MONITORING SOIL DROUGHT

(A REVIEW)

Assoc.Prof. Vesselin Alexandrov

Sofia October, 2006

1

CONTENTS p.

1. INTRODUCTION 3

2. MEASUREMENT OF SOIL MOISTURE 3

2.1. Soil-water content 4

2.1.1. Direct measurements 4

2.1.2. Indirect methods 4

2.1.2.1. Radiological methods 4

2.1.2.1.1. Neutron attenuation 5

2.1.2.1.2. Gamma absorption 5

2.1.2.2. Soil-water dielectrics 6

2.1.2.2.1. Time-domain reflectometry 6

2.1.2.2.2. Microwave probe 7

2.1.3. Emerging technologies 7

2.1.3.1. Pulsed nuclear magnetic resonance (PNMR) 7

2.1.3.2. Remote sensing 8

2.2. Soil-water potential instrumentation 8

2.2.1. Tensiometers 9

2.2.2. Resistance blocks 9

2.2.3. Psychrometers 9

3. MONITORING AGRICULTURAL (SOIL) DROUGHT – STATE OF

THE ART 10

3.1. Monitoring soil drought abroad 10

3.1.1. Soil moisture, indices and indicators 12

3.1.2. Remote sensed indices (Satellite monitoring) 17

3.1.3. Soil-water balance based models 19

3.2. Monitoring soil drought in Bulgaria 22

3.2.1. Direct soil moisture measurements and drought indices 23

3.2.2. Remote sensed indices 27

3.2.3. Soil-water balance based models 28

3.3.4. Monitoring (soil) drought in Bulgaria through international cooperation -

Drought Management Centre for South-Eastern Europe 29

4. REFERENCES 31

2

MONITORING SOIL DROUGHT

1. INTRODUCTION

Drought is one of the most severe and extreme weather events affecting more people than

any other form of natural disaster (e.g. Wilhite, 2000). Given the consequences and pervasiveness of drought, it is important to assess drought severity. However, the precise quantification of drought is difficult as no universal drought estimation method (e.g. drought indices, hydrological or soil water balance models) can be defined through the complexity of the problem. The American Meteorological Society (1997) suggests that the time and space processes of supply and demand are the two basic processes that should be included in an objective definition of drought, and thus in the derivation of drought estimation methods. Common to all types of drought is a lack of precipitation (e.g. WMO, 1993). From a meteorological standpoint, drought is associated with dry spells of varying lengths and degrees of dryness. The basic measure of drought is inadequate precipitation for a particular activity (i.e. crop growth, irrigation supply, reservoir level).

In the scientific literature four types of droughts are commonly distinguished: meteorological or climatological, hydrological, agricultural, and socioeconomic (e.g. Rasmussen et al., 1993, Wilhite and Glantz, 1985). Meteorological drought results from a shortage of precipitation, while hydrological drought describes a deficiency in the volume of water supply (e.g. Wilhite, 2000). Agricultural drought relates to a shortage of available water for plant growth, and is assessed as insufficient soil moisture to replace evapotranspirative losses (e.g. WMO, 1975).

Agriculture is probably the most vulnerable economic sector to extreme weather events such as drought and many other economic sectors of our society depend on agroecosystems, which is a specific form of ecosystem adapted by humans for food production. As agriculture is an important economic factor in many countries, drought can have a number of economic and socio-economic consequences (e.g. CAgM, 1992, 1993; WMO, 1995, 2001) such as loss of income in agriculture and food industry, significant higher costs for water and production techniques (e.g. irrigation systems). Of all natural disasters, droughts globally occur most frequently, have the longest duration, cover the largest areas, and cause the greatest loss to agricultural production (e.g. WMO, 1997). Although drought is a natural component of climate, in arid and semi-arid climatic regions, it can also occur in areas that normally receive adequate precipitation (e.g. Li and Makarau, 1994). Most places in the world can be affected by agricultural droughts which reduce the availability of water required in agricultural production, but duration and intensity vary greatly from one climatic zone to another (e.g. Wilhite, 1993). Because of climate changes that could change climatic variability including precipitation pattern, extreme weather events such as drought are likely to occur more frequently in different spatial and time scales in future (e.g. IPCC, 2001b). 2. MEASUREMENT OF SOIL MOISTURE

One of the most significant factors influencing crop yield and watershed performance is the amount of water stored in the soil mantle. This soil moisture information is essential for determining irrigation schedules, for the evaluation of water and solute fluxes, and for partitioning of net solar radiation into latent and sensible heat components. Determination of soil moisture is of great concern to a number of agricultural disciplines. Soil moisture is the main indicator for the agricultural drought phenomenon assessment. To satisfy the widespread need of determining soil moisture status, a number of commercially-available instruments have been developed. The most commonly used and a few state-of-the-art instruments as well as techniques used instruments are discussed below.

3

2.1. Soil-water content 2.1.1. Direct measurements The simplest and most widely used method for measuring soil-water content is the gravimetric technique. Because this method is easy and based on direct measurements, it is the standard with which all other procedures are compared. Gravimetric soil moisture, is typically determined on a dry mass basis. In order to determine soil-water content, soil samples are removed from the field with the most convenient tool. Typical tools include shovels, spiral hand augers, bucket augers, as well as power-driven coring tubes. The soil samples are then placed in a leak-proof, tare-weighed container suitable for transporting to a laboratory and drying in an electrically heated oven. The samples and container are weighed in the laboratory both before and after drying, the difference being the mass of water originally in the sample. The drying procedure consists in placing the open container in an electric oven at 105°C until the mass stabilizes at a constant value. The time required varies from 16 to 24 hours. However, if the soil samples contain considerable amounts of organic matter, excessive oxidation may occur and some of the organic matter will be lost from the sample. Although the specific temperature at which excessive oxidation occurs is difficult to specify, lowering the oven temperature from 105 to 70°C seems to be sufficiently low to avoid significant loss of organic matter. Microwave oven drying for the determination of gravimetric water contents can also be used effectively (Gee and Dodson, 1981). In this method, soil water temperature is quickly raised to boiling point where it remains constant for a period of time due to the consumption of heat in vaporizing water. However, the temperature rapidly rises as soon as the energy absorbed by the soil water exceeds that consumed for vaporizing the water. Caution should be used with this method as temperatures can become so high that they can melt plastic containers if stones are present in the soil sample. Although rarely used, there are other methods for the direct measurement of soil-water content. However, they are limited to special purposes and emergencies. One of these methods involves placing the soil in a tared container with a perforated bottom and weighing it to determine the wet mass. The soil samples are irrigated with methanol, which will eventually displace the water. The methanol is then ignited, and the procedure is repeated at least one more time. The sample is then again weighed to determine its dry mass. The amount of methanol needed to displace the water depends on a number of factors, such as the size of the sample, its water content, and its texture. The latter method is very susceptible to error as volatile soil components may be lost.

2.1.2. Indirect methods The capacity of soil to retain water is, among other variables, a function of soil texture and structure. In removing a soil sample, the soil being evaluated will be disturbed, and its water-holding capacity altered. Indirect methods of measuring soil water are beneficial as they allow information to be collected at the same location for each observation without disturbing the soil-water system.

2.1.2.1. Radiological methods Two general radiological methods are widely used and available for measuring soil-water content. One is the neutron scatter method, which is based on the inter-action of high-energy (fast) neutrons and the nuclei of hydrogen atoms in the soil. The other method utilizes the attenuation of gamma rays as they pass through soil. Both instruments use portable equipment for taking

4

measurements at permanent observation sites and require careful calibration, preferably with the soil in which the equipment is to be used.

When using any radioactive emitting device, some precautions are necessary. All rules regarding radiation hazard laid down by the manufacturers and health authorities must be observed. When the guidelines and regulations are followed, there is no need to fear exposure to excessive radiation levels, regardless of the frequency of use. None the less, whatever the type of radioactive emitting device is used, the operator should wear some type of film badge that will enable the exposure levels to be evaluated and recorded on a monthly basis.

