The intercept is an important measure of water vapor at both the field and scene levels

The slope acts as a measure of moisture advection as a factor of wind at the field-level.At the scale of an individual field, the intercept quantifies the build-up of moisture over a field, while at the scale of the entire study site, the spatial pattern of intercepts highlight advection of moisture across the scene. The trajectory is equivalent to the azimuth of the water vapor trend at the field-level. To assess the strength of the modeled, fitted surface, r-squared and p-values were also calculated. Only fields that had statistically significant linear trends were analyzed. The water vapor occurring above an example field and its corresponding fitted plane are shown in Figure 4.4. Water vapor concentrations were explored as their distributions vary by day and within the scene. Within each scene 1,000 random pixels were selected and a Pearson’s R was calculated to analyze the correlation between GV fraction and LST with water vapor concentrations in order to test Hypothesis A. GV fraction was obtained from MESMA and LST from the corresponding MASTER imagery. If expected correlations are found,planting blueberries in a pot these correlations would be indicative of water vapor relating to the surface beneath it.

Green vegetation transpires and produces water vapor, which will lead pixels with more vegetation to have higher water vapor. These surfaces should also have lower temperatures as evapotranspiring plants shed energy through latent heat. We tested Hypothesis B by examining patterns of water vapor intercepts against prevailing wind direction. Over the study area, we expected the water vapor concentration, as quantified through the intercept of the fitted water vapor plane, to increase downwind due to moisture advection. For example, if the wind is blowing from the North, we would expect fields in the southern part of the study area to show higher intercepts than fields in the northern part of the study area. We evaluated this hypothesis in each of the three years by mapping out intercepts in the study area and qualitatively assessing their relationship to the calculated wind direction. At the field level, we analyzed gradients of water vapor as they vary over agricultural fields in line with expectations of vapor as conceptualized in Figure 4.1 and as explained through Hypotheses C through K in Section 1. As such, we tested Hypotheses C, D, E, and F by evaluating the relationship between wind speed and direction with the slope of water vapor. Even if pixel or scene-level trends were not identified in an image, we included all dates of imagery in the field-level analysis as we hypothesize that trends may be happening at variable scales so null results at one level does not preclude significant results at another. The trends of water vapor above fields will be a factor of both wind speed and direction.

We expected to find that, within fields predominately covered in green vegetation , the relationship between water vapor slope and wind would show a quadratic relationship with relatively high or low winds creating water vapor gradients less steep than winds that are of an “intermediate” magnitude. Higher winds will move water vapor at a faster rate, which will lead to shallower gradients. However, this concept should only hold once the winds reach a certain threshold magnitude and a stable directionality as light and/or inconsistent winds will not produce any gradients. To test this hypothesis we plotted wind magnitude against water vapor slope in each of the three years. We also expected to find water vapor surfaces that aligned in directionality with the wind. We calculated the difference between the estimated wind direction and the trajectory of the water vapor above each field as the directional difference. For those fields that had directional differences of less than 30° and a statistically significant slope of vapor, we analyzed their characteristics such as crop type and GV fraction to understand what types of fields our set of hypotheses holds for. Second, we tested the impact of field size on water vapor slope in fields of >50% GV to examine Hypothesis G. We plotted field size against water vapor gradient while hypothesizing that we would find a positive relationship. Steeper gradients would be expected above large fields as they have a larger surface area over which the vapor can advect. Third, we observed the relationship between GV fraction and water vapor slope in order to test hypotheses H and I.

