Skip to content

Posts from the ‘Irrigation’ Category

Stem Water Content Changes Our Understanding of Tree Water Use

In an update to our previous blog, “Soil Moisture Sensors in a Tree?”, we highlight two current research projects using soil moisture sensors to measure volumetric water content (VWC) in tree stems and share why this previously difficult-to-obtain measurement will change how we look at tree water usage.

stem water content

Researchers explore the feasibility of inserting capacitance soil sensors in tree stems as a real-time measurement.

Soil Moisture Sensors in Tree Stems?

In a recent research project, Ph.D. candidate Ashley Matheny of the University of Michigan used soil sensors to measure volumetric water content in the stems of two species of hardwood trees in a northern Michigan forest: mature red oak and red maple.  Though both tree types are classified as deciduous, they have different strategies for how they use water. Oak is anisohydric, meaning the species doesn’t control their stomata to reduce transpiration, even in drought conditions.  Isohydric maples are more conservative. If the soil starts to dry out, maple trees will maintain their leaf water potential by closing their stomata to conserve water.  Ashley and her research team wanted to understand the different ways these two types of trees use stem water in various soil moisture scenarios.

Historically, tree water storage has been measured using dendrometers and sap flow data, but Ashley’s team wanted to explore the feasibility of inserting a capacitance-type soil sensor in the tree stems as a real-time measurement.  They hoped for a practical way to make this measurement to provide more accurate estimations of transpiration for use in global models.  

Scientists measured volumetric water content in the stems of two species of hardwood trees in a northern Michigan forest: mature red oak and red maple.

Measurements used

Ashley and her team used meteorological, sap flux, and stem water content measurements to test the effectiveness of capacitance sensors for measuring tree water storage and water use dynamics in one red maple and one red oak tree of similar size, height, canopy position and proximity to one another (Matheny et al. 2015). They installed both long and short soil moisture probes in the top and the bottom of the maple and oak tree stems, taking continuous measurements for two months. They calibrated the sensors to the density of the maple and oak woods and then inserted the sensors into drilled pilot holes.  They also measured soil moisture and temperature for reference, eventually converting soil moisture measurements to water potential values.

Results Varied According to Species

The research team found that the VWC measurements in the stems described tree storage dynamics which correlated well with average sap flux dynamics.  They observed exactly what they assumed would be the anisohydric and isohydric characteristics in both trees.  When soil water decreased, they saw that red oak used up everything that was stored in the stem, even though there wasn’t much available soil moisture.  Whereas in maple, the water in the stem was more closely tied to the amount of soil water. After precipitation, maple trees used the water stored in their stem and replaced it with more soil water.  But, when soil moisture declined, they held onto that water and used it at a slower rate.

stem water content

Researchers want to figure out the appropriate level of detail for tree water-use strategy in a global model.

Trees use different strategies at the species level

The ability to make a stem water content measurement was important to these researchers because much of their work deals with global models representing forests in the broadest sense possible.  They want to figure out the appropriate level of detail for tree water-use strategy in a global model. Both oak and the maple are classified as broadleaf deciduous, and in a global model, they’re lumped into the same category. But this study illustrates that if you’re interested in hydrodynamics (the way that trees use water), deciduous trees use different strategies at the species level.  Thus, there is a need to treat them differently to produce accurate models.

Read the full study in Ecosphere.

Reference: Matheny, A. M., G. Bohrer, S. R. Garrity, T. H. Morin, C. J. Howard, and C. S. Vogel. 2015. Observations of stem water storage in trees of opposing hydraulic strategies. Ecosphere 6(9):165.

Next week: Learn about more research being done using soil moisture sensors to measure volumetric water content in tree stems.

Get more information on applied environmental research in our

Electrical Conductivity of Soil as a Predictor of Plant Response (Part 2)

Salt in soil comes from the fertilizer we apply but also from irrigation water and dissolving soil minerals.  If more salt is applied in the irrigation water than is leached or taken off in harvested plants, the soil becomes more saline and eventually ceases to support agricultural production (see part 1).  This week, learn an effective way to measure electrical conductivity (EC) in soil.

Electrical conductivity

Salt in irrigation water reduces its water potential, making it less available to the plant.

How to Measure Electrical Conductivity of the Soil Solution

As mentioned above, the earliest measurements of solution conductivity were made on soil samples, but it was found to be more reliable to extract the soil solution and make the measurements on it. When values for unsaturated soils are needed, those are calculated based on the saturation numbers and conjecture about how the soil dried to its present state. Obviously a direct measurement of the soil solution conductivity would be better if it could be made reliably.