2.1.2.1.1. Neutron attenuation There are two types of neutron soil moisture detecting device a soil surface meter and a depth probe. In both devices, high-energy (fast) neutrons are emitted and are eventually slowed down upon their interaction with matter (resulting in neutron thermalization) (Visvalingam and Tandy, 1972). The hydrogen nuclei, having about the same mass as neutrons, are by far the most effective soil components in slowing down neutrons upon collision. As a result, the density of slow neutrons in the vicinity of the neutron probe is nearly proportional to the volumetric soil-water content. The slow or thermalized neutrons form a cloud around the neutron-emitting device where its density and size represent an equilibrium between the emission rate of fast neutrons and those thermalized. Within each neutron-emitting device is a thermalized neutron detector which determines the density of the thermalized neutron cloud. Unfortunately, the volume encompassed by the thermalized neutron cloud varies substantially with water content. For example, in wet soil, the radius of influence may be only 15 cm, while in dry soil, the radius may increase to 35 cm. Because the volume being measured varies with water content, this method lacks high resolution, making it impossible to localize water-content discontinuities. A particular problem occurs at the soil interface on account of the soil-air discontinuity. As a result, the neutron probe is not used in the top 18 cm of soil. However, the neutron surface meter is used exclusively for measuring water contents in the soil surface (0–30 cm). Unfortunately, where the soil surface is rough, precision falls off dramatically. A neutron depth probe comprises a radioactive source of high-energy neutrons, and a detector of slow thermalized neutrons, typically in a cylindrical form. The probe is attached by cable to the main electronics so that the probe can be lowered into a previously installed access tube. Although several arrangements of source-detector are possible, it is best to have a probe with a double detector and a central source. This arrangement allows for a more spherical zone of influence and leads to a more linear response with soil-water content. The neutron surface meter usually has a thermalized neutron detector laid horizontally on the soil surface with a fast neutron source behind it. The access tube should be seamless and thick enough (typically 1.25 mm) to be rigid, but not so rigid that the access tube itself is responsible for thermalizing neutrons. The access tube must be made of a non- corrosive material, such as stainless steel, aluminium, or some plastics, but polyvinylchloride should be avoided as it absorbs slow neutrons. The probe should be capable of being inserted into the tube without risk of jamming; usually a 4-cm diameter tube is sufficient. Care should be taken in installing the access tube to ensure that it is not bent. Additionally, no air voids should exist between the access tube and the soil matrix. Approximately 15 cm of the tube should extend beyond the soil surface as the box containing the electronics fits on top of the access tube. All access tubes should be fitted with a removable cap to keep rainwater from entering the tubes.

2.1.2. 1.2. Gamma absorption Whereas the neutron-attenuation method measures the volumetric water content in a large sphere, gamma absorption scans a 1-cm layer. Although it has a high degree of resolution, the small

5

soil volume evaluated will exhibit more spatial variation due to soil heterogeneities (Gardner and Calissendorff, 1967). The single-probe gamma device measures attenuation by reflection and is no longer widely used. However, the dual-probe gamma device which measures both soil density and water content is still a widely accepted instrument. Changes in gamma attenuation for a given mass absorption coefficient and absorber thickness can be related to changes in total density. As the attenuation of gamma rays is due to mass, it is not possible to determine water content unless the attenuation of gamma rays responding to dry soil density is known. Additionally, the dry density of the soil must remain unchanged with changing water content. If the dry soil density is known, then the soil-water content can be determined from the difference between the total and dry density values. Unlike neutron attenuation, gamma-ray attenuation enables a high spatial resolution. Vertical measurements at 2.5 cm can be made with excellent precision. It also has the advantage of making accurate measurements 2.5 cm below the air-surface interface. Additional caution should be taken with the use of gamma-emitting devices as they are potentially more dangerous than the neutron-emitting devices. The manufacturer will provide a shield which should be used at all times. The only time the probe leaves the shield is when it is lowered into the access tube.

2.1.2.2. Soil-water dielectrics Because of the dramatic difference in the dielectric constants of water and dry soil (approximately 80 and 3.5, respectively), theoretical and empirical relationships relating soil volumetric water content to the dielectric constant of the soil-water system have been proposed. This approach allows reliable, fast, non-destructive measurements of the volumetric water content, without the potential hazard associated with radioactive emitting devices. In addition, these methods lend themselves to being fully automated for large-scale data-acquisition programs. At present, two newly developed instruments which evaluate soil-water dielectrics are commercially available and are being used on an international scale. The first instrument utilizes time-domain reflectometry (TDR) technology, while the other measures the dielectric constant at a specific microwave frequency.

2.1.2.2.1. Time-domain reflectometry Time-domain reflectometry is a relatively new method which determines the dielectric constant of the soil by measuring the transmittal time of an electromagnetic pulse launched along a pair of parallel rods of known length embedded in the soil. As the sampling area is essentially a cylinder around the parallel probes, a large soil volume is examined. Theoretically, the dielectric constant is sensitive to soil surface area; however, time-domain reflectometry does not appear to be sensitive enough to require calibration for the range in surface areas typically found in soils. Generally, the parallel probes are separated by 5 cm and can vary in length from a few to over 30 cm. Additionally, the rods making the probe can be of any metallic substance; stainless steel is most frequently used. Although some care should be taken to ensure that the probes are parallel, slight deviations do not affect the resultant dielectric readings. In theory, the attenuated time-domain reflectometry signal should be able to measure both soil-water content and the salinity independently from a single reading; however, this work is still in its infancy. Additional work is being evaluated which allows this technique to be automated by examining the water content from a buried set of probes, each placed horizontally at a different depth. The probes are then linked through a multiplexing device attached to a field data logger.

6

2.1.2.2.2. Microwave probe The microwave dielectric probe utilizes an open-ended coaxial cable and a single reflectometer at the probe tip to measure amplitude and phase at a particular frequency (typically in the microwave region). Soil measurements are referenced to air, and typically calibrated with dielectric blocks and/or liquids of known dielectric properties. One advantage of using the liquids for calibration is that a perfect electrical contact between the probe tip and the material can be maintained (Jackson, 1990).

As a single, small probe tip is used, only a small volume of soil is ever evaluated. As a result, this method is excellent for laboratory or point measurements but is likely to be subject to spatial variability problems if used on a field scale. Additionally, the probe evaluates a small soil volume; therefore, soil contact is critical.

2.1.3. Emerging technologies Due to recent engineering advances, new methods are being developed which allow for the rapid measurement of soil moisture conditions. Two recent developments in soil moisture measurements are the use of pulsed nuclear magnetic resonance and microwave remote sensing.

2.1.3.1. Pulsed nuclear magnetic resonance (PNMR) Still in the research and development stage, the use of PNMR may have practical application in the near future (Paetzold et al., 1987). This measurement approach focuses on the interaction between hydrogen nuclear magnetic moments and a magnetic field. The sensor unit consists of an electromagnetic, radio-frequency coil, and a tuning capacitor. Essentially, this method allows for the instantaneous measurement of the volumetric water content in soil - independent of texture - organic matter content, and soil density. The magnetic moment of a nucleus which contains an odd number of protons/neutrons behaves like a spinning bar magnet. When placed in a static magnetic field, the magnetic moment precesses about an axis parallel to the applied magnetic field. If an oscillating magnetic field equal to the precession frequency of a hydrogen atom is applied at right angles to the static magnetic field, it will force the magnetic moments of hydrogen to precess in phase. The oscillating magnetic field is produced by the radio-frequency generator. The amount of energy adsorbed by the sample can, then, be measured, as well as the decay signal of the oscillating field. The analysis of the resultant adsorption and decay signals yields information concerning the spin-spin and spin-lattice relaxation times which, in turn, are used to calculate the amount of hydrogen in the sample. A tractor fitted with a prototype PNMR device has already been built and tested. This device could be used to determine soil-water content at the time of planting, or could be used to collect ground data for calibrating remote sensing instruments. Although the tractor PNMR system can accurately evaluate approximately 5 cm of surface soil moisture, precision drops off dramatically with increasing depth. The magnetic field must be homogeneous for PNMR techniques to work effectively, and obtaining a homogeneous magnetic field in undisturbed soil is the greatest limitation of this technique. Laboratory PNMR instruments can be purchased but they are generally too expensive for practical applications.

2.1.3.2. Remote sensing Measurements from space-borne instruments utilizing remote sensing techniques will be available in the near future for evaluating soil-water content, estimation of evapotranspiration rates,

7

and evaluation of plant stress on a watershed scale (Jackson and Schmugge, 1989). Although infrared and microwave energy levels have been widely studied, only the microwave region has the potential for obtaining direct quantitative soil moisture measurements from a space platform.

Microwave techniques can be separated into passive (radiometric) and active (radon) radiation. Passive microwave techniques focus on analysing the natural microwave emissions from the Earth’s surface, while active radiation refers to measuring the attenuation of a radar backscattering signal. Both approaches are based on the large differences that exist between the dielectric properties of liquid water and dry soil, and both are conducive to monitoring surface soil-water content over large areas of land.

2.2. Soil-water potential instrumentation To date, only instruments capable of measuring the matrix potential are sufficiently inexpensive and reliable to use in a field-scale monitoring programme. In each case, there are severe limitations to the range in water potential in which the instrument functions properly. Care must, therefore, be exercised if osmotic potentials are significant.

