Two-thirds or more of the pistachio and almond fields showed water vapor trends that were in agreement with the wind direction while less than a third of walnut and grape fields did. These differences may be attributable to differences in the crops themselves such as rate of ET or health of the plants, or these differences could be attributable to their position within the study scene as the crops are not evenly disbursed throughout the area. Hypothesis G predicted that larger fields would show steeper slopes of vapor than smaller fields because moisture has a larger area over which to build up. We looked at correlations between field size and water vapor slope in each of the three years and found temporally stable results that contradicted our hypothesis. In 2013, 2014, and 2015 water vapor slope and field size showed inverse correlations of -0.35, -0.34, and -0.25 respectively. Smaller fields show steeper slopes than larger fields, generally.Water vapor imagery holds information regarding the land surface, the atmosphere, and their interactions, and thus offers a valuable dataset with which to observe and quantify key fluxes between the two. In this study, we expanded upon the work of Ogunjemiyo et al. by testing several hypotheses to investigate how AVIRIS estimates of water vapor vary with the surface properties and atmospheric conditions in the Central Valley of California. In this section,blueberries in containers growing the key findings will be interpreted, the challenges of this methodology will be discussed, and thoughts on how to move this work forward will be presented.The main findings of the study are summarized in Table 4.4.
The results supported that water vapor imagery will show coupling with the land surface at the pixel-level and advection at the scene-scale, under certain atmospheric conditions . Further, results supported hypotheses that water vapor gradients will form over fields as a function of wind speed and direction . However, hypotheses that field-level ET would be detectable through water vapor slopes were not supported . At the pixel level, we found results that supported Hypothesis A in two of the three scenes studied. The 2013 and 2015 dates of imagery showed promising correlations to the ground surface below with water vapor increasing with increasing fractions of green, healthy vegetation, while the 2014 imagery did not show any coupling. One hypothesis for this difference is the timing of the flights. While the 2013 and 2015 images were acquired between 11 AM and noon, the 2014 imagery was flown near 3 PM. This difference in timing suggests that the boundary layer may stay closer to the surface earlier in the day, allowing for greater study of the surface below. Another hypothesis is differences in the atmospheres between those dates such as the air temperature or humidity, which could affect the water vapor over the scene. These differences may also explain why 2014 had significant field-level trends with respect to wind speed and direction while pixel-level correlations with GV were not found. If the 2014 scene shows water vapor that is higher in the air column and more coupled to the wind than to the surface, we would expect to find field-level patterns that act as a function of wind while we would not expect to find surface and water vapor coupling. Our results support this hypothesis. By calculating field-level water vapor intercepts and evaluating them over the scene, we found evidence that supported Hypothesis B, particularly in the 2013 scene.
In that image, small-scale trends of vapor across the study scene showed patterns that were consistent with advection of moisture, as shown in Fig. 4.7. This finding suggests that, as air moves across the Central Valley, crop ET adds vapor to the water column, which builds up in the downwind direction. These results, therefore, show that water vapor imagery could be of use for regional water resource accounting in agricultural areas where water vapor imagery could help quantify latent heat fluxes. Hypotheses C, D, and E were supported by significant quadratic relationships between wind magnitude and slope that suggest that water vapor slopes only occur when the wind is strong enough to create such trends but weak enough that the water vapor slope is not too shallow. Work by Ogunjemiyo et al. found their conceptual model held best when winds were at 1.17 to 1.24 m/s. The one case that showed no water vapor patterns was with an August image with 3.91 m/s winds. The winds in our study were between 1.5 and 3 m/s, in between Ogunjemiyo’s values of 1.24 to 3.91 m/s. The quadratic result of our wind speed vs. slope curves suggest that relationships might hold best at intermediate speeds around 2.2-2.5 m/s, and that the wind speed in Ogunjemiyo were lighter than their optimal speed for creating water vapor slopes. Additionally, some fields showed water vapor trends in line with our hypotheses regarding water vapor trajectory and wind direction, supporting Hypothesis F. As shown in Fig 3.7 of the previous chapter, within pure GV pixels there exists a large range of temperatures that suggest that not all green crop fields are transpiring, possibly due to water shortages mid-summer during a severe drought. Therefore, we hypothesize that fields that show directional agreement between water vapor patterns and wind could be indicative of those fields that were actively transpiring at the time of flight. In agreement with this hypothesis, within those fields that were directionally aligned, we found large variability in the prevalence of crop types. As seen in Table 4.2, two-thirds or more of the alfalfa, almond and pistachios align in trajectory with the wind. Interestingly, these crops are also known for their high water use. Shivers et al. shows that, in comparison to other deciduous fruits, subtropical fruits or grapes, alfalfa, almonds and pistachios are higher water users.