Two approaches have been made to this measurement. The first uses platinum electrodes embedded in ceramic with a bubbling pressure of 15 bars. Over the plant growth range the ceramic remains saturated, even though the soil is not saturated, allowing a measurement of the solution in the ceramic. As long as there is adequate exchange between the ceramic and the soil solution, this measurement will be the EC of the soil solution, pore water EC.

Salt in soil comes from the fertilizer we apply, irrigation water and dissolving soil minerals.

The other method measures the conductivity of the bulk soil and then uses empirical or theoretical equations to determine the pore water EC. The ECH2O 5TE uses the second method. It requires no exchange of salt between soil and sensor and is therefore more likely to indicate the actual solution electrical conductivity. The following analysis shows one of several methods for determining the electrical conductivity of the saturation extract from measurements of the bulk soil electrical conductivity.

Mualem and Friedman (1991) proposed a model based on soil hydraulic properties. It assumes two parallel conduction paths: one along the surface of soil particles and the other through the soil water. The model is

Equation 1

Here σb is the bulk conductivity which is measured by the probe, σs is the bulk surface conductivity, σw is the conductivity of the pore water, θ is the volumetric water content, θs is the saturation water content of the soil and n is an empirical parameter with a suggested value around 0.5. If, for the moment, we ignore surface conductivity, and use eq. 1 to compute the electrical conductivity of a saturated paste (assuming n = 0.5 and θs = 0.5) we obtain σb = 0.35σw. Obviously, if no soil were there, the bulk reading would equal the electrical conductivity of the water. But when soil is there, the bulk conductivity is about a third of the solution conductivity. This happens because soil particles take up some of the space, decreasing the cross section for ion flow and increasing the distance ions must travel (around particles) to move from one electrode of the probe to the other. In unsaturated soil these same concepts apply, but here both soil particles and empty pores interfere with ion transport, so the bulk conductivity becomes an even smaller fraction of pore water conductivity.

Electrical conductivity

When water evaporates at the soil surface, or from leaves, it is pure, containing no salt, so evapotranspiration concentrates the salts in the soil.

Our interest, of course, is in the pore water conductivity. Inverting eq. 1 we obtain

Equation 2

In order to know pore water conductivity from measurements in the soil we must also know the soil water content, the saturation water content, and the surface conductivity. The 5TE measures the water content. The saturation water content can be computed from the bulk density of the soil

Electrical conductivity

Equation 3

Where ρb is the soil bulk density and ρs is the density of the solid particles, which in mineral soils is taken to be around 2.65 Mg/m3 . The surface conductivity is assumed to be zero for coarse textured soil. Therefore, using the 5TE allows us to quantify pore water EC through the use of the above assumptions. This knowledge has the potential to be a very useful tool in fertilizer scheduling.

Electrical Conductivity is Temperature Dependent

Electrical conductivity of solutions or soils changes by about 2% per Celsius degree. Because of this, measurements must be corrected for temperature in order to be useful. Richards (1954) provides a table for correcting the readings taken at any temperature to readings at 25 °C. The following polynomial summarizes the table

where t is the Celsius temperature. This equation is programmed into the 5TE, so temperature corrections are automatic.

Electrical conductivity

Soil salinity has been measured using electrical conductivity for more than 100 years.

Units of Electrical Conductivity

The SI unit for electrical conductance is the Siemen, so electrical conductivity has units of S/m. Units used in older literature are mho/cm (mho is reciprocal ohm), which have the same value as S/cm. Soil electrical conductivities were typically reported in mmho/cm so 1 mmho/cm equals 1 mS/cm. Since SI discourages the use of submultiples in the denominator, this unit is changed to deciSiemen per meter (dS/m), which is numerically the same as mmho/cm or mS/cm. Occasionally, EC is reported as mS/m or µS/m. 1 dS/m is 100 mS/m or 105 µS/m.


Richards, L. A. (Ed.) 1954. Diagnosis and Improvement of Saline and Alkali Soils. USDA Agriculture Handbook 60, Washington D. C.

Rhoades, J. D. and J. Loveday. 1990. Salinity in irrigated agriculture. In Irrigation of Agricultural Crops. Agronomy Monograph 30:1089-1142. Americal Society of Agronomy, Madison, WI.