2.2.1. Tensiometers The most widely used and least expensive water-potential measuring device is the tensiometer. Tensiometers are simple, generally consisting of a porous ceramic cup and a plastic cylindrical tube connecting the porous cup to a recording device which seals the top of the cylinder. In view of their universal availability and low cost, a detailed description of their construction is unnecessary. The tensiometer establishes a quasi-equilibrium condition with the soil-water system. The porous ceramic cup acts as a membrane through which water flows, and as such, must remain saturated if it is to function properly. Consequently, all the pores in the ceramic cup and the cylindrical tube are initially filled with de-aerated water. Once in place, the tensiometer will be subject to negative soil-water potentials, causing water to move from the tensiometer into the surrounding soil matrix. The water movement from the tensiometer will create a negative potential or suction in the tensiometer cylinder which will register on the recording device. The recording device can be a pressure transducer (Marthaler, et al., 1983), a Bourdon-type vacuum gauge, or a simple U-tube filled with water and/or mercury. On the other hand, if the soil receives water, the soil-water potential may increase to where water moves from the soil back into the tensiometer, resulting in a less negative water potential reading. This exchange of water between the soil and the tensiometer, as well as the tensiometer’s exposure to negative potentials, will cause dissolved gases to be released by the solution, forming air bubbles. The formation of air bubbles will alter the pressure readings in the tensiometer cylinder and will result in faulty readings. Consequently, the cylinders occasionally need to be refilled and de-aired with a hand-held vacuum pump. Before installation, but after the tensiometer has been filled with water and degassed, the ceramic cup must remain wet. Wrapping the ceramic cup in wet rags or inserting it into a container of water will keep the cup wet during transport from the laboratory to the field. In the field, a hole of the appropriate size and depth is prepared. The hole should be large enough to be snug on all sides of the cylinder and long enough for the tensiometer to extend several centimetres above the soil surface. Since the ceramic cup must remain in contact with the soil, it is generally beneficial to prepare a thin slurry of mud from the excavated site and to pour it into the hole before inserting the tensiometer. Care should also be taken to ensure that the hole is backfilled properly, thus eliminating any depressions that may lead to ponded conditions adjacent to the tensiometer. The latter precaution will minimize any water movement down the cylinder walls, which would produce unrepresentative soil-water conditions. Tensiometers can measure only the matrix potential, because solutes can move freely through the porous cup. However, tensiometers can be purchased with additional features such as

8

electrodes, placed either inside the ceramic cup or just above the ceramic chamber, thus allowing electrical conductivity within the tensiometer to be determined simultaneously. Obviously, it may take some time for these tensiometers to equilibrate with the soil environment. Another limitation is that the tensiometer has a practical lower limit of about –80 kPa. Beyond –100 kPa, water will boil at ambient temperature, forming water vapour bubbles which will destroy the vacuum inside the tensiometer cylinder. The cylinder and recording portion of the tensiometer allow for appreciable changes in volume. Under drought conditions, appreciable amounts of water can move through the tensiometer to the soil. Thus, tensiometers can alter the very condition they were designed to measure. Typically, when the tensiometer acts as an irrigator, so much water is lost through the ceramic cups that a vacuum in the cylinder cannot be maintained, and the tensiometer gauge will be inoperative. Additional support of this process comes from excavated tensiometers which have accumulated large numbers of roots in the proximity of the ceramic cups. The tensiometer is also sensitive to temperature. Although only a small portion of the tensiometer is exposed to ambient conditions, the interception of solar radiation may induce thermal expansion of the tensiometer cylinder. Additionally, temperature gradients from the soil surface to the ceramic cup may result in thermal expansion or contraction of the cylinder, thus inducing false water-potential readings. To minimize these effects, the tensiometer cylinder should be constructed of non-conducting materials and readings should be taken at the same time every day, preferably in the early morning.

2.2.2. Resistance blocks Electrical resistance blocks, although insensitive to water potentials in the wet range, are excellent companions to the tensiometer. Electrical resistance blocks consist of electrodes encased in some type of porous material that will reach a quasi-equilibrium state with the soil. The most common block materials are gypsum, nylon fabric, and fibreglass (Perrier and Marsh, 1958).

Resistance blocks are relatively inexpensive and are good for field investigations. However, they do need to be calibrated before installation. This is generally accomplished by saturating the blocks in distilled water and then subjecting them to a predetermined pressure in a pressure-plate apparatus. After equilibration at a specific pressure, readings are taken, and the block is exposed to successively greater pressure potentials. This procedure should be repeated for at least five different pressures before installation. As the resistance-block calibration curves change with use, they need to be calibrated both before installation and after each investigation.

Unfortunately, resistance blocks are slow to equilibrate with soil, generating water-potential estimates that are more closely associated with the soil-drying curve. Consequently, this method is subject to errors where soil hysteresis may be an important factor. There is also a problem with shrinking and swelling soil which will break contact with the blocks. In addition, this approach determines water potential as a function of electrical resistance, and is sensitive to soil salinity. If saline conditions do exist, it must be remembered that added salts will decrease resistance, falsely indicating a wetter soil. The gypsum blocks are less sensitive to salts because the electrodes are consistently exposed to a saturated solution of calcium sulphate. However, gypsum blocks tend to deteriorate faster than fibreglass blocks.

When installing the resistance blocks it is best to dig a small trench for the lead wires before preparing the hole for the blocks. This will minimize water movement along the wires to the block, which could result in erroneous readings.

2.2.3. Psychrometers Thermocouple psychrometers do not measure the soil-water potential directly, but measure the vapour phase with which it is in equilibrium (Rawlins, 1972). As a result, psychrometers are quick to equilibrate with the soil environment. As with electrical resistance blocks, this method is

9

not sensitive to wet conditions but is well suited to a dry soil environment. It also lends itself to automated data acquisition. Psychrometers consist of a miniature thermocouple placed within a small chamber. The thermocouple is cooled off by the Peltier effect, condensing water on a wire junction. As water evaporates from the junction, its temperature decreases and a current is produced which is measured by a voltmeter. Consequently, these measurements are quick to respond to changes in soil-water potential, but are very sensitive to temperature and salinity. 3. MONITORING AGRICULTURAL (SOIL) DROUGHT – STATE OF THE ART 3.1. Monitoring soil drought abroad According to Wilhite and Svoboda (2000), the primary tasks for the monitoring processes are:

• To adopt an applicable definition for grading of drought. Many drought indices exist in consequence of drought definition multiplicity. Therefore, it is important to evaluate the performance of several drought indices in drought situation, for which comparative case studies can be accomplished. The evaluation include the index potential for using in early warning conditions and identifying the different effects of drought on hydrological features, ground water table, yield and state economy.

• regionalization for drought management. Climatic, land use, hydrological, topographical factors have to be considering to create homogenous areas for drought management.

• develop drought monitoring system. In the Central European region the meteorological observations belong mostly to the national meteorological services. This fact has a couple of positive influences: a. in the consequence of the general automatization process at the meteorological institutes, fully or partly automatic observation systems begun to be implemented. b. If the largest national meteorological network has one owner, than there is a good possibility to have the same instruments, same measuring time, observing rules, etc, resulting in a quite homogenous data. The same statement is valid for the hydrological observations as well, even if the meteorological and hydrological networks may belong to different institutes. • inventory data qantity and quality from current observation networks. Many indicators must be monitored beyond the meteorological and hydrological ones: agricultural, industrial, etc.

• determine data needs of primary users. The effectivity of the new systems are the highest, when primary users participate in the process of development.

• dissemination. The information delivery system has to take the information to the corresponding user on time. The deliverer can be transmitting system (broadcasting, TV), telecommunication system (phone, fax), but the most hopeful is probably the Internet. Many applications run on the Internet already, and the number of requests is growing continuously from the user side, too. During the past decade, there has been significant progress in drought monitoring strategies in the United States (Svoboda and Hayes, 2004). Most of these developments have improved the temporal and spatial resolution of monitoring drought conditions. Much of this development has stemmed from the Internet, which has provided near real time access to data and improved information sharing, as well as the development of satellite technology, Geographic Information Systems (GIS), and super computing capabilities. Each of these technological developments has helped improve our capability to monitor drought. Recently, decision support systems incorporating drought monitoring have begun to be developed.

