Geology topics

Stephen P. Boyte

Publications and source records attributed to Stephen P. Boyte.

21 records · Page 2Linked to original sources

USGS2013

The integration of geophysical and enhanced Moderate Resolution Imaging Spectroradiometer Normalized Difference Vegetation Index data into a rule-based, piecewise regression-tree model to estimate cheatgrass beginning of spring growth

Cheatgrass exhibits spatial and temporal phenological variability across the Great Basin as described by ecological models formed using remote sensing and other spatial data-sets. We developed a rule-based, piecewise regression-tree model trained on 99 points that used three data-sets – latitude, elevation, and start of season time based on remote sensing input data – to estimate cheatgrass beginning of spring growth (BOSG) in the northern Great Basin. The model was then applied to map the location and timing of cheatgrass spring growth for the entire area. The model was strong ( R 2 = 0.85) and predicted an average cheatgrass BOSG across the study area of 29 March–4 April. Of early cheatgrass BOSG areas, 65% occurred at elevations below 1452 m. The highest proportion of cheatgrass BOSG occurred between mid-April and late May. Predicted cheatgrass BOSG in this study matched well with previous Great Basin cheatgrass green-up studies.

Great Basin↗

USGS2013

Influence of management and precipitation on carbon fluxes in greatplains grasslands

Suitable management and sufficient precipitation on grasslands can provide carbon sinks. The net carbon accumulation of a site from the atmosphere, modeled as the Net Ecosystem Productivity (NEP), is a useful means to gauge carbon balance. Previous research has developed methods to integrate flux tower data with satellite biophysical datasets to estimate NEP across large regions. A related method uses the Ecosystem Performance Anomaly (EPA) as a satellite-derived indicator of disturbance intensity (e.g., livestock stocking rate, fire, and insect damage). To better understand the interactions among management, climate, and carbon dynamics, we evaluated the relationship between EPA and NEP data at the 250 m scale for grasslands in the Central Great Plains, USA (ranging from semi-arid to mesic). We also used weekly estimates of NEP to evaluate the phenology of carbon dynamics, classified by EPA (i.e., by level of disturbance impact). Results show that the cumulative carbon balance over these grasslands from 2000 to 2008 was a weak net sink of 13.7 g C m −2 yr −1 . Overall, NEP increased with precipitation (R 2 = 0.39, P < 0.05) from west to east. Disturbance influenced NEP phenology; however, climate and biophysical conditions were usually more important. The NEP response to disturbance varies by ecoregion, and more generally by grassland type, where the shortgrass prairie NEP is most sensitive to disturbance, the mixed-grass prairie displays a moderate response, and tallgrass prairie is the least impacted by disturbance (as measured by EPA). Sustainable management practices in the tallgrass and mixed-grass prairie may potentially induce a period of average net carbon sink until a new equilibrium soil organic carbon is achieved. In the shortgrass prairie, management should be considered sustainable if carbon stocks are simply maintained. The consideration of site carbon balance adds to the already difficult task of managing grasslands appropriately to site conditions. Results clarify the seasonal and interannual dynamics of NEP, specifically the influence of disturbance and moisture availability.

Ecological Indicators↗

USGS2012

Ecosystem performance monitoring of rangelands by integrating modeling and remote sensing

Monitoring rangeland ecosystem dynamics, production, and performance is valuable for researchers and land managers. However, ecosystem monitoring studies can be difficult to interpret and apply appropriately if management decisions and disturbances are inseparable from the ecosystem's climate signal. This study separates seasonal weather influences from influences caused by disturbances and management decisions, making interannual time-series analysis more consistent and interpretable. We compared the actual ecosystem performance (AEP) of five rangeland vegetation types in the Owyhee Uplands for 9 yr to their expected ecosystem performance (EEP). Integrated growing season Normalized Difference Vegetation Index data for each of the nine growing seasons served as a proxy for annual AEP. Regression-tree models used long-term site potential, seasonal weather, and land cover data sets to generate annual EEP, an estimate of ecosystem performance incorporating annual weather variations. The difference between AEP and EEP provided a performance measure for each pixel in the study area. Ecosystem performance anomalies occurred when the ecosystem performed significantly better or worse than the model predicted. About 14% of the Owyhee Uplands showed a trend of significant underperformance or overperformance ( P <0.10). Land managers can use results from weather-based rangeland ecosystem performance models to help support adaptive management strategies.

Oregon;Idaho↗