Get more information on applied environmental research in our


Electrical Conductivity of Soil as a Predictor of Plant Response

Plants require nutrients to grow, and if we fail to supply the proper nutrients in the proper concentrations, plant function is affected. Fertilizer in too high concentration can also affect plant function, and sometimes is fatal.

electrical conductivity

Plant function is affected by nutrient concentration.

Most of us have had the experience of fertilizing some part of a lawn too heavily, perhaps by accident, and killing grass in that part of the lawn. Generally, it isn’t the nutrients themselves that cause the damage, it is their effect on the water. Salt in the water reduces its water potential making it less available to the plant. The salt therefore causes water stress in the plant.

Salt in soil comes from the fertilizer we apply, but also from irrigation water and dissolving soil minerals. Relatively small amounts are removed with the plants that are harvested, but most leaches with the water out of the bottom of the soil profile. When water evaporates at the soil surface, or from leaves, it is pure, containing no salt, so evapotranspiration concentrates the salts in the soil. If more salt is applied in the irrigation water than is leached or taken off in harvested plants, the soil becomes more saline and eventually will cease to support agricultural production. Thousands of acres have been lost from production in this way, and production has been drastically reduced on tens of thousands of additional acres.

electrical conductivity

Thousands of acres have been lost from over-fertilization.

Soil Salinity and Electrical Conductivity

Soil salinity has been measured using electrical conductivity for more than 100 years. It is common knowledge that salty water conducts electricity. Whitney and Means (1897) made use of that fact to measure the concentration of salt in soil. Early methods made measurements directly on a soil paste, but the influence of the soil in the paste on the measurement was not fully understood until recently, leading to uncertainty in the measurements. By about 1940 the accepted method for determining soil salinity was to make a saturated paste by a specified procedure, extract solution from the paste, and measure the electrical conductivity of the solution (Richards, 1954). The measurement is referred to as the electrical conductivity of the saturation extract. These values were then correlated with crop response.

Richards (1954) defined 4 soil salinity classes, as shown in Table 1. Crops suitable for these classes are also listed by Richards, but a much more extensive list is given by Rhoades and Lovejoy (1990). For example, bean is listed as a sensitive crop. It can only be grown without yield damage in soils with EC below 2 dS/m. Barley is a tolerant crop. It can be grown without much yield reduction in any soil up to EC of 16 dS/m.

Table 1: Salinity classes for soils

Two other columns are shown in the table. The “salt in soil” column shows how much salt is required to salinize a soil. In terms of the total soil mass, only a small percentage change is needed to make a big difference in salinity, but this would still represent a large addition of fertilizer. A 200 kg/ha addition of fertilizer would represent a fairly high rate. If this were incorporated into the top 15 cm of soil, it would represent

electrical conductivity

This wouldn’t cause much change in soil salt percentage.

The other column shows osmotic potential of the saturation extract. To give some reference for this number, remember that the nominal permanent wilt water potential of soil is -1500 kPa. Osmotic potentials of plant leaves vary widely depending on species, but -1500 kPa is a kind of median value. The values in the table may seem small compared to the permanent wilt (PW) value, but remember that these are values at saturation. When a soil is saturated, water quickly drains to a “drained upper limit” (UL) water content which is around half the saturation value. The useful water storage of the soil is between the UL and the PW or lower limit water content, which, again, is about half the UL. The concentration of salts at the UL is about the same as at saturation because the water drained away, but the water loss between the UL and PW is typically by evapotranspiration, so little or no salts are lost. The concentration at the lower limit is therefore twice that shown in Table 1, which is significant compared to the permanent wilt water potential. Likewise the osmotic potential of the soil solution after fertilizing with 200 kg/ka and mixing wouldn’t change much, but the same amount of fertilizer concentrated in a band near seed would have a much larger effect.

Next Week: Read part 2 of Electrical Conductivity as a Predictor of Soil Response.

Get more information on applied environmental research in our


Avocado Growers in Kenya Fight Drought with Recycled Water Bottle Irrigation (Part 2)

Dr. Brent Clothier, Dr. Steve Green, Roberta Gentile and their research team from Plant and Food Research in New Zealand are working in Kenya to alleviate the poverty of the many small-holder farmers who grow avocados in the Central Highlands of Kenya (see part 1). This week, read about an inexpensive irrigation solution for these farmers and how the researchers are developing a plan to manage nutrients.