10

One of the best examples of a new drought monitoring tool is the U.S. Drought Monitor map [http://drought.unl.edu/dm]. Produced on a weekly basis and operational since the summer of 1999, the Drought Monitor has become an accepted tool for drought assessment by the public, media and decision makers. Using an integrated approach, the product is not an index itself, but a composite indicator that uses several input parameters based on a ranking percentile in order to gauge the severity of drought in a given area. Based on four drought classes and an abnormally dry class, it utilizes information from snowpack, streamflow, soil moisture, precipitation, drought indices, satellite-derived vegetation indices, and several others indicators. In addition, an experimental effort has been underway between scientists in Canada, Mexico and the United States to produce a monthly North American Drought Monitor [http://www.ncdc.noaa.gov/oa/climate/monitoring/drought/nadm/index.html] map using the same methodology as the U.S. Drought Monitor, only tailored to the data available on a near real-time basis in Mexico and Canada. Canada has also a Drought Watch: http://www.agr.gc.ca/pfra/drought/index_e.htm (Fig. 1)

Fig. 1. On-farm surface water supply in Canada (source: http://www.agr.gc.ca/pfra/drought/index_e.htm)

There has been a recent emphasis on improving the usefulness of the U.S. Drought Monitor information for the users or decision makers, especially through decision support systems. States, for example, are capitalizing on the improved spatial and temporal resolution of information to develop state-based products critical to their specific needs. Oklahoma has developed a climate Mesonet that provides detailed climate information and a variety of products that can be used for

11

drought monitoring. Within the Oklahoma Mesonet system, the users can decide which statewide products are most valuable for their needs. The National Drought Mitigation Center is involved in a couple of projects that provide user-defined maps and data from a variety of drought indices [http://nadss.unl.edu] or satellite-derived vegetation conditions [http://gisdata.usgs.gov/website/Drought_Monitoring/viewer.asp]. At the gis.usgs.gov link, users can identify a variety of layers and resolutions for display, which can be used to aid in their decision making. In spite of these advancements, challenges in drought monitoring remain. There are still temporal and spatial data resolution issues as decision makers struggle to respond to drought conditions appropriately and in a timely fashion. In addition, the general lack of observed soil moisture and groundwater measurements impair our ability to determine the true severity and impacts of drought. Opportunities for cooperation and partnerships continue to be necessary in order to improve data networks and their quality. Other challenges include the need for a better understanding of the relationship between specific drought indicators and the severity of various impacts, and improved drought prediction. Perhaps the greatest challenge is in finding money to fund drought-related research.

It is recommended that European countries place greater emphasis on drought monitoring and the development of proactive mitigation plans to reduce the impacts of future drought episodes (Wilhite, 2004). The experiences of the United States and other countries can facilitate this process. Sharing information on drought monitoring, planning, mitigation, and policy issues will be beneficial for all parties and will likely stimulate greater progress on drought preparedness in the future. An example of the automatically interactive, Internet based system is the automatical irrigation advisory system, developed at the Hungarian Meteorological Service, by financial support of Ministry for Agriculture and Rural Development. The system calculates daily evapotranspiration (Penman-Monteith-formula) and uses this result and precipitation measurements to derive the accumulated climatological water sortage. From water shortage the water demand of different plants can be calculated. This system is the first operating automatically, interactive and freely accessible system on Internet in Hungary. This type of system can be a basis of a simple and real time drought monitoring system (Szalai, 2004). Quite simply, drought kills. Accurate monitoring of agricultural/soil droughts helps manage them, minimize losses attributed to them, and mitigate their extreme forms, which some countries face even today. There are many scientific papers and books across the world dialing with this topic. For example the book entitled “Monitoring and Predicting Agricultural Drought: A Global Study ” and edited by Vijendra et al. (2005) presents the basic concepts of agricultural drought, various remote sensing techniques used to monitor them, and efforts by international organizations to check them. The international contributors of the 34 papers cover the basic concepts of agricultural drought and its monitoring, prediction and analysis, techniques and examples of remote sensing such as passive and active microwave systems, and case studies from the Americas, Europe, Russia, the Near East, Africa, Asia, and Australia.

3.1.1. Soil moisture, indices and indicators The drought is supposed to be one of the most dangerous natural hazards nowadays even in the area of Central Europe. There is no drought-oriented monitoring system of soil moisture in Slovakia (Orfánus and Šútor, 2004). During the last decades, there were organized some partial and purpose-oriented monitoring systems of soil moisture. In the area of Rye Island (boundary area with Hungary) for example, there has been monitored soil water content in the unsaturated zone of soil as an indicator of impacts of Gabčíkovo Dam built on Danube River on surrounding ecosystems. The water regime of extremely heavy soils is monitored in the East-Slovakian Lowland. However, these systems cannot provide information relevant for drought evaluation. Regardless, an example

12

how can the monitored data of soil moisture be related to estimated drought-indicators (hydrolimits) was done (Orfánus and Šútor, 2004). The purpose of operational agro-meteorological program in Romania is monitoring the agro-meteorological phenomena of thermal, hydric and mechanic stress/risk in order to identify in due time the areas the most vulnerable to their occurrence and the dissemination of information towards the users aiming at making the right decision to prevent and mitigate the effects upon the crop efficiency. Soil moisture is the main indicator for the agricultural drought phenomenon assessment (Mateescu et al, 2004). In Romania the calculation stage for the soil moisture parameter are: a. Daily collection and processing of the primary meteorological data/fast flow from the ORACLE database/the AGRO-SYNOP Application b. Calculation of the potential evapotranspiration-the Penman-Monteith method/the CROPWAT model c. Decade (10 days) processing of the data regarding the soil moisture (mc/ha) coming from the weather stations integrating an agrometeorological program/winter wheat and maize/0-20, 0-50 and 0-100 cm depth d. The calculation of soil moisture reserve (mc/ha) (%/capacity of useful soil water -CAu) and the soil water deficit (mc/ha)/simple model of water balance computation The following keywords can be related to the above monitoring system: the way of space plotting - application based on remote sensing and GIS techniques; the execution interval – weekly. reasons/current monitoring, evolution in dynamics, prognostic estimate, real time warning, recommendations to prevent and mitigate of the agricultural drought; Important/mobile soil moisture measuring system. Figure 2 represents the spatial plotting of the soil moisture content at 0-20 cm depth.

Fig. 2. Spatial plotting of the soil moisture content in Romania at 0-20 cm depth / 22 September 2004 (source: Mateescu et al, 2004).

13

Because there is no single definition for drought, its onset and termination are difficult to determine. We can, however, identify various indicators of drought, and tracking these indicators provides us with a crucial means of monitoring drought. Determining which indicators to use poses more difficulties for planners: should they rely on data collected for specific parameters (such as streamflow and snowpack), or should they select one or more indices, which incorporate and weigh various types of data in various combinations? Equally important in choosing these indicators is a consideration of the type or types of water shortage facing the planner—an index or parameters well suited to agricultural concerns are of limited use to urban planners. Agricultural drought indices can be defined as rainfall pattern related to other climatic parameters and accompanied by soil water reduction, which obviously affects the soil water available for crops. To create agricultural drought indices, some of the ‘rainfall indices’ can be related, for example, to soil and crop type, crop status and climatological parameters such as air temperatures, air humidity and wind. A soil water deficit within the rooting zone can result in crop water stress, depending on the crop status and climatological factors affecting transpiration and evaporation (e.g. FAO, 1977). Agricultural drought can be therefore determined by a period of reduced plant growth with a prolonged and abnormal soil water deficiency. Sophisticated agrometeorological drought indices that are widely used in the USA are the Palmer Moisture Anomaly Index (Z index) and Crop Moisture Index. The Palmer indices are a soil moisture algorithm calibrated for relatively homogeneous regions. The Crop Moisture Index, updated weekly by the Climate Prediction Center, is a derivative of the Palmer Drought Index, designed to reflect quickly changing soil moisture conditions (Fig. 3).

14

Fig. 3. PDSI and CMI indices for the USA

(source: http://drought.unl.edu/dm)

These drought indices are currently monitored at a large spatial resolution (several thousand km2). Further, these drought indices are primarily based on precipitation deficits and are thus good indicators for monitoring large scale meteorological drought. However, agricultural drought depends on soil moisture and evapotranspiration deficits. Hence, two drought indices, the Evapotranspiration Deficit Index (ETDI) and Soil Moisture Deficit Index (SMDI), were developed by Narasimhan (2004). based on evapotranspiration and soil moisture deficits, respectively (Fig. 4). A Geographical Information System (GIS) based approach was used to simulate the hydrology using soil and land use properties at a much finer spatial resolution (16km2) than the existing drought indices.

15

Fig. 4. Spatial distribution of Soil Moisture Deficit Index (SMDI) in Texas, USA. a) 46th week of 1988 with standard deviation of 1.00 b) 24th week of 1990 with a standard deviation of 1.5.

(source: Narasimhan, 2004).

Тo investigate the usefulness of the SPI drought index for detecting and monitoring agricultural drought, the soil moisture component was investigated in Hungary (Szalai, 2004). Soil moisture is one of the most important limiting factors of plant production in Hungary.With soil moisture, the winter half of the year (October to March) was excluded. Generally there is very little evaporation from the soil, and soils are usually saturated at the end of winter regardless of the

16

amount of winter precipitation. Monthly soil moisture data were used for April through September at a depth of 0.5 m. The strongest relationships is with the 2-month SPI. There is a large scatter of soil moisture values around the relationship, indicating that non-meteorological factors (e.g., agronomic factors) also play a role in determining soil moisture.

