 2013-06 | en_US |
dc.date.issued | 2013-10-04 | en_US |
dc.date.submitted | June 2013 | en_US |
dc.description.abstract | Grasslands are the most important ecosystem to humanity, as they are responsible for feeding that majority of the human population. These are also very large ecosystems; they cover approximately 40% of the surface of the earth (Loveland et al., 1998), making ground-based surveys for monitoring grassland health and productivity extremely time consuming. Remote sensing has the advantage of providing reliable and repeatable observations over large swaths of land; however, optical sensors exploiting the visible and near infrared regions of electromagnetic (EM) spectrum will be unable to collect information from the ground if clouds are present (Wang et al., 2009). Imaging radar sensors, the most common being synthetic aperture radar (SAR), have the advantage of being able to image the ground even during cloudy conditions. The longer wavelengths of EM energy used by the SAR sensor are able to penetrate clouds while shorter wavelength used by optical sensors are scattered. A grassland monitoring tool based on SAR imagery would have many advantages over an optical imagery system, especially when SAR data becomes widely available. To demonstrate the feasibility of grassland monitoring using SAR, this study experimented with a set of dual-polarimetric SAR imagery to extract several grassland biophysical parameters such as soil moisture, canopy moisture, and green grass biomass over the mixed grassland in southwestern Saskatchewan. Soil moisture was derived from these images using the simple Delta Index (Thoma et al., 2006) first developed for a sparsely vegetated landscape. The Delta Index was found to explain 80% of the variation in soil moisture, in this vegetated landscape. Canopy moisture was modeled using the water cloud model (Attema and Ulaby, 1978). This model has a similar explanatory power of R2 = 0.80. This study found that only the photosynthesizing green grass biomass had a significant relationship with the canopy moisture model. However, only about 40% of the variation in green grass biomass can be explained by canopy moisture alone. The cross-polarized ratio developed from the dual polarimetric images was found to reflect the plant form diversity of the grassland. Biophysical parameters extracted from optical satellite imagery, Landsat-5 in the case of this study, were compared to those derived from the SAR images. This comparison revealed that the SAR images were superior in sensitivity to soil and canopy moisture. Optical imagery was found to be more sensitive to green canopy cover. An approach combining the results from both sensors showed an improvement in green grass biomass estimation (Adjusted R2 = 0.71). | en_US |
dc.identifier.uri | http://hdl.handle.net/10388/ETD-2013-06-1193 | en_US |
dc.language.iso | eng | en_US |
dc.subject | SAR | en_US |
dc.subject | Grasslands | en_US |
dc.subject | soil moisture | en_US |
dc.subject | canopy moisture | en_US |
dc.subject | biomass | en_US |
dc.title | Developing a Grassland Biomass Monitoring Tool Using a Time Series of Dual Polarimetric SAR and Optical Data | en_US |
dc.type.genre | Thesis | en_US |
dc.type.material | text | en_US |
thesis.degree.department | Geography and Planning | en_US |
thesis.degree.discipline | Geography | en_US |
thesis.degree.grantor | University of Saskatchewan | en_US |
thesis.degree.level | Masters | en_US |
thesis.degree.name | Master of Science (M.Sc.) | en_US |