The period of water stress in October is at the time of main flowering.

Recycled Water Bottles Provide a Solution

When the team was visited Kenya in early March, the Long Rains had not arrived, and the trees were under water stress. The researchers sought to reduce the impact of drought by using a prototype of a portable drip-irrigation system they developed. They used ‘old’ 20 liter drinking water bottles to deliver water to the trees at 4 L/hr.


20 L water bottles used for tree irrigation.

The bottles can be refilled and moved from tree to tree. By measuring water content in the soil, the team found that the 20 L of drip irrigated water lasted in the soil about 2 days. When the period was increased to 4 days, the root water uptake was reduced over days 3 and 4 after wetting. Thus they recommended the bottle be recharged and reapplied every two days. This enables the bottle to be used on another tree on the intervening day and should help the farmers to reduce the worst impacts of the drought while waiting for the Long Rains to arrive.


Refilling the water bottles.

Replacing Low Soil Nutrients

In another phase of the experiment, Dr. Clothier’s team surveyed soil and plant nutrient contents in the main avocado production regions to assess the current fertility status of the farms. Soils in this region are classified as Nitisols, deep red soils with a nut-shaped structure and high iron content (Jones et al. 2013). These soils have low levels of organic matter and low pH. Soil sampling revealed a decrease in pH and increase in organic matter with altitude in the Kandara valley. This observed gradient is likely attributable to the higher amounts rainfall received in the higher altitudes of the valley, which can increase organic matter production and leach base cations from the soil. Soil and leaf nutrient analyses of the monitoring farms showed similar trends in nutrient availability. There are also low levels of the macronutrients nitrogen and phosphorus and the micronutrient boron in these soils. These nutrients are essential for avocado growth and production. One challenge to improve avocado productivity is finding ways to improve soil nutrient availability and tree nutrition.


An example of the benefits of a secure revenue-stream: One farmer purchased a new cow, which enables him to meet the nutrient requirements of more avocado trees.

A Plan for Managing Nutrients

The majority of the small-holder farms supplying avocados to Olivado use organic production methods. This means organic amendments such as plant residues, composts and animal manures are required to replenish the nutrients that are exported from the farms and improve soil fertility. Livestock have the potential to provide nutrient amendments for a considerable number of avocado trees. Even better, the input of organic materials will build-up soil organic matter levels, which benefit soil conservation, water holding capacity, pH buffering, and soil biological activity.

The researchers are developing simple nutrient budgets for these avocado trees using yield and fruit nutrient concentration data to assess the quantity of nutrients being exported off-farm in the harvested crop. Using the nutrient concentrations of locally available organic amendments, they will provide recommendations on the amount of organic material needed to sustain soil fertility.

Nutrient balances will be incorporated into a decision support tool to assist small-holder farmers in enhancing their soil and plant nutrition. These budgets will be enhanced by further characterizing the nutrient composition and quantities of available organic matter amendments in the region. The researchers are working to improve these nutrient budget estimates with data specific to the avocado farms in the region. They will also set up demonstration farms to evaluate the production responses to recommended nutrient management practices.

To find out more about Kenyan avocado research contact Brent Clothier: .

(This article is a summary/compilation of several articles first printed in WISPAS newsletter)


Jones, A., Breuning-Madsen, H., Brossard, M., Dampha, A., Deckers, J., Dewitte, O., Gallali, T., Hallett, S., Jones, R., Kilasara, M., Le Roux, P., Micheli, E., Montanarella, L., Spaargaren, O., Thiombiano, L., Van Ranst, E., Yemefack, M., Zougmore, R., (eds.) 2013. Soil Atlas of Africa. European Commission, Publications Office of the European Union, Luxembourg. 176 pp.

Get more information on applied environmental research in our


Avocado Growers in Kenya Fight Drought with Recycled Water Bottle Irrigation

Dr. Brent Clothier, Dr. Steve Green, Roberta Gentile and their research team from Plant and Food Research in New Zealand are working in Kenya to alleviate the poverty of the many small-holder farmers who grow avocados in the Central Highlands of Kenya. These farmers have old and very large avocado trees. The fruit from these trees are purchased by the company Olivado EPZ who presses over 1300 small-holders’ avocados for oil. Dr. Clothier and his team are investigating how to increase the productivity of the farmers’ avocado trees and increase the quality of the fruit so they yield more oil.


Small-holder farmers grow avocados in the Central Highlands of Kenya.

