Improving irrigation requires smart data gathering to help growers make better choices in the field. Measuring in situ creates high-resolution, temporal data enabling us to see clearly what’s happening over time—but only at a single point. Satellites show data across a large spatial scale but are hampered by revisit frequencies, clouds, and resolution limits.
Often we see information in a silo, looking at one type of data or another. The challenge to researchers is how to connect across these scales and combine the information to make better irrigation decisions. In this webinar, Dr. Colin Campbell explores the future of irrigation and research he’s been doing with collaborators at Brigham Young University. Learn:
How researchers are combining in situ, drone, and satellite measurements to extract key information
In the conclusion of our three-part series on improving NDVI sensor data (see part 2), we discuss how to correct for limitations which occur in high leaf area index (LAI) conditions.
Where there’s a large amount of vegetation, NDVI tends to saturate.
NDVI Limitations – High LAI
NDVI is useful in the midrange of LAI’s as long as you don’t have strong soil effects, but as you approach an LAI above 4, you lose sensitivity. In figure 6, loss of sensitivity is primarily due to a saturation in the red band. Measurements were taken in a wheat canopy and a maize canopy. The near-infrared reflectance is sensitive across the entire spectrum of the wheat and maize canopies, but the red saturates relatively quickly. Where the red starts to saturate is where the NDVI starts to saturate.
Figure 6: Gitelson (2004) J. Plant Phys
Note: NDVI saturates at high LAI’s, however, if your purpose is to get at the fractional interception of light, NDVI tends not to have the saturation issue. In Figure 7, Fpar or the fractional interception of light of photosynthetically radiation is nearly complete far before NDVI saturates. This is because canopies are efficient at intercepting light, and once we get to an LAI of about 4, most of the light has been intercepted or absorbed by the canopy. Thus, incremental increases in LAI don’t significantly affect the FPar variable.
Figure 7: Fractional interception of light is near complete at an LAI around 4. (Gamon et al. (1995) Eco. Apps)
Solution 3- WDRVI
One solution for the NDVI saturation issue is called the Wide Dynamic Range Vegetation Index (WDRVI). Its formulation is similar to NDVI, except for a weighting coefficient that can be used to reduce the disparity between the contribution of the near infrared and red reflectance.
In the WDRVI, a is multiplied by the near-infrared reflectance to reduce its value and bring it closer to the red reflectance value. In doing so, it balances out the red and the near-infrared contribution to the vegetation index.
Figure 8: (Gitelson (2004) J. Plant Phys)
a can range anywhere from 0 to 1. Figure 8 shows that as we use a smaller value of a, we get an increasing linear response of the wide dynamic vegetation index to LAI.
The only drawback of the WDRVI is that the selection of a is subjective. It’s something that you experiment on your own until you find a value for a that is optimal for your solution. People tend to err on the side of a very low value simply because they’ll get closer and closer to a linear response to LAI as a decreases.
Solution 4 – Enhanced Vegetation Index
The enhanced vegetation index (EVI) was designed to enhance sensitivity in high biomass ecosystems, but it also attempts to reduce atmospheric influences. This was a vegetation index created for the purposes of a satellite-based platform. There’s a lot of atmosphere to look through from a satellite to the ground, and sometimes the aerosols in the atmosphere affect the reflectances in the red and the near infrared regions causing spurious observations. The EVI also tries to reduce sensitivity of the index to soil. Thus the EVI is a kind of solution to both extremes.
In the EVI equation, the two major inputs are near infrared and red reflectances. C1 , C2, and L are all parameters that can be estimated, but the blue band is something that has to be measured. Most NDVI sensors are two band sensors, so you don’t have that information in the blue. Plus, with satellites, the blue band is relatively noisy and doesn’t always have the best quality data, thus EVI has limited value.
Solution 6: EVI2 (Enhanced Vegetation Index 2)
Those problems led a scientist named Jiang to come up with a solution. Jiang observed quite a bit of autocorrelation between the red band and the blue band, so he decided to try and formulate EVI without the blue band in what he called the EVI2 (Enhanced Vegetation Index 2). if you’re interested in the mathematics, we encourage you to read his paper, but here we give you the equation in case you’re interested in using it.