Plotting the percentages in Table 4.2 with the water application values of each crop from Table 2.10 , we find a positive correlation . This result suggests that high water-using crops show water vapor patterns that align with the wind more frequently, a finding that may be indicative of ET. While pixel-level correlations, scene-level advection, and relationships between wind and water vapor were supported by results, we found that Hypotheses G-K were not supported. These hypotheses all emanated from a larger idea that water vapor slope would correlate with ET. We did not find evidence of that. First, we found a temporally-consistent negative relationship between field size and water vapor slope which was in opposition to Hypothesis G. One explanation is that the signal of water vapor movement actually becomes more diffuse and less concentrated over larger fields due to inconsistencies of wind that tend to move the water vapor around and even it out instead of creating gradients. Patterns in smaller fields would be more closely tied to shorter-term wind patterns which would create gradients. Another explanation,planting blueberries in containers suggested by that fact that some crops showed a positive correlation between LST and slope, is that rather than advection of evapotranspired moisture downwind over individual fields, ET causes an accumulation of water vapor over the field. This idea will be explored further in section 4.4.1.3. Second, we did not find positive correlations between GV fraction and water vapor slope as postulated in Hypotheses H, I and J. If green vegetation transpires and adds to the water vapor above the fields, we expected this addition of water vapor to be quantifiable through the slope above it. We found no correlation between the two, even when results were segmented by field size and GV fraction. We used 50% GV as the cutoff to demarcate sparsely vegetated fields from highly vegetated fields, as is consistent with previous studies . However, we found that the average fractional GV coverage of fields that showed good alignment between wind direction and water vapor directionality was around 45%. Therefore, future studies may want to consider a lower GV threshold or a segmentation of fields into multiple GV classes. Third, we did not find consistent positive linear relationships between expected crop transpiration rates and water vapor slope as hypothesized in Hypothesis K. Water vapor patterns were as expected at the field level, in response to wind. However, water vapor patterns were not as expected in response to the surface properties of field size, GV fraction, and ET rate. We had hypothesized that field-level water vapor slopes can be used to infer crop transpiration, but did not find evidence supporting that hypothesis. Rather, our results suggested that water vapor accumulation from transpiration was more dominant than the advection signal at the field level. The rate of ET has been found to stay constant with downwind distance across a field, even if warm, dry air is being advected to a vegetated field . If plants are transpiring at a constant rate and winds are not strong enough or stable enough in directionality to evenly disburse it, the concentration of water vapor above the field would increase relatively evenly throughout the field, leading to a diminished slope.
Crops are also more aerodynamically rough than an empty soil field , and the resultant turbulence caused by vegetation creates eddies and atmospheric mixing that may muddle signals of field-level advection discernable above smoother landscapes. The hypothesis of water vapor accumulation is supported by results that found a positive relationship between LST and slope for some crops, a negative relationship between field size and slope, and a weak positive correlation between water vapor intercept and GV fraction in 2013 and 2015. Therefore, the results of this study lead us to new conceptual understanding that the magnitude of water vapor as assessed though the intercept of a fitted plane may be a more accurate indicator of latent heat flux. However, underlying heterogeneity of the landscape and scaling issues, as discussed below, prohibited isolated analyses of intercepts in this study area. There is error within all water vapor estimates regardless of which retrieval method is used, and the estimates vary significantly from model to model . However, Ben-Dor et al. found that, of six different water vapor retrievals, ACORN estimated water content with acceptable accuracy and, importantly for our study, it was one of only two models that did not show significant changes in water vapor by vegetation fraction. Therefore, the positive correlations found in years 2013 and 2015 between water vapor and vegetation fraction is assumed to be a product of coupling between the landscape and the atmosphere, rather than an artifact of the retrieval. There is also error within wind estimates. Wind direction and magnitude can change significantly within a small period of time, making estimations of wind within the study scene at the time of the flight particularly difficult. For example, the Button willow meteorological station registered a wind direction of 353 degrees at 21:28 UTC and 51 degrees ten minutes later at 21:38 UTC on June 3, 2014 when imagery was being captured. With wind directionality that changes frequently, water vapor will not show a uniform trend in a single direction . Furthermore, with a difference of 100 degrees within ten minutes, estimating wind direction inherently contains a large degree of error. Moreover, with uncertainty as to the accuracy of the wind estimations at the meteorological stations, interpolating from this data to the entire scene will have even a greater degree of error. Unlike Ogunjemiyo et al. who studied water vapor over a relatively homogeneous area of transpiring poplars, this study evaluated water vapor as it varies across a very diverse agricultural landscape with many different crop species, green vegetation cover, and irrigation regimes. As such, Ogunjemiyo’s conceptual model illustrated an ideal relationship between water vapor and vegetation at the field-scale that may not hold when introduced in our complex study area. First, interactions between water vapor occurring over two diverse, adjacent fields may alter the vapor deficit and stomatal response of a single crop field and result in water vapor trends that do not hold with Ogunjemiyo’s model. The schematic in Figure 4.15A illustrates one possible interaction in which a transpiring field is upwind of a non-transpiring field.