Reducing Leaf Area to Avoid Water Stress

Because of the age and size of these trees, harvesting of the avocados is difficult and time consuming, and through dropped fruit, the quality of the avocados can be comprised. In addition, any dry season water-stress negatively impacts fruit filling. The research team performed some initial remedial pruning of these trees to develop a more manageable and productive tree form. They sought to assess whether the reduced leaf area would enable the trees to avoid water stress during the dry season of January through March between the short and long rainy seasons. They removed 30-40% of the central limbs of the avocado tree to create a more open canopy form.

The team instrumented two trees with heat-pulse sap-flow probes. One tree was left unpruned and the tree in the photo above was pruned. The tree that was pruned was using between 300-400 liters per day, as expected for a tree of that large size. The unpruned tree was smaller in size, and it was using between 150-250 liters per day during May and June. The selective limb pruning resulted in the rate of water-use dropping to 200-300 liters per day, a drop of 100 liters per day.


The more open canopy form of the pruned avocado tree.

Determining Tree Water Use During Rainy and Dry Seasons

The team also measured the water-use of four trees of different sizes during the entire season using the compensation heat-pulse method and soil water content. They found the trees’ water-use doubled with the arrival of the Short Rains and then began to decline in early January after the rains ended. The trees were under a degree of water stress prior to the arrival of the (short) Short Rains, and as the weak Short Rains ended early, the trees again went into water stress with only occasional respite due to isolated rainstorms in January and February.

This pattern of water stress presents a challenge for sustaining high levels of avocado production. The period of water stress in October is at the time of main flowering, and researchers who were there noted a carpet of aborted flowers on the orchard floor. They also noticed that the fruit were smaller at one farm than those higher up in the Central Highlands where rainfall is higher and more frequent. Thus, to improve production it is imperative to mitigate the impacts of drought, and this needs to be done without reference to any infrastructure for irrigation.

Next week: Read about an inexpensive irrigation solution for these farmers and how the researchers are developing a plan to manage nutrients.

(This article is a summary/compilation of several articles first printed in WISPAS newsletter)

Get more information on applied environmental research in our


Soil Moisture Sensors: Why TDR VS. Capacitance May Be Missing the Point (Part 2)

Dr. Colin S. Campbell discusses whether TDR vs. capacitance (see part 1) is the right question, the challenges facing soil moisture sensor technology, and the correct questions to ask before investing in a sensor system.

It’s easy to overlook the obvious question: what is being measured?

What are You Trying to Measure?

When considering which soil water content sensor will work best for any application, it’s easy to overlook the obvious question: what is being measured?  Time Domain Reflectometry (TDR) vs. capacitance is the right question for a researcher who is looking at the dielectric permittivity across a wide measurement frequency spectrum (called dielectric spectroscopy). There is important information in these data, like the ability to measure bulk density along with water content and electrical conductivity. If this is the desired measurement, currently only one technology will do: TDR. The reflectance of the electrical pulse that moves down the conducting rods contains a wide range of frequencies.  When digitized, these frequencies can be separated by fast fourier transform and analyzed for additional information.

The objective for the majority of scientists, however, is to simply monitor soil water content instantaneously or over time, with good accuracy. There are more options if this is the goal, yet there are still pitfalls to consider.


Considerable research has been devoted to determining which soil moisture sensors meet expectation.

Each Technology Has Challenges

Why would a scientist pay $100+ for a soil volumetric water content (VWC) sensor, when there are hundreds of soil moisture sensors online costing between $5 and $15? This is where knowing HOW water content is measured by a sensor is critical.

Most sensors on home and garden websites work based on electrical resistivity or conductivity. The principle is simple: more water will allow more electrons to flow. So conductivity will change with soil water content. But, while it’s possible to determine whether water content has changed with this method, absolute calibration is impossible to achieve as salts in the soil water will change as the water content changes. A careful reading of sensor specs will sometimes uncover the measurement method, but sometimes, price is the only indication.

Somewhere between dielectric spectroscopy and electrical resistance are the sensors that provide simple, accurate water content measurement. Considerable research has been devoted to determining which of these meet expectation, and the results suggest that Campbell Scientific, Delta-T, Stevens, Acclima, Sentek, and METER (formerly Decagon Devices), provide accurate sensors vetted by soil scientists. The real challenge is installing the sensors correctly and connecting them to a system that meets data-collection and analysis needs.