We separated fields into groups of similar field size to control for the impact of this factor and then studied the correlation between green vegetation cover and water vapor slope and intercept within each of those groups. We hypothesized that fields with lower vegetation cover would show a poor relationship between GV fraction and water vapor slope and/or intercept while fields containing a majority GV fraction would have a positive correlation with water vapor slope and/or intercept. We used a 50% GV threshold as was set in Shivers et al. . Field-level correlations between GV and intercept would be expected in situations with low winds and higher build-up of water vapor whereas strong correlations between GV and slope would be expected if consistent, moderate winds created advection of moisture across fields. Positive correlations would indicate that fields with more transpiring vegetation are adding more moisture to the air than less vegetated fields. A higher concentration of water vapor would be confirmed though a positive correlation with water vapor slope if winds are consistent and moderate, or an increase in intercept if winds are faint and/or variable. Fourth, this study evaluated Hypotheses J and K by evaluating the slopes and intercepts of the fitted water vapor surfaces over fields of different irrigated crop species. These intercepts indicate the magnitude of water vapor above a field while the slope is indicative of the trend of vapor over a field. A one-way ANOVA was performed to assess differences in slopes between the crop species, and results were evaluated with expected ET rates. ET rates were approximated using the expected crop ET coefficient for irrigated crops for June in the Southern San Joaquin Valley of California in a dry year . We expect crops that transpire more to have significantly higher slopes than crops with lower ET rates. To further examine expected patterns of water vapor as it relates to ET while controlling for some level of complexity within the scene,raspberries in pots we chose three crops that are prevalent in the study area and looked at their LST as it related to water vapor slope. We explored water vapor over fields of alfalfa, almonds, and cherries. We included all fields which had a fractional green cover of 50% or more. We aimed to investigate the hypothesis that fields with lower temperatures would have steeper water slopes. Fields with lower LST are assumed to be healthier and less stressed than those with higher LST because plants that have adequate water will transpire and cool themselves . The three dates of imagery showed different spatial trends of water vapor. In 2013, water vapor showed a clear increasing trend from southwest to northeast, which is noticeable but not as defined in 2015 . The 2014 and 2015 scenes showed decreasing water vapor values in the northernmost portion of the scene as the Central Valley transitions into the mountains and the elevation increases. Besides the decreasing water vapor in the northernmost part of the scene, the remainder of the 2014 image is not indicative of any other trends. When observing the imagery at a larger scale, the water vapor from 2013 and 2015 shows strong coupling with the ground surface below with agricultural field boundaries clearly defined. This result may be indicative of surface-atmosphere interactions or simply an artifact of the reflectance retrieval. In contrast, the 2014 imagery shows patterns of vapor that are more resonant of vapor or clouds that do not relate directly to the surface structure below it. We hypothesize that the difference may be attributable to the moisture level of the atmosphere, the differences in the timing of image acquisition, or the height of the water vapor in the scene. The 2014 imagery had both the driest atmosphere at 10.6 mm and also was the image that was acquired latest in the day.

Given the appearance of the water vapor imagery, we hypothesize that the water vapor in 2014 was located well above the terrain while the water vapor in the 2013 and 2015 images were lower in the atmosphere, closer to the terrain. If our study site had larger elevation gradients, we could test this hypothesis with the method laid out in Roberts et al. . However, the flatness of our study area precludes such an analysis. Computation of water vapor intercepts and interpolation of wind directionality allowed for comparison between water vapor abundance and patterns of wind as laid out in Hypothesis B. Figure 4.7 shows the directionality of the wind and the water vapor intercept maps side-by-side for comparison. Of the three dates, the 2013 imagery shows the most clear pattern of advected moisture that generally agrees with the wind map, especially in the northern portion of the study area. The intercept map shows water vapor concentration increasing from south to north while the wind direction map shows a south to north trend of wind in the northern part of the study area. As crops transpire and water vapor advects, theintercepts above fields show increasing moisture. The southern portion of the study area shows less agreement with winds, indicating winds coming from the northeast but a water vapor gradient increasing from west to east. We hypothesize that this may be due to differences in temporal scales or wind interpolation error, as noted in the discussion. The 2014 and 2015 images show water vapor that are not as clear in their trends. The 2014 wind map shows winds primarily from the north and west. The northern winds do generally agree with water vapor intercepts that seem to increase from north to south. The 2015 water vapor intercepts show patterns that are somewhat similar to 2013 with a general south to north increase in moisture, except for in the most northern portion of the flight line. Variability in winds makes evaluation between intercepts and trends challenging. Moreover, while the wind map is a snapshot at the time of flight, the intercept map likely represents a trend of water vapor over a time period of many hours, which further complicates analyses. However, results show some approximate agreement between winds and advected moisture, especially in 2013. Hypotheses C-F proposed expected relationships between the directionality of water vapor and its slope with both wind magnitude and wind direction. When looking at fields that were predominately covered in green vegetation , we found patterns that were somewhat consistent with our hypotheses that a moderate wind speed would show higher slopes than very low wind speeds or high wind speeds. Although r-squared values were low, each year showed a significant quadratic relationship between water vapor slope and wind magnitude . The 2014 image also had a significant linear trend, but the quadratic relationship showed a higher r-squared. Because wind speeds were lower in 2014, on average, than the other two years, we hypothesize that 2014 would have shown a more definitive quadratic trend if the 2014 scene had more higher wind speed values. These quadratic trends, although accompanied by considerable spread, are in line with our hypotheses.Analyzing these directionally aligned fields by GV cover and crop type in each year, we found no significant characteristics related to GV when these sub-selected fields were compared to all fields in the study. Examining histograms of GV fraction within the fields that showed directional agreement, no discernable pattern was found. High GV fields were as likely to align in trajectory with wind direction as the low GV fields. In fact, the mean GV of the selected fields were 0.45, 0.46 and 0.43 for the three years, in comparison to 0.47 for the average of all fields in the study. However, segmentation by crop type did show some interesting results. Looking at nine of the most prevalent crops, large differences are seen in the percentage of these crops that showed directional agreement with the wind .