3.1.2. Remote sensed indices (Satellite monitoring) The cost of natural disasters is increasing; their impact is invariably higher in developing countries and where people are concentrated. The risk of drought is a major concern in Romania. There are several technological trends that can serve to decrease the vulnerability to disasters. These include: better understanding of hazardous processes, improved analytical methods and communications. Since orbital sensing technologies have undergone unprecedented development, the use of multispectral satellite data in conjunction with traditional means is able to ensure the improvement of the classical determination methods of the agrometeorological parameters, greatly contributing to improve management of drought. The European and American orbital platforms (e.g. NOAA/AVHRR, SPOT/VEGETATION, ERS, LANDSAT, EOSTerra/ Aqua,QuikScat, ADEOS-2,etc.), equipped with different optical or radar sensors offer a high quality information with frequent repeat coverage (Fig. 5).

Fig. 5. Cover page of the proceeding of the conference on RS and geoinformation processing

Soil moisture is a key state variable of the global energy and water cycle. Accurate assessment of the soil moisture state and its spatial and temporal dynamics is therefore of crosscutting importance for many disciplines. Considerable technological progress during the last three decades leave no doubt that remote sensing from space afford the possibility of obtaining frequent global sampling of soil moisture over large fractions of the earth surface. Only recently the first global soil moisture data set based on remotely sensed data has been presented. The data set

17

was derived from measurements taken by the ERS scatterometer, a coarse resolution active microwave instrument. First validation studies, using soil moisture data from global climate models and field data, indicate the good overall performance of the derived information. Wagner (2004) gives an overview of our experiences gained with the ERS scatterometer as a basis to understand the potential of the Advanced Scatterometer ASCAT, which will be the operational continuation of the ERS scatterometers onboard EUMETSATs Polar System METOP Currently two soil moisture products are derived from ERS scatterometer data, surface soil moisture and profile soil moisture. The surface moisture is a relative measure of soil moisture in the first centimetre of the soil ranging between 0 and 100, representing the degree of saturation. Soil moisture can only be retrieved under snow-free conditions. Over dense tropical forest retrieval is not possible which affects about 6.5 % of the land surface area. Also in sand desert areas spurious effects are observed in the backscattered signal. These are related to azimuthal viewing effects which are currently not correctly treated in the retrieval and therefore masked. The other product is the profile soil moisture or Soil Water Index (SWI). The SWI is a relative measure of the soil moisture content of the 1 meter soil layer ranging between 0 and 100 (Fig. 6).

Fig.6. Mean monthly Soil Water Index derived from ERS scatterometer data for January, April, July, and October. Brown colour tones indicate dry conditions, blue colour tones indicate wet

conditions. (source: Wagner, 2004)

Scatterometers are designed to continuously record and transmit data to the ground stations. Due to the low data bit rate, the processing load is moderate and an operational application of the proposed technique is realistic. A prototype near real time operator for soil moisture retrieval has already been tested and implemented for the ERS scatterometer to monitor soil moisture conditions over Africa and Central Asia for the Food and Agriculture Organisation of the United Nations. The system has been operated for nearly six month and delivered soil moisture data on a weekly basis. Unfortunately, the system had to be discontinued after the ERS scatterometer mission has temporarily been stopped due to technical problems. Given the very similar technical characteristics of the ERS and the Advanced scatterometer it is possible to use the existing algorithms to deliver operational 25 km soil moisture products in

18

quasi-real time (2-3 hours after reception) with an expected accuracy of about 0.05 m3m-3. Every day to every other day the Advanced scatterometer could deliver an update of the status of the regional soil moisture conditions within a few hours after data reception. Technically near real time processing can be started immediately after the Advanced scatterometer is in an operational phase. The required datasets for retrieving soil moisture from ASCAT data can be set up using historic ERS scatterometer measurements (for reliable soil moisture retrieval knowledge of backscattering characteristics for each point of the earth surface is required which is extracted from long time series spanning several years of data). Given the similarities between the sensors, using the historic backscatter data should not constrain the retrieval of soil moisture from ASCAT data. An optimisation of the retrieval will be achieved by frequent reanalysis of the historic backscatter knowledgebase incorporating data from ASCAT. By satellite data drought and desertification indices are calculated in Romania and mapped with GIS technologies, e.g. Spectral Vegetation Indices (SVI), The Normalized Difference Vegetation Index (NDVI); The Modified Soil Vegetation Index (MSAVI); The soil heat flux (G); The soil moisture index (SMI), useful to characterize the actual drought or desertification status of the ground; The Water Deficit Index (DFI) (Stancalie, 2004) – Figure 7.

Fig.7. Soil Moisture Assessment over the Balkan peninsula using AMSR-E Data (source: Stancalie, 2004.)

3.1.3. Soil-water balance based models

The approach of analysing the effect of drought on agroecosystems using dynamic crop models has the advantage to include all relevant drought impact factors of the soil-crop-atmosphere system over short time periods. The simulation of soil and crop water balance is a crucial point in dynamic crop growth models and is achieved using various methods of different complexity (e.g.

19

Boogaard et al. 1998; Eitzinger et al., 2000a; Hoogenboom, 2000; Jamieson et al., 1998a; Penning de Vries et al., 1989; Sirotenko, 1983; Tsuji et al., 1994, 1998). Crop simulation models for a variety of crops and applications including soil water balance assessment have been described (e.g. Kunkel, 1990; Robinson and Hubbard, 1990; Steiner et al., 1991; Wilhite, 1993). Jamieson et al. (1998b), for example, have compared the models AFRCWHEAT2, CERES-Wheat, Sirius, SUCROS2 and SWHEAT with measurements of wheat grown under drought conditions. Such models have proved to be useful in analysing drought effects on agroecosystems in specific locations. They are often combined with remote sensing (e.g. van der Keur et al., 2001) and other data in geographic information systems to aid regional assessment of drought (e.g. White, 1999). A lysimeter experiment conducted on three soil types in a main agricultural production region of Austria in Marchfeld, was used to test the performance of the three widely used crop models, CERES, SWAP and WOFOST (Eitzinger et al., 2004). The soils included chernozem, sandy chernozem and fluvisol with a 2.0m profile depth. Daily measurements of the soil water content were taken using TDR probes (one per 0.3m of depth) in six replicates for each soil type. The analysis was carried out for winter wheat and spring barley grown on the site during seasons 2000 and 2001 and included a detailed comparison of the simulated and measured soil water contents as well as an analysis of seasonal soil water balances, root front velocities and an evaluation of the modeled crop yields. All three models simulated soil water content in the profile with similar results. Both CERES and SWAP mimicked the soil water content dynamics well in the top 0.3m of the soil. The study shows that the multiple layer approach models (SWAP or CERES) including more sophisticated estimation methods for root growth and soil water extraction should be preferred in comparable environments. For the monitoring of drought in agriculture in Italy, Marletto (2005) is using the transpiration deficit, defined as difference between maximum and actual transpirations, computed by means of a soil water balance model. To evaluate drought in agriculture the transpiration deficit becomes significant if it stays abnormally high for a long period.

The Grassland Information System and Monitoring (ISOP) is applied in order to assess also the soil drought in France (MeteoFrance, 2005). Some years ago the French Ministry for Agriculture and Fisheries, through its department for sample survey and statistical studies (SCEES), requested assessments of year-to-year variability of grassland production in France, with forage production regions as unit areas of assessment. The National Institute for Agronomic Research (INRA) and Meteo-France proposed a joint integrated system (i.e. ISOP) based on STICS, a multi-crop simulation model, developed by INRA. The ISOP objectives are to detect crisis situations – significantly low productions due to unfavourable soil-climate conditions – and to provide operationally early global assessments of forage production or its anomalies. The ISOP applies appropriate interpolation techniques of weather data, a soil map provided by INRA, characteristics of management practices such as: fertilization, grazing, silage, attribution of available soil water, etc. The ISOP outputs include both temporal and spatial monitoring - assessment of objective information on forage production (Fig. 8), e.g.: • ratio to mean (statistical) values • Alert maps – spatial overview for day D – once per month • time profiles for relevant drought-striken RFP, on request

The ISOP end users for now are: the Ministry for Agriculture and Fisheries through: the central department for sample survey and statistical studies; the bureau for agricultural disasters; the departmental services for Agriculture, Fisheries and Forestry.