Installation Techniques Affect Accuracy

Studies show there is a difference between mid-priced sensor accuracy when tested in laboratory conditions. But, in the field, sensor accuracy is shown to be similar for all good quality probes, and all sensors benefit from site specific soil calibration. Why? The reason is associated with the principle upon which they function. The electromagnetic field these sensors produce falls off exponentially with distance from the sensor surface because the majority of the field is near the electrodes. So, in the lab, where test solutions form easily around sensor rods, there are differences in probe performance.  In a natural medium like soil, air gaps, rocks, and other detritus reduce the electrode-to-soil contact and tend to reduce sensor to sensor differences. Thus, picking an accurate sensor is important, but a high quality installation is even more critical.


Improper installation is the largest barrier to accuracy.

Which Capacitance Sensor Works Best?

Sensor choice should be based on how sensors will be installed, the nature of the research site, and the intended collection method. Some researchers prefer a profile sensor, which allows instruments to be placed at multiple depths in a single hole. This may facilitate fast installation, but air gaps in the auger pilot hole can occur, especially in rocky soils. Fixing this problem requires filling the hole with a slurry, resulting in disturbed soil measurements. Still, profile sensor installation must be evaluated against the typical method of digging a pit and installing sensors into a side-wall. This method is time consuming and makes it more difficult to retrieve sensors.

New technology that allows sensor installation in the side of a 10 cm borehole may give the best of both worlds, but still requires backfill and has the challenge of probe removal at the end of the experiment.

The research site must also be a consideration. If the installation is close to main power or easily reached with batteries and solar panels, your options are open: all sensors will work. But, if the site is remote, picking a sensor and logging system with low power requirements will save time hauling in solar panels or the frustration of data loggers running out of batteries.


Often times it comes down to convenience.

Data Loggers Can Be a Limitation

Many manufacturers design data loggers that only connect to the sensors they make. This can cause problems if the logging system doesn’t meet site needs. All manufacturers mentioned above have sensors that will connect to general data loggers such as Campbell Scientific’s CR series. It often comes down to convenience: the types of sensor needed to monitor a site, the resources needed to collect and analyze the data, and site maintenance. Cost is an issue too, as sensors range from $100 to more than $3000.

Successfully Measure Water Content

The challenge of setting up and monitoring soil water content is not trivial, with many choices and little explanation of how each type of sensor will affect the final results. There are a wealth of papers that review the critical performance aspects of all the sensors discussed, and we encourage you to read them. But, if soil water content is the goal, using one of the sensors from the manufacturers named above, a careful installation, and a soil-specific calibration, will ensure a successful, accurate water content measurement.

For an in-depth comparison of TDR versus capacitance technology, read: Dielectric Probes Vs. Time Domain Reflectometers

For an understanding of how capacitance sensors compare to other major contemporary sensor technologies, watch our Soil Moisture 201 webinar.

Soil Moisture Sensors: Why TDR vs. Capacitance May Be Missing the Point

Time Domain Reflectometry (TDR) vs. capacitance is a common question for scientists who want to measure volumetric water content (VWC) of soil, but is it the right question?  Dr. Colin S. Campbell, soil scientist, explains some of the history and technology behind TDR vs. capacitance and the most important questions scientists need to ask before investing in a sensor system.

TDR vs. Capacitance

TDR began as a technology the power industry used to determine the distance to a break in broken power lines.

Clarke Topp

In the late 1970s, Clarke Topp and two colleagues began working with a technology the power industry used to determine the distance to a break in broken power lines.  Time Domain Reflectometers (TDR) generated a voltage pulse which traveled down a cable, reflected from the end, and returned to the transmitter. The time required for the pulse to travel to the end of the cable directed repair crews to the correct trouble spot. The travel time depended on the distance to the break where the voltage was reflected, but also on the dielectric constant of the cable environment.  Topp realized that water has a high dielectric constant (80) compared to soil minerals (4) and air (1).  If bare conductors were buried in soil and the travel time measured with the TDR, he could determine the dielectric constant of the soil, and from that, its water content.  He was thus able to correlate the time it took for an electromagnetic pulse to travel the length of steel sensor rods inserted into the soil to volumetric water content. Despite his colleagues’ skepticism, he proved that the measurement was consistent for several soil types.

TDR vs. Capacitance

TDR sensors consume a lot of power. They may require solar panels and larger batteries for permanent installations.