When Jiang calculated his EVI2 and compared it to the traditional EVI (Figure 9), it was nearly a one to one relationship. For all intents and purposes EVI2 was equivalent to EVI. Since this avoids blue band, it offers some exciting possibilities as it reduces to just using the two inputs of NIR and red bands to calculate NDVI.
NDVI Sensor Summary
NDVI measurements have considerable value, and though there are extremes where NDVI performs poorly, even in these cases there are several solutions. These solutions all use the near infrared and the red bands, so you can take an NDVI sensor, obtain the raw values of NIR and red reflectances and reformulate them in one of these indices (there are several other indices available that we haven’t covered). So if you’re in a system with extremely high or low LAI, try to determine how near infrared and red bands can be used in some type of vegetation index to allow you to research your specific application.
Last week we discussed Normalized Difference Vegetation Index (NDVI) sampling across a range of scales both in space and in time, from satellites sampling the entire earth’s surface to handheld small sensors that measure individual plants or even leaves (see part 1). This week, learn about NDVI applications, limitations, and how to correct for those limitations.
Limitations of the Normalized Difference Vegetation Index tend to occur at the extremes of the spectrum.
People use NDVI to infer things like leaf area index (LAI) or fractional light interception (FPAR) of a canopy. Some scientists also associate NDVI with biomass or yield of a crop. People also use NDVI to get a sense of phenology (general temporal patterns of greenness), as well as where vegetation occurs or how much vegetation is in a particular location.
In Figure 4, you can see how the reflectance spectrum at a given canopy LAI changes with leaf area index, decreasing in the visible range while increasing in the near infrared.
At very low LAI’s, the reflectance spectrum is relatively undifferentiated between red and NIR (black line), but when LAI is high, there’s a strong absorption of red light by chlorophyll with a strong reflectance in the NIR. In fact, as LAI increases, there’s an ever-increasing reflectance in the near infrared around 800 nm.
Limitations of the Normalized Difference Vegetation Index tend to occur at the extremes of the spectrum. Any time there’s very low vegetation cover (majority of the scene is soil), NDVI will be sensitive to that soil. This can confound measurements. On the other extreme, where there’s a large amount of vegetation, NDVI tends to saturate. Notice the negligible difference between spectra at a leaf area index (LAI) of 3 (purple) versus 6 (green). Indeed, in a tropical forest, NDVI will not be sensitive to small changes in the LAI because LAI is already very high. However, several solutions exist.
Solution 1-Soil Adjusted Vegetation Index
Figure 5 shows the results of a study taking spectral measurements of different vegetation indices across a transect of bare soil. Moving from dry clay loam to wet clay loam, we see a very strong response of NDVI due to the wetness of the soil; undesirable if we’re measuring vegetation. We’re not interested in an index that’s sensitive to changes in soil or soil moisture. However, there are a few other indices plotted in figure 5 with much lower sensitivities to variations in the soil across the transect.
Figure 5: Qi et al. (1994) Rem. Sens. Env.
The first one of those indices is the Soil Adjusted Vegetation Index (SAVI). The equation for SAVI is similar to NDVI. It incorporates the same two bands as the NDVI—the near infrared and the red.
Soil Adjusted Vegetation Index (Huete (1988) Rem. Sens. Env.)
The only thing that’s different, is the L parameter. L is a soil adjustment factor with values that range anywhere from 0 to 1. When vegetation cover is 100%, L is 0 because there’s no need for a soil background adjustment. However, when vegetation cover is very low, that L parameter will approach one. Because it is difficult to measure exactly how much vegetation cover you have without using NDVI, we can modify the NDVI so it’s not sensitive to soil by guessing beforehand what L should be. It’s common practice to set L to an intermediate value of 0.5. You can see in Figure 5 the Soil Adjusted Vegetation Index or SAVI has a much lower sensitivity to the soil background.
Solution 2- Modified SAVI
The next vegetation index is the modified SAVI (MSAVI). The SAVI equation contains an L parameter that we have to estimate—not an accurate way of handling things. So a scientist named Key developed a universal optimum for L. We won’t get into the math, but he was able to simplify the SAVI equation to where there’s no longer a need for the L parameter, and the only inputs required are the reflectances in the near infrared and the red.
Modified SAVI (Qi et al. (1994) Rem. Sens. Env.)
This was a pretty significant advance as it circumvented the need to estimate or independently measure L. When Key compared SAVI to MSAVI, there was virtually no difference between the two indices in terms of their sensitivity to the amount of vegetation and their response to the soil background.
MSAVI compares well with SAVI in terms of dynamic range and noise level (Qi et al. (1994) Rem. Sens. Env.)
Modern technology has made it possible to sample Normalized Difference Vegetation Index (NDVI) across a range of scales both in space and in time, from satellites sampling the entire earth’s surface to handheld small sensors that measure individual plants or even leaves.
Figure 1: NDVI is sensitive to the amount of vegetation cover that is present across the earth’s surface.
NDVI – Global
The broadest way to think of NDVI is data obtained from an earth orbiting satellite. In the figure above, you can see highly vegetated areas that have high NDVI values represented by dark green colors across the globe. Conversely, areas of low vegetation have low NDVI values, which look brown. NDVI is sensitive to the amount of vegetation cover that is present across the earth’s surface.
NDVI – Local
How might NDVI be useful at the plot level? Figure 2 below shows a successional gradient where time zero is a bare patch of soil, or a few forbs or annual grasses. If we leave that patch of ground for enough time, the vegetation will change: shrubs may take over from grasses and eventually we might see a forest. Across a large area, we may also move from grasslands to forest. In an agricultural system, there is yearly turnover of vegetation—from bare field to plant emergence, maturity, and senescence. This cycle repeats itself every year. Within these growth cycles NDVI helps to quantify the canopy growth that occurs over time as well as the spatial dynamics that occur across landscapes.
Figure 2: Seasonal growth plotted against spatiotemporal variation
Spectral Reflectance Data
So where does NDVI come from? In Figure 3, the x-axis plots wavelength of light within the electromagnetic spectrum; 450 to 950 nm covers both the visible region and a portion of the near infrared. On the y-axis is percent reflectance. This is a typical reflectance spectrum from green vegetation.
The green hyperspectral line is what we would expect to get from a spectral radiometer. Reflectance is typically low in the blue region, higher in the green region, and lower in the red region. It shifts dramatically as we cross from the visible to the near infrared. The two vertical bars labeled NDVI give you an idea of where a typical NDVI sensor measures within the spectrum. One band is in the red region and the other is in the near-infrared region.
NDVI capitalizes on the large difference between the visible region and the near infrared portion of the spectrum. Healthy, growing plants reflect near-infrared strongly. The two images on the right of the figure above are of the same area. The top image is displayed in true color, or three bands–blue, green and red. The image below is a false color infrared image. The three bands displayed are blue, green, and in place of red, we used the near infrared. The bright red color indicates a lot of near infrared reflectance which is typical of green or healthy vegetation.
The reason NDVI is formulated with red and near infrared is because red keys in on chlorophyll absorption, and near infrared is sensitive to canopy structure and the internal cellular structure of leaves. As we add leaves to a canopy, there’s more chlorophyll and structural complexities, thus we can expect decreasing amounts of red reflectance and higher amounts of near-infrared reflectance.
How Do We Calculate the NDVI?
The Normalized Difference Vegetation Index takes into account the amount of near-infrared (NIR) reflected by plants.It is calculated by dividing the difference between the reflectances (Rho) in the near-infrared and red by the sum of the two. NDVI values typically range between negative one (surface water) and one (full, vibrant canopy). Low values (0.1 – 0.4) indicate sparse canopies, while higher values (0.7 – 0.9) suggest full, active canopies.
The way we calculate the percent reflectance is to quantify both the upwelling radiation (the radiation that’s striking the canopy and then reflected back toward our sensor) as well as the total amount of radiation that’s downwelling (from the sky) on a canopy. The ratio of those two give us percent reflectance in each of the bands.
Next Week:Learn about NDVI applications, limitations, and how to correct for those limitations.