20

Fig. 8. General outline of ISOP

21

Nain et al. (2002) concluded from the preceding result and discussion that agriculture drought monitoring with crop simulation model has edge over other conventional and popular drought monitoring approach such as Standardized Precipitation Index (SPI). Though there is flexibility in SPI analysis and indices can be generated ranging from month to years. Still SPI have limitations that it cannot account for water deficit caused by evapotranspiration, deep percolation, and run off. The Crop Simulation model based Drought Indices (CSDI) has additional advantage that it also accounts for the type and stage of crop on evapotranspiration losses. SPI do not consider the intensity factor and temporal distribution of rainfall within the base unit, for example month. The intensity factor is necessary to calculate the water losses through runoff, while temporal distribution is necessary to calculate the water stress during the crop growth cycle. For example a good amount of precipitation in very early stage of month may cause drought like condition in later phase (especially in sandy soils), while similar amount of precipitation well spread over the month will save the crop from any stress. SPI will show the similar values as it only consider the total amount or precipitation during the base unit such as month. This is reflected in the present results as Crop Simulation based Drought Indices (CSDI) could identify three crop yield deficient years out of four yield deficient years in the 13 years observed data set. 3.2. Monitoring soil drought in Bulgaria During the last years the system for soil monitoring in Bulgaria has been developed. It becomes an important source of information to be used in future policy-making. After the improvement of the technology in respect to soil monitoring, it is already possible to establish an updated database. By the means of this database, it would be possible more accurately to assess the trends as well as to determine the physical dimensions and measures related to soil protection, farms – served, soil areas – treated, tons of soils – rescued, etc. (Kolev, 2005) It is considered that the country has a well developed system for soil monitoring (MOEW, 2003). For example: Bulgaria manages a National Monitoring System (NMS) and an information system on components and factors of the environment, including soils. The quality control of the soils is among the NMS, which covers the whole territory and supports a database on national and regional scales. The Ministry of Environment and Water (MOEW) controls the NMS exploitation. The NMS is leading by the Executive Environmental Agency. The Regional Centers on Environment and Water (RCEW) in the country control the soil quality in respect to erosion, acidity and salinity. For example since 2003 the RCEW in Stara Zagora monitors soil salinity (NUTS4 strategy plan for 2007-2013, 2006) According to the EC requirements on forests protection against air pollution – documents 3528/86, 526/87, 1696/87, 1697/87, 2157/92, 1091/94, 307/97, 1390/97, 2995/89, 836/94, 690/95, 1390/97 – MOEW assesses the negative impact of air pollution in the country and measures for its reduction. Bulgaria was involved in an international programme for an extensive monitoring and executes soil measurements in 280 points across the country, following the adopted methodology (Republic of Bulgaria, 2004). The soil drought monitoring in the country is implemented by various organizations, for example research ones such as: - Institute of meliorations and mechanization – national center in the field of irrigation - Soil institute “Pushkarov” – with recent national projects, including investigations on soil degradation (e.g. erosion, acidity, salinity, etc.) and changes of the soil physical conditions under drought. Development of systems and levels of soil monitoring is also included in these projects. All this can be incorporated as elements of the foreseen monitoring of the system “soil-vegetation-atmosphere” in the context of the UNCCD. - Agricultural Institute in General Toshevo, Agricultural University in Plovdiv, etc.

22

It is necessary to point out that the above organizations implement soil drought monitoring (including measurement of soil moisture), mainly in limited regions or locations. It is considered that the main national system for direct measurements of soil moisture is based in the National Institute of Meteorology and Hydrology (NIMH).

3.2.1. Direct soil moisture measurements and drought indices NIMH at the Bulgarian Academy of Sciences is the main provider of the scientific research and operational activities in the field of meteorology, agrometeorology and hydrology in the country. The guiding principles of these activities, especially the operational ones, are the Technical Regulations of the World Meteorological Organization (WMO) of the UN. The Department of Meteorology is the main methodological coordinator and performer of operational service, applied research and theoretical scientific investigations at the National Institute of Meteorology and Hydrology in the field of meteorological stations' network, meteorological and agrometeorological databases, climatology and agrometeorology.

The Division of Agrometeorology: provides agrometeorological data, information, analyses, consultations, expert assessments, scenarios and predictions for the needs of governmental organizations and institutions; develops agrometeorological expert assessments, consultations, scenarios and predictions due to official users' requests from various sectors such as agriculture and environment, as well as to private companies and interested citizens; provides methodical and technical support as well as optimization of the agrometeorological network, measurements and observations in Bulgaria, where the NIMH has responsibilities; investigates the know-how in the field of methods for agrometeorological measurements, observations as well as data control and proposition of measures for their introduction at the NIMH agrometeorological activities; evaluates the current status of the agrometeorological network as well as suggestions of measures and strategies for its development, improvement and optimization; updates the methodological handbooks on agrometeorological measurements, observations as well as data processing, stations' files, metadata, etc., according to the respective normative documents and activities of the World Meteorological Organization; manages, maintains and controls the agrometeorological instruments for measurements and observations as well as related technical support; implements specific agrometeorological experiments, measurements and observations across the country; establishes, develops, manages and maintains the NIIMH-BAS agrometeorological database; maintains the agrometeorological (hard copies) archives. The NIMH agrometeorological network (Fig. 9) does not expand. Furthermore, some of the agrometeorological and phenological stations are closed down, because of lack of funds. The agrometeorological service has the following design: 1.Central agrometeorological service by the Institute of Meteorology and Hydrology offered to the central government departments and specially to the Ministry of Agriculture. 2.Agrometeorological services offered by the four branches of the Institute of Meteorology and Hydrology - branch Pleven - in Northeast part of Bulgaria, branch Varna - the eastern part of the country, branch Plovdiv - in the southern part of Central Bulgaria and the Kjustendil branch which is in the south - Western Part of the country. 3.Regional agrometeorological services by the 23 hydrometeorological observatories of regional administrative bodies, agricultural observations and farmers. 4. Kinds and types of services The agrometeorological services are accomplished in the following manners: -Daily agrometeorological evaluations of agricultural crop growth and opportunities for work in the field. -Weekly agrometeorological evaluation of the conditions for agricultural crop growth as well as of the conditions for carrying out hydrological activities. It is distributed by National radio broadcaster.

23

-Monthly bulletin about the meteorological conditions, soil status (including assessment of soil moisture) and the phenological stages of agrometeorological analysis of agrometeorological conditions by the means of tables, maps (Fig. 10) and graphics. They are published by the National Institute of Meteorology and Hydrology and are then distributed among the subscribers.

Fig. 9. Agrometeorological network (source: http://plovdiv.meteo.bg)

-Monthly agrometeorological forecasts about the conditions for the basic agricultural crop growth and the conditions for carrying out agrotechnical events. They are distributed by the TV and broadcaster on the Bulgarian National Radio (the Horizont program). They are printed in newspapers. -Utilisation of agrometeorological data in agricultural production and economical effectiveness.

The NIMH agrometeorological service provides measurements for soil moisture monitoring. The gravimetric method of measurement (described above) is applied (Fig. 11). These measurements have got discrete characteristic in time and space. Usually these measurements are carried on limited areas in three - five repetition for all 10 cm layers down to depth 100 cm every ten days.

24

Fig.10. Soil moisture (in %)under spring crop, 27 July and 27 August 2006. The red colour shows

the areas affected by soil drought at the end of August (source: www.meteo.bg)

25

Fig. 11. Stages of soil moisture determination

Various drought indices are also explored in Bulgaria (e.g. Koleva et al., 2004). Some of these indices are related to agricultural drought (e.g., Agroclilmatic atlas, 1982; Kercheva, M., 2004) – Figures 12 and 13.

Fig. 12. Index of wetness during the potential crop growing season

26

(source: Agroclilmatic atlas, 1982).

Fig. 13. Long-term variations of PDSI in Plovdiv, Sofia and Gorna Orjahovitza (source: Kercheva, M., 2004).

3.2.2. Remote sensed indices

The problem of continuous soil moisture monitoring is of great importance because of the increased necessity of creating Agrometeorological Automatic Management and Informational Systems (AAMIS) in agriculture (Kazandjiev and Georgiev, 1993). Every national agrometeorological service provides measurements for soil moisture monitoring. These measurements have got discrete characteristic in time and space. The results received from these measurements are distributed on area from 200 to 500 km2. This method of soil moisture presentation does not give any idea about space variability of mechanical, water - physical and chemical soil characteristics. Application of the soil moisture estimations by utilisation remote sensing data removes this defect, because data registrations are provided practically for all over the area. To trace the dynamics of Soil Brightness Indices in dependence of the soil moisture, the condition of damp was simulated by application of artificial damping of soils samples (Kazandjiev, 1993). After the measurement of spectral reflectance coefficients of the samples (for determination of the soil moisture) they were damped up to Full Soil Moisture Capacity. These measurements were repeated two or three times a day, several days on end and after that the whole cycle was repeated again. The result is - the increasing of soil moisture in the layer leading to the decreasing of the meanings of Spectral Reflectance Coefficients. Because the soil moisture determination by Soil Brightness Indices use is not enough precise its application is recommended in emergency or for express estimation. The priorities of the method are: quickness, objective assessment of the obtained information all over the given territory that is of great importance for operational practice. The Remote Sensing Applications Center (ReSAC) in Bulgaria was established with the support of the Food and Agriculture Organisation (FAO) of the UN in 1998. Main tasks of ReSAC are to develop Remote Sensing and GIS applications to agricultural and environmental management,

27

land cover/land use, soil and forest inventory, water resources, environmental hazards, urban planning, infrastructure, participation in regional and international projects and cooperation.