TDR Technology is Accurate, but Costly

In the years since Topp et al.’s (1980) seminal paper, TDR probes have proven to be accurate for measuring water content in many soils. So why doesn’t everyone use them? The main reason is that these systems are expensive, limiting the number of measurements that can be made across a field. In addition, TDR systems can be complex, and setting them up and maintaining them can be difficult.  Finally, TDR sensors consume a lot of power.  They may require solar panels and larger batteries for permanent installations. Still TDR has great qualities that make these types of sensors a good choice.  For one thing, the reading is almost independent of electrical conductivity (EC) until the soil becomes salty enough to absorb the reflection.  For another, the probes themselves contain no electronics and are therefore good for long-term monitoring installations since the electronics are not buried and can be accessed for servicing, as needed.  Probes can be multiplexed, so several relatively inexpensive probes can be read by one set of expensive electronics, reducing cost for installations requiring multiple probes.

Many modern capacitance sensors use high frequencies to minimize effects of soil salinity on readings.

Advances in Electronics Enable Capacitance Technology

Dielectric constant of soil can also be measured by making the soil the dielectric in a capacitor.  One could use parallel plates, as in a conventional capacitor, but the measurement can also be made in the fringe field around steel sensor rods, similar to those used for TDR.  The fact that capacitance of soil varies with water content was known well before Topp and colleagues did their experiments with TDR.  So, why did the first attempt at capacitance technology fail, while TDR technology succeeded? It all comes down to the frequency at which the measurements are made.  The voltage pulse used for TDR has a very fast rise time.  It contains a range of frequencies, but the main ones are around 500 MHz to 1 GHz.  At this high frequency, the salinity of the soil does not affect the measurement in soils capable of growing most plants.  

Like TDR, capacitance sensors use a voltage source to produce an electromagnetic field between metal electrodes (usually stainless steel), but instead of a pulse traveling down the rods, positive and negative charges are briefly applied to them. The charge stored is measured and related to volumetric water content. Scientists soon realized that how quickly the electromagnetic field was charged and discharged was critical to success.  Low frequencies led to large soil salinity effects on the readings.  This new understanding, combined with advances in the speed of electronics, meant the original capacitance approach could be resurrected. Many modern capacitance sensors use high frequencies to minimize effects of soil salinity on readings.  

TDR vs. Capacitance

NASA used capacitance technology to measure water content on Mars.

Capacitance Today is Highly Accurate

With this frequency increase, most capacitance sensors available on the market show good accuracy. In addition, the circuitry in them can be designed to resolve extremely small changes in volumetric water content, so much so, that NASA used capacitance technology to measure water content on Mars. Capacitance sensors are lower cost because they don’t require a lot of circuitry, allowing more measurements per dollar. Like TDR, capacitance sensors are reasonably easy to install. The measurement prongs tend to be shorter than TDR probes so they can be less difficult to insert into a hole. Capacitance sensors also tend to have lower energy requirements and may last for years in the field powered by a small battery pack in a data logger.   

In two weeks: Learn about challenges facing both types of technology and why the question of TDR vs. Capacitance may not be the right question.

Get more information on applied environmental research in our

New Weather Station Technology in Africa-3

The Trans African Hydro and Meteorological Observatory (TAHMO) project expects to put 20,000 microenvironment monitors over Africa in order to understand the weather patterns which affect that continent, its water, and its agriculture. In the conclusion of our 3 part series, we  interview Dr. John Selker about his thoughts on the project.


The economics of weather data value may be going up because we’re reaching a cusp in terms of humanity’s consumption of food.

In your TEDx talk you estimate that US weather stations directly bring U.S. consumers  31 billion dollars in value per year. Can Africa see that same kind of return?

Even more.  The economics of weather data value may be going up because we’re reaching a cusp in terms of humanity’s consumption of food.  Africa, one could argue, is the breadbasket for this coming century.  Thus, the value of information about where we could grow what food could be astronomical.  It’s very difficult to estimate.  One application of weather data is crop insurance.  Right now, crop insurance is taking off across Africa. The company we’re working with has 180,000 clients just in Kenya.  When we talked about 31 billion dollars in the U.S., that is the value citizens report, but you need to add to that protection against floods, increased food production, water supply management, crop insurance and a myriad of other basic uses for weather data.  In Africa, the value of this type of protection alone pays for over 1,000 times the cost of the weather stations.