With the recent news coverage of the SMAP (Soil Moisture Active Passive) satellite launch, researchers may wonder: what does remote sensing mean for the future of in situ measurements? We asked two scientists, Drs. Colin Campbell and Chris Lund, for answers to this complex question. Here’s what they had to say:
What is SMAP?
SMAP is an orbiting earth observatory that estimates soil moisture content in the top 5 cm of soil over the entire earth. The mission is three years long with measurements taken every 2-3 days. This will allow seasonal changes around the world to be observed over time, improving our ability to manage water resources and better parameterize land surface models. SMAP determines the amount of water found between the minerals, rocky material, and organic particles found in soil by measuring the ability of radar to penetrate the soil. The wetter the soil is, the less the radar will penetrate. SMAP has two different sensors on the platform: an L band aperture radar with a resolution of about a kilometer when it’s looking straight down (the pixel size is about 1 km by 1 km), combined with a passive radiometer with about 40 km of resolution. This combination creates a synthetic product that takes advantage of the sensitivity of the radiometer.
What does SMAP mean for in situ soil water content measurement?
It’s all about scale: In some ways, comparing in situ to SMAP measurements is like comparing apples to…well…mountain-sized apples. The two forms of measurement use vastly different scales. In situ soil moisture sensors measure water content at the volume of several liters of soil, maximum. Even the sensor with the largest field of sensitivity, the neutron probe, can only integrate a volleyball-sized volume. On the other hand, SMAP measures at a resolution of 1 km2, which is larger than the size of a quarter section, a large field for many farmers. Global soil moisture maps will allow scientists using SMAP to look at big picture applications like weather, climate and hydrological forecasting, drought, and flooding, while more detailed in situ measurements will tell a farmer when it’s time to water, or help researchers discover exactly why plants are growing in one location versus another. The difference in spatial scale makes the two forms of measurement useful for very different research purposes and applications. However, there are applications where the two measurements can be complementary. Most notably, in situ measurements are often temporally rich while being spatially poor. But, SMAP can be used to scale in situ measurements to areas where in situ measurements are absent. In situ measurements can also be used as a source of validation data for SMAP-derived values for any location where both in situ and SMAP measurements overlap. Thus, there is opportunity for synergy when pairing SMAP and in situ measurements.
Satellite image in Winter.
What can SMAP do that in situ measurement can’t?
Scientists say they’ve seen a relationship between the top 5 cm of soil moisture and some factors related to climate change and weather. Because in situ soil sensors sample across a spatial footprint of a few meters, it can be very difficult to use their data to say anything about processes occurring across broad spatial scales; two liters of soil is not going to tell you anything about weather or flooding. SMAP can help us better understand the interaction between the land surface and atmosphere, improving our understanding of the global water cycle as well as regional and global climate. This will help with forecasting crop yield, pest pressure, and disease…that’s big picture research.
The productivity of a forest also may depend on the general soil moisture measured by SMAP. For instance, if we got an idea of the soil moisture and greenness of a forest, we could tie together the approximate water availability and the resulting biomass accumulation with incoming solar radiation. Better biomass accumulation models could lead to better validation of global carbon cycle models.
SMAP will also be able to detect dry areas across the U.S. and challenges they might present. Surface runoff that leads to flooding could also be predicted as scientists will be able to see where soils reach saturated conditions.
In other applications, people working on global water or energy budgets have to parameterize the land surface in terms of how wet or dry it is. That’s the big advantage of SMAP’s relatively new data sets. Any time you’re running a regional climate model you have to parameterize what the soil moisture is in order to partition surface heat flux into sensible and latent heat flux. If there’s a lot of available water, it’s weighted more toward evaporation and less toward sensible heat flux. In areas where there’s little available water and low evaporation, you get high surface temperatures and sensible heat flux. So SMAP will be important for model parameterization as we haven’t had a good global data set for soil moisture until now.
In situ sensors show how much water is lost from the root zone and what is still left.
What can in situ sensors do that SMAP can’t?
In irrigated agriculture, farmers need to know when and how much to irrigate. In situsensors give them this information by showing how much water was lost from the root zone and what is still left. SMAP is unable to tell you what’s down in the root zone; it only reaches to 5 cm. Additionally, 1 km resolution is larger than most irrigation blocks. These factors mean that it will be difficult to make irrigation decisions from SMAP alone.
Scientists using in situ sensors are concerned with the soil moisture available in a local area because their time resolution is excellent and they have the ability to resolve what’s happening in particular conditions related to crops or natural systems. Natural systems are often heterogeneous, meaning there may be adjacent areas with different types of vegetation including trees, shrubs, and grass. Tree roots may grow deep while grass roots are shallow. Being able to look over all these different areas without averaging them together, as SMAP does, is critical in some applications.
What about geotechnical applications? Literature suggests SMAP output can help predict landslides. It is more likely that it can only see when the soil is generally saturated and generate a warning. But in slopes that are at risk of landslides, in situ monitoring with sensors such as tensiometers to measure positive pore water pressure may be more useful for determining when a slide is imminent.
SMAP, like in situ water content measuring systems, is also limited by the fact that it measures the amount, not the availability, of water. If it measures 23% water content in a certain area, that measurement may not tell us what we want to know. A clay soil at 23% VWC will be close to wilting point while a sand would be above the plant optimal range. SMAP doesn’t measure the energy status of water (water potential), so even if SMAP tells us a field has water content, that water might not be readily available. Water availability must be determined through a pedo-transfer function or moisture release curve appropriate for a specific soil type (It is possible to overlay SMAP data on soil type data to estimate energy state, but this might not be fine enough resolution to be useful).
How do SMAP and in situ instruments work together? The key is ground truthing in situ soil moisture measurements with SMAP type satellites and vice versa. Ground-based measurements at specific locations can be matched with satellite information to extrapolate over a field and gain confidence in the small continuous scale alongside the larger infrequent scale. It’s analogous of a video camera recording one plant continuously while a single shot camera snaps whole-field pictures every day. With the SMAP “single-shot” we can say, something changed from time A to time B, but we don’t know what happened in the middle (rain event, etc.). In situ measurements will tell us the details of what happened in between each snapshot. Putting both data sets together and matching trends, we can show correlation and complete the soil moisture picture. Basically, In situ measurements provide temporally rich information about soil moisture from a postage stamp-sized area of earth’s surface (driven by highly localized conditions), whereas SMAP gives us the ability to monitor broad scale spatiotemporal patterns across all of earth’s surface (driven by synoptic conditions).
Dr. Christopher Lund is a research scientist and product manager for METER’s new irrigation management instrumentation group. He has more than a decade of experience working with land surface flux measurements, terrestrial water budgets, and soil-vegetation-atmosphere transfer scheme modeling. Prior to joining METER, he served as a research scientist on the NASA-CSUMB SIMS (Satellite Irrigation Management Support) Project, a multi-year collaboration between the California Department of Water Resources, NASA, and CSU Monterey Bay providing California growers with novel irrigation decision support tools. Dr. Lund’s current research focuses on developing cost-effective irrigation management instrumentation for commercial markets. Dr. Lund will be giving a talk on innovations in agricultural remote sensing at the Third Professional Workshop on Technology For Irrigation Scheduling. He will talk about his work with the SIMS team and what growers can do with remote sensing data to estimate things like evapotranspiration. He’ll also address how to improve those estimates by combining them with field measurements from ground based instrumentation such as soil moisture sensors.
Image: USGS Landsat Project Website
“The advantage of satellite remote sensing is that it allows you to look at many fields at once and also integrate across spatial variability. The down side is it doesn’t give you access to everything you might want for irrigation management, so there are certain things you have to measure on the ground. When it comes to remote sensing data and ground measurements, I don’t think it’s an either/or situation. I think the future is hybrid products utilizing both remote sensing and ground based measurements,” he says.
He will also speak on how satellite derived NDVI data can benefit from new inexpensive ground based-sensors like the SRS. This enables scientists to make sure that their satellite NDVI data accurately reflect what’s happening on the ground.
The seminar will be held at the Third Professional Workshop On Technology For Irrigation Scheduling on February 11, 2015 at the CREA auditorium, Calle Jose Galan Merino Sevilla, Spain.