3.2.3. Soil-water balance based models A soil water balance model using for prediction and management of irrigation regime in the country was used (Krafti et al., 1980). A computer water balance sub-model MICCIM based on the model of van Keulen (1986) for agricultural production was also developed (Kolev, 1994). An assessment of drought impact on soil water balance and agricultural potential of Bulgaria using the Belgium simulation model WAVE was done (Popova et al., 1995). WAVE model is a series of computer sub-models simulating the behavior of water, solutes and soil-plant continuum. Crop water use and crop yield is simulated by SWAT sub-model that is capable to calculate water balance and mimic crop growth. SWAT sub-model integrates earlier packages SWATRER and SUCROS. The DSSAT developed by the IBSNAT (International Benchmark Sites Network for Agrotechnology Transfer) is using in Bulgaria to predict the performance of cereal (maize, winter wheat, barley, rice) and legume (dry bean, soybean and peanuts) crops (IBSNAT, 1993; Dukov et al., 1994; Slavov et al., 1996). The cereal, or CERES DSSAT family of crop models take into account the water balance hat simulates the daily evaporation, runoff, percolation and crop uptake under fully irrigated conditions and rainfed conditions. The water balance sub-model provides feedback that influences the development of crop growth processes. In the DSSAT legume GRO models as in the CERES models, a one dimensional soil-water model simulates water availability to the plants based on processes of runoff, percolation and redistribution of water. The above models are also sensitive to irrigation management options. In order to run the CERES and GRO water balance sub-models an information for soil moisture at the beginning of the simulation and soil profile description is needed. Usually, data of soil moisture measured every 10 days in the agrometeorological network at the National Institute of Meteorology and Hydrology are used. Since 1978 the Russian "Weather-Yield" dynamical crop model has been used in the Division of Agrometeorology at the National Institute of Meteorology and Hydrology to simulate growth, development and yield formation of different crops such as maize, winter wheat and sunflower. Modeling the water regime of the model consists of two stages: 1.) absorption of rainfall and water of irrigation applied using the gravitational method; 2.) simulation of soil moisture dynamics (Slavov et al., 1994). A similar model simulating the dynamics of soil moisture during the maize growing season between wilting point and fully saturated soil is presented in (Georgiev and Alexandrov, 1993). The blocks of evapotranspiration calculation using a modified Penman-Monteith equation, precipitation assimilation and soil moisture movement in different layers are concerned. Daily meteorological data of air temperature, air humidity deficit, sunshine duration and precipitation quantity are used for calculating soil moisture profile. The water stress as the ratio between actual and potential evapotranspiration (potential based on Thornthwaite approach, actual based on soil water dynamics calculation in ROIMPEL crop model) have been recently computed. Figure 14 shows the average (for the period 1961-1990) values of soil water stress in Bulgaria.

28

Fig. 14. Average annual soil water stress in Bulgaria average (1961-1990) 3.3.4 Monitoring (soil) drought in Bulgaria through international cooperation - Drought Management Centre for South-Eastern Europe

Drought is seen as a worsening threat. The regional implementation Annex to the United Nations Convention to Combat Desertification (UNCCD) for Central and Eastern Europe (CEE)-Annex V, by its Article 5 encourages regional cooperation between the affected countries of the region, with the aim to complement and increase the efficiency of national action programmes. Article 6 of Annex V stipulates that Parties of the region shall, individually or jointly promote the strengthening of scientific and technical cooperation networks, of monitoring indicators and of information systems at all levels. At a regional meeting for northern Mediterranean, central and eastern European and other affected country parties 23-26 July 2002, Geneva, Switzerland the following conclusions and recommendations were adopted: - Desertification assessment in some countries of the region is still at a preliminary stage. The

weakness of networking among scientific institutions, the absence of an operational early warning system on drought and soil moisture, limited exchange of data and work carried at varying geographic scales represent some of the difficulties that continue to challenge progress in monitoring drought and desertification

- Better coordination and sharing of relevant information and data at national, subregional and regional level to mitigate the adverse effects of drought should be developed.

29

- Early warning systems are under further development in Northern Mediterranean. Some countries have established national databases of information on monitoring desertification and drought. A Mediterranean database of information on monitoring desertification and drought is being set up. Continuity of such monitoring and assessment will depend on the availability of financial resources. In some countries, desertification impact indicators are being prepared and used. Social and economic indicators are being incorporated into desertification risk scenarios.

- Work on benchmarks and indicators is being carried out in line with the orientations of the European Commission. The future European soil monitoring system should be based on common legislation as statutory action and should include a set of parameters on desertification and land degradation.

The Regional Meeting for strengthening cooperation in the field of land resources management in Central and Eastern Europe held in Minsk, Belarus, in December 2003 discussed the issue of implementing the UNCCD at subregional level. The participants expressed the need for establishing a Balkan Drought Monitoring Center, and requested the UNCCD Secretariat to organize a workshop to discuss this issue. The UNCCD Secretariat organized a Technical workshop on drought preparedness in the Balkans within the context of the UNCCD in Poiana Brasov, Romania, 25-26 October 2004. The countries, participating at the Workshop agreed on the need to establish a subregional centre relating to drought management issues in the framework of the UNCCD. Further to the request of the Workshop, the UNCCD Secretariat in cooperation with WMO and at the invitation of the Ministry of Environment and Waters of Bulgaria organized in Sofia, Bulgaria from 26 to 28 April 2006 the second Technical workshop on the establishment of a subregional centre relating to drought in South-eastern Europe in the context of the UNCCD. The main objective of the second workshop was to reach an agreement on the main issues relating to the establishment and functioning of a Centre. The participants at the workshop in Sofia adopted a framework proposal on the establishment of a subregional “Drought Management Centre for South-Eastern Europe” (DMCSEE) within the context of the UNCCD and World Meteorological Organization (WMO) mandate. The aims and objectives of the Centre are:

(a) To serve as an operational centre for South-Eastern Europe for drought preparedness, monitoring and management; (b) To create and coordinate a subregional network of National Meteorological and Hydrological Services (NMHSs) and other relevant institutions; (c) To coordinate and provide the operational guidelines that will assist the NMHSs and other relevant institutions in the subregion to interpret and apply drought-related products; (d) To prepare drought monitoring and forecast products and make them available on near real-time basis to relevant institutions in participating countries; (e) To promote and strengthen the technical and scientific capacity for drought preparedness, monitoring and management in participating countries; (f) To facilitate the exchange of knowledge, experience and best practices on drought issues; (g) To enhance synergies among NMHSs, national UNCCD coordinating bodies, other international organizations and the scientific community on drought issues; (h) To enhance the implementation of the UNCCD in the context of drought preparedness, monitoring and management, in particular in working out a national drought strategy; (i) To collaborate actively with international research frameworks and programmes, to ensure the full participation of the South-Eastern European countries in such frameworks and programmes.

In Geneva, 26 September 2006, Slovenia was elected as the host country regarding to this Drought Management Drought Center for Southeastern Europe in the context of the United Nations Convention to Combat Desertification. It is expected that Bulgaria will efficiantly participate in the Centre’s activities especially in the field of drought monitoring where soil drought monitoring is an important part.

30

4. REFERENCES Agroclilmatic atlas of Bulgaria, 1982. Sofia (in Bulgarian)

American Meteorological Society (AMS), 1997. Meteorological drought - Policy statement. Bull. Amer. Meteor. Soc., 78, 847-849.

American Society of Agronomy, 1976: Methods of Soil Analyis, Part 1, Physical and Mineralogical Methods (A. Klute, ed.). Americal Society of Agronomy, Madison, Wisconsin, 188 pp.

Boogard, H., C. van Diepen, R. Roetter, J. Cabrera and H van Laar, 1998. WOFOST 7.1 User’s guide for the WOFOST 7.1 crop growth simulation model and WOFOST control Center 1.5. SC-DLO, IRRI, Wageningen, 144 pp.

Brunini, O. and Thurtell, G. W., 1982: An improved thermocouple hygrometer for in situ measure-ments of soil water potential. Soil Science Society of America Journal, 46, pp. 900–904.

Commission of Agricultural Meteorology (CAgM), 1975. Drought and Agriculture. Technical note No.138., WMO No.392, WMO, Geneva, 125 pp.

Commission of Agricultural Meteorology (CAgM), 1992. Monitoring Assessment and Combat of Drought and Desertification. WCP/CAgM, Report No. 47, WMO/TD 505, Geneva 112 p.

Commission of Agricultural Meteorology (CAgM), 1993. Report of the RA II Working Group on Agricultural Meteorology, Part IV: Drought and Desertification. Report No. 52, WMO/TD-No. 524, pp. 1-43.

Doorenbos, J. and Pruitt, W. O., 1976: Crop water requirements. Irrigation and Drainage, 24, Food and Agriculture Organization of the United Nations, Rome.