Another application for weather data is that in Africa, the valuation of land itself is uncertain. So if, because of weather station data, we find that a particular microclimate is highly valuable, suddenly land goes from having essentially no value to becoming worth thousands of dollars per acre.  It’s really difficult to estimate the impact the data will have, but it could very well end up being worth trillions of dollars.  We have seen this pattern take place in central Chile, where land went from about $200/hectare in 1998 to over $3,000/ha now due to the understanding that it was exceptionally suited to growing pine trees, which represented a change in land value exceeding $3 billion.

Does the effect of these weather stations go beyond Africa?

There’s limited  water falling on the earth, and if you can’t use weather data to invest in the right seeds, the right fertilizer, and plant at the right time in the right place, you’re not getting the benefit you should from having tilled the soil.  So for Africa the opportunity to improve yields with these new data is phenomenal.  

In terms of the world, the global market for calories is now here, so if we can generate more food production in Africa, that’s going to affect the price and availability of food around the world.  The world is one food community at this point, so an entire continent having inefficient production and ineffective structures costs us all.


If we can generate more food production in Africa, that’s going to affect the price and availability of food around the world.

You’re collecting data from Africa. Is it time to celebrate yet?

I think this is going to be one of those projects where we are always chilling the champagne and never quite drinking it.  It is such a huge scope trying to work across a continent.  So I would say we’ve got some stations all over Africa, we’re learning a lot, and we’ve got collaborators who are excited.  We have reason to feel optimistic.  It will be another five years before I’ll believe that we have a datastream that is monumental.  Right now we’re still getting the groundwork taken care of.  By September of this year we expect to have five hundred of stations in place, and then two years from now, over two thousand. This will be a level of observation that will transform the understanding of African weather and climate.


This is a project of hundreds of people across the world putting their hands and hearts in to make this possible.

How do you deal with the long wait for results?  

In science there is that sense you get when you want to know something, and you can see how to get there.  You have a theory, and you want to prove it.  It kind of captures your imagination.  It’s a combination of curiosity and the potential to actually see something happen in the world: to go from a place where you didn’t know what was going on to a place where you do know what’s going on.  I think about Linus Pauling, who made the early discoveries about the double helix.  He had in his pocket the X-ray crystallography data to show that the protein of life was in helical form, and he said, “In my pocket, I have what’s going to change the world.”  When we realized the feasibility of TAHMO, we felt much the same way.”  

Sometimes in your mind, you can see that path: how you might change the world.  It may never be as dramatic as what Pauling did, but even a small contribution has that same excitement of wanting to be someone who added to the conversation, who added to our ability to live more gracefully in the world.  It’s that feeling that carries you along, because in most of these projects you have an idea, and then ten years later you say, “why was it that hard?”  

Things are usually much harder than your original conception, and that energy and curiosity really helps you through some of the low points in your projects.  So, curiosity has a huge influence on scientific progress.  Changing the world is always difficult, but the excitement, curiosity, and working with people, it all fits together to help us draw through the tough slogs.  In TAHMO, I cannot count the number of people who have urged us to keep the effort moving forward and given a lift just when we needed it most.  This is a project of hundreds of people across the world putting their hands and hearts in to make this possible.  Having these TAHMO supporters is an awesome responsibility and concrete proof of the generosity and optimism of the human spirit.

Learn how you can help TAHMO.

How to Get More From Your NDVI Sensor (Part 3)

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.

NDVI Sensor

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.

NDVI Sensor

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.

NDVI Sensor

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). It’s 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.  

NDVI Sensor

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.

NDVI Sensor

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.

NDVI Sensor

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 go read his paper, but here we give you the equation in case you’re interested in using it.

NDVI Sensor

Figure 9

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.

Get more information on applied environmental research in our

Get More From Your NDVI Sensor (Part 2)

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.

NDVI Sensor

Limitations of the Normalized Difference Vegetation Index tend to occur at the extremes of the spectrum.

NDVI Sensor

NDVI Applications

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.

NDVI Sensor

Figure 4

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. If fact, as LAI increases, there’s an ever-increasing reflectance in the near infrared around 800 nm.

NDVI Limitations

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 an 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.

NDVI Sensor

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.

NDVI Senso

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.  

NDVI Sensor

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.

NDVI Sensor

MSAVI compares well with SAVI in terms of dynamic range and noise level (Qi et al. (1994) Rem. Sens. Env.)

Next week:  Learn about solutions for high LAI.

Get more information on applied environmental research in our

%d bloggers like this: