A coarse resolution WRF simulation is run for the entire period to be downscaled, while for only a subset of that period a nested version of the same model is run at high resolution. The period over which the coarse and high-resolution runs overlap is called the training period, while the remaining portion is termed downscaling period. For each time of the latter, the best matching coarse estimates over the training period are found. The downscaled solution is then constructed from the set of high-resolution values that correspond to the best matching coarse analogs. This method is based upon Delle Monache et al.. The WRF simulation used telescoping, one-way interacting computational grids. Their respective horizontal grid increments are 20 km and 4 km, with the 4 km grid centered over California. The initial and lateral boundary conditions are specified using MERRA-2. The 20 km grid was run for the entire 01 Jan 1980–31 Dec 2015 period, and generated output every hourly, while the nested 4 km grid was run only during the last year of the full simulation . The high resolution downscaled dataset is constructed for the entire 36-year period using the 4 km resolution training data and the 20 km simulation . The result is an hourly time series at each 4 km grid point for January 1st 1980 to December 31st 2015. Wind speed and direction at hub heights, indoor plant table including 50 m, 80 m, 140 m, are output. DNV GL served solely as a data provider, and is not responsible for any results from this data.
The Modern-Era Retrospective analysis for Research and Applications, Version 2 is a reanalysis product for the satellite era using the Goddard Earth Observing System Data Assimilation System Version 5 produced by Global Modeling and Assimilation Office at NASA. MERRA-2 integrates several improvements over the first version MERRA product. For the fields used in this study, the spatial resolution is ~55 km with 3-hourly output frequency from 1980 to present. Vertical interpolation of MERRA-2 data was performed to calculate hub height wind speed. Variables used in vertical interpolation were extracted from two subsets: 3-hourly instantaneous pressure level assimilation and hourly instantaneous single level assimilation .The Integrated Surface Database from NOAA’s National Centers for Environmental Information were used for assessment of hourly 10 m wind speed from model and reanalysis. The ISD observational stations are distributed globally, with the highest concentration of stations found in NorthAmerica. Stations across California that provide full year data were selected. As not all stations had continuous temporal coverage between 1980 to 2000, each year was calculated separately so as to maximize the number of available stations. To compare 10 m wind speeds from model and reanalysis datasets to ISD, the nearest grid point values to each of the ISD stations was used. Coastal stations were neglected in the analysis of 10 m winds, due to coastal biases that tend to occur in near-surface coarse resolution reanalysis. These biases tend to emerge because similarity theory is typically employed to extract 10 m wind speeds, which produces distinctly different results over the ocean and land surface.
The wind speed at each wind farm location was determined using nearest grid point values to each wind farm site. To obtain hub-height wind vectors, vertical interpolation was performed on 3-hourly VR-CESM, 3-hourly MERRA-2, 6-hourly CFSR, and 3-hourly NARR products from 1980 to 2000. As mentioned above, hub-height wind output is available directly from the DNV GL Virtual Met data product. Vertical interpolation of VR-CESM data uses the 3D wind field on hybrid surfaces and 10 m altitude wind speed, which is computed from similarity theory. For VR-CESM data, the interpolation procedure is as follows: the CAM5 hybrid coordinates are first converted to pressure coordinates within the column being analyzed, the height of each pressure surface above ground level is computed by subtracting the surface geopotential height from the geopotential height of the model level, two model levels that bound the desired interpolation altitude are selected or, if the interpolation altitude is below the lowest model level, the lowest model level and 10 m wind speed field are used, and logarithmic interpolation is applied to obtain the wind speed at the desired interpolation altitude. The interpolation was done by fitting a log equation with the two levels bounding the altitude to be calculated, then with the log profile, interpolating the wind at desired altitude. Vertically interpolated wind speed from MERRA-2, CFSR, NARR, and sounding observations all followed a similar procedure, and were calculated at three hub heights . Figs. 1–4 show the interpolated hub-height wind speed at 50 m and 140 m, respectively, at northern and southern California.
For wind speed at 80 m, and further wind speed analysis, please refer to the cosubmitted research article. Wind turbines can contribute to energy via the electric power system. This contribution is the total amount of usable energy supplied by the turbine per year . The capacity factor is often defined as actual power output divided by the max amount of wind power that can be generated through the system. This wind speed and CF relationship is not continuous since there is a discontinuous minimum and maximum wind speed required to begin and cease wind power production , and this is represented with different power curves associated with each of the wind farm sites. The calculated CF at each wind farm site is based on different characteristic power curves at that site, and do not include electrical losses during the power generation process. The normalized power curves at each wind farm sites, with each value corresponding to a 1 m/s wind speed bin increment starting from 0 m/s, are listed in Table 1. To calculate the CF, wind speed is multiplied with the corresponding power curve value from the corresponding wind speed bin, and then times 100 to convert the percentage values.Following these Renovation policies, known as Doi Moi, Vietnam focused on improving infrastructure and shifting the physical environment in the delta. The government focused on direct domestic and foreign investment to support a “big push” of government investment, a common approach in developing countries joining the global economy . In fact, cold war politics were played out through communist or democratic control over the “sinew of development”: infrastructural projects . The irrigation network in Vietnam was an example of a postcolonial and post-conflict technique for economic and social development. The global success story in the late 1980s for infrastructural progress, the taming of nature, was the Tennessee Valley. In fact, American engineer David Lilienthal designed the Mekong Delta Development Program, which systematized the control of water through canals and levees . This plan was implemented in the 1990s, when the majority of the canal system was built. The intent was to create a highly productive, intensified, triple-rice cropping delta environment, which the farmers of the delta enacted by quadrupling production of rice per hectare .Now the pendulum has swung and heads of state are reevaluating the economic sustainability of intensified agriculture. In the international arena, the debate between intensification and extensification is raging. On the one hand are advocates of “ecoagriculture,”also known as the “land sharing” camp, which encourages wildlife-friendly and “pro-poor” production systems to limit the extent of agricultural impacts on ecosystem health . On the other hand are the “land sparing” conservationists who find that ecosystem integrity is far superior without human systems influence, plant growing stand and thus advocate for intensification of existing agricultural lands . In Vietnam, this debate is taking form in the policy arena, as development projects and the Ministry of Agriculture and Rural Development attempt to peel back the triple-cropping system that relies on irrigation water to push out a third crop during the dry season. For example, a recent IUCN project in the Mekong River Delta focuses on seasonally removing dykes to allow natural flooding to occur, which forces a shift to a completely different production regime.
As a function of the agrarian changes described above, Vietnam is now experiencing a rural-to-urban migration trend. On top of being a regional rice production giant, Vietnam is now a powerful electronics and textile export economy. Families who cannot make ends meet farming, or wish for a non-farming future for their children through education, use migration as a coping strategy. These environmental and economic changes have had huge impacts on Vietnamese families, shifting how mothers and fathers interact with their children and each other, where family members live, and how farming and family is “done” . Men, women, children, and families are shifting their personal and familial identities in response to environmental, economic, and social pressures.The story told in this dissertation takes place in Southern Vietnam, in the Mekong River Delta. It is a landscape blanketed with thick productive, green rice fields, and peppered with vegetable gardens and small villages. “As husband and wife” is a phrase repeatedly invoked to encapsulate a sense of togetherness in Vietnamese families and on Vietnamese farms. Shared feelings of endurance and dedication, shared work and child care, shared responsibility. Spouses attend funerals and celebrations as “husband and wife.” They often raise their children “as husband and wife.” And, importantly to this story, they make financial and migration decisions “as husband and wife.” However, there is a tension in this sense of family cohesion, which is equally as important to this story, because farm fields are often not planted,tended, and harvested “as husband and wife.” In many cases, farmscapes are strictly divided into spaces tended by husband or wife. And men and women do not gain access to and control of resources “as husband and wife,” necessarily. It is this tension, this intersection of rural agricultural landscapes and gender dynamics, where I do my research. The research approaches agrarian change from three scales: regional production practices, household livelihoods, and individual identities. At a regional scale, Chapter 2 uses remotely sensed radar data, combined with in-situ moisture readings, to determine water-saving practice adoption through change detection of a time series wetness index. The research illustrates a water-saving practice adoption likelihood scale across the delta, indicating promise for the change detection methodology. My contribution in Chapter 2 is to explore a beta transformation and change detection approach, referred to as a “multi-temporal SAR for SWC change detection” that uses multiple passes from the same incidence angle and polarization. By exploring and refining this methodological approach, the research builds a toolbox for harnessing remote sensed data in agricultural research. At a household scale, Chapter 3 is a gender disaggregated plot-level study that uses a binary logistic regression to determine if livelihood approaches on male- and female-managed plots influence adoption of farming practices. We find that gendered plot management is directly associated with SI and CI practice adoption, and there is an indirect gendered impact due to unequal access between the sexes to natural and human capitals that are associated with increased SI adoption. Chapter 3 contributes empirical evidence towards understanding the tension between CI and SI practice adoption in the Vietnamese Mekong River Delta. To conclude, we highlight specific opportunities for improvements in female uptake of SI practices through our suggested policy reform. Vietnam plans to scale back its annual triple rice crop production in the Mekong River Delta over the next several years, making these gendered differences in sustainable practice adoption crucial to a smooth transition into a sustainable food system in the context of global change. Finally, at an individual scale, Chapter 4 uses a spatial intersectionality approach, a grounded theory, exploring identities and spaces traversed by migrant families in the city and country. There are three gaps in the gender and migration literature in Vietnam. First, these studies have almost exclusively taken place in Northern Provinces of Vietnam, which I attempt to fill by conducting this research in Ho Chi Minh City. Second, they predominantly focus on women’s identity as the migrant, which I address by analyzing men and women’s experiences in the context of the mobile family. Thirdly, most recent studies focus on abstract identities such as “husband” or “father.” I use space as the primary focus in which identities compete with each other, within a body, and within a family. Space and identity are coimplicated; so choosing to work and live in separate spaces has profound impacts on gender identity, gender relationships, and performative or lived family values. This research disrupts assumptions of gender roles by taking a spatial look at intersecting identities.