Drungil, C. E. C., Abt, K. and Gish, T. J., 1989: Soil moisture determination in gravelly soils with time domain reflectometry. Transactions of the American Society of Agricultural Engineers, 32, pp. 177–180.

Dukov, R. G. Georgiev and V. Alexandrov, 1994. Optimization of Soybean Irrigation Regime by a Computer Model. Proceedings of the 17th European Conference on Irrigation and Drainage, Yield Response to Water as Affected by Irrigation Schedule, Varna, Bulgaria, Vol. 2, pp. 231-237.

Eitzinger, J, V. Alexandrov, V. Cajic and H. Formayer, 2000a. A Site Specific Study on the Potential Range of Climatic Change Impact on Soil Water Balance and Crop Production under Consideration of Various Models. Proceedings of the 3rd European Conference on Applied Climatology "Tools for the environment and man of the year 2000" Pisa, Italy, 6 pp. (CD version).

Eitzinger J., Trnka M., Hosch J., Zalud Z., Dubrovsky M., 2004: Comparison of CERES, WOFOST and SWAP models in simulating soil water content during growing season under different soil conditions. Ecological Modelling 171, 223-246

FAO, 1977. Crop water requirements. Technical paper No. 24. Hoogenboom, G., 2000. Contribution of agrometeorology to the simulation of crop production and

its application. Agricultural and Forest Meteorology, 103: 137-157. Gardner, W. H. and Calissendorff, C., 1967: Gamma ray and neutron attenuation measurements of

soil bulk density and soil water content. Proceedings of the Symposium on Techniques in Soil Physics and Irrigation Studies, Istanbul. International Atomic Energy Agency, Vienna, pp. 101–113.

Gee, G. W. and Dodson, M. E., 1981: Soil water content by microwave drying: a routine procedure. Soil Science Society of America Journal, 45, pp. 1234–1237.

Georgiev, G. and V. Alexandrov, 1993. Simulation of soil moisture dynamics. IDOJARAS, Quarterly Journal of the Hungarian Meteorological Service, Vol. 97, No. 1, pp. 43-50.

31

Georgiev, G., V.Alexandrov, and N.Slavov, 1993. Initial Calibration and Validation of a Winter Wheat Simulation Model in Bulgaria. Bulgarian Journal of Meteorology and Hydrology, Vol. 4, No. 4, pp. 205-211.

International Benchmark Sites Network for Agrotechnology Transfer (IBSNAT), 1993. The IBSNAT Decade. Department of Agronomy and Soil Science, College of Tropical Agriculture and Human Resources, University of Hawaii, Honolulu, 178 pp.

Intergovernmental Panel on Climate Change (IPCC), 2001b. Climate Change 2001b: The Scientific Basis. Contribution of Working Group 1 to the Third Assessment Report of the Intergovernmental Panel on Climate Change (Houghton, J.T., Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell, and C.A. Johnson (eds.)). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 881pp.

Jackson, T. J., 1990: Laboratory evaluation of a field-portable dielectric/soil moisture probe. IEEE Transactions on Geoscience and Remote Sensing, 28, pp. 241–245.

Jackson, T. J. and Schmugge, T. J., 1989: Passive microwave remote sensing system for soil moisture: some supporting research. IEEE Transactions on Geoscience and Remote Sensing, 27, pp. 225–235.

Jamieson, P.D., M.A.Semenov, I.R. Brooking and G.S. Francis, 1998a. Sirius: a mechanistic model of wheat response to environmental variation. European Journal of Agronomy, 8: 161-179.

Jamieson, P.D., J.R.Porter, J.Goudrian, J.T.Ritchie, H.van Keulen and W.Stol, 1998b. A comparison of the models AFRCWHEAT2, CERES-Wheat, Sirius, SUCROS2 and SWHEAT with measurements from wheat grown under drought. Field Crops Research, 55: 23-44.

Jury, W. A. and Roth, K., 1990: Transfer Functions and Solute Movements Through Soil. Birkhauser Verlag AG, Basle, Switzerland, 228 pp.

Kazandjiev V. and G. Georgiev, 1993. Automated Agrometeorology System for Expert Assessment and information service, paper on the Conference “Environment-Earth sciences”, 29 March - 02 April 1993. (in Bulgarian)

Kazandjiev, V. 1993. Remote Sensing Soil Research in Visible and NIR Range Waves and Estimation of Soil Moisture, Bulgarian Journal of Meteorology and Hydrology, Vol.3, No.5.

Kercheva, M., 2004. Information base for modeling components of soil water balance and assessment of agroecological risks. PhD thesis (in Bulgarian)

Kolev, B., 1994. Using an Universal Model for Simulation of Evaporation from Water Surface and Potential Evapotranspiration, Proceedings of the Fifth Symposium on Physics and Agriculture, pp. 119-123 (In Bulgarian).

Kolev, K., 2006. The role of the scientific community in respect to sustainable soil management.UNDP

Krafti, G., A. Sadovski, I. Hristov and B. Kolev, 1980. Prediction and Management of Irrigation Regime. Journal of Plant Sciences, No.4, pp. 135-143 (In Bulgarian).

Kunkel, K.E., 1990. Operational soil moisture estimation for the Midwestern United States. Journal of Applied Meteorology, 29: 1158-1566.

Li, K. and A. Makarau (eds.), 1994. Drought and Desertification: Reports to the Eleventh Session of the Commission for Climatology. WMO/TD 605, Geneva, Switzerland, 68 p.

Marletto, V., 2005. Transpiration deficit: a tool for agricultural drought monitoring Marthaler, H. P., et al., 1983: A pressure transducer for field tensiometers. Soil Science Society of

America Journal, 47, pp. 624–627. Mateescu, E., Ion Poiana, V. Pescaru, Oana Oprea, Daniel Alexandru, 2004. Agrometeorological

monitoring system – decision-making instrument to prevent and mitigate the agricultural

32

drought in Romania. Proceedings of the Technical Workshop on Drought Preparedness in the Balkans within the context of UNCCD. UNCCD Secretariat, Bon, Germany, (CD)

MeteoFrance, 2005. Grassland Information System and Monitoring (ISOP) MOSW, 2003. National Self assessment of the capacities of Bulgaria for global environmental

management. UNCCD Nain, A.,S, Kurt-Christian Kersebaum and Wilfried Mirschel, 2002. Are Meteorological Parameters

Based Drought Indices Enough for Agricultural Drought Monitoring: A Comparative Study of Drought Monitoring with SPI and Crop Simulation Model

Narasimhan, B, 2004. Development of indices for agricultural drought monitoring using a spatially distributed hydrologic model. PhD thesis Texas university

Nielsen, D. R., et al., 1971: Soil Water. American Society of Agronomy, Madison, Winconsin, 175 pp.

NUTS4 strategy plan for 2007-2013, 2006 Orfánus, Tomáš and Július Šútor, 2004. Monitoring of soil moisture with respect to soil-drought

indicators and potential water storage in the unsaturated zone of soil, Proceedings of the workshop on drought monitoring in agriculture. Austria

Paetzold, R. F., Gish, T. J. and Jackson, T. J., 1987: NMR measurements of soil water content. Proceedings of the International Conference on the Measurement of Soil and Plant Water Status, Logan, Utah, 6–10 July 1978, 1, pp 255–260.

Penning de Vries, F.W.T, D.M. Jansen, H.F.M ten Berge and A.Bakema, 1989. Simulation of ecophysiological processes of growth in several annual crops. Pudoc, Wageningen, 271 pp.

Perrier, E. R. and Marsh, A. W., 1958: Performance characteristics of various electrical resistance units and gypsum materials. Soil Science, 86, pp. 140–147.

Popova, Z., M.Vancloster, J.Diels and J.Feyen, 1995. Assessment of Drought impact on Soil Water Balance and Agricultural Potential of Bulgaria using Simulation Model WAVE. In. Tsiourtis (ed.), 1995. Water Resources Management under Drought or Water Shortage Conditions, Balkema, pp. 87-92.

Rawlins, S. L., 1972: Theory of thermocouple psychrometers for measuring plant and soil water potentials. In: Brown, R. W. and van Haveren, B. P., Psychrometry in Water Relations Research. Utah Agricultural Experiment Station, Utah State University, Logan.

Rasmussen, E.M., Dickinson, R.E., Kutzbach, J.E., Cleaveland, M.K., 1993. Climatology. Handbook of Hydrology, D.R. Maidment, Ed., McGraw-Hill, 5.1-5.51.

Republic of Bulgaria, 2004, Negotoiatins on chapter 7 “Agriculture”. Intergovernmental conference on Bulgarian EU joining.

Robinson, J.M. and K.G.Hubbard, 1992. Soil water assessment model for several crops in the High Plains. Agronomy Journal, 82: 1141-1148.

Sirotenko, O., 1983. Development and applications of dynamic simulation models in agrometeorology. CAgM Reprt No.13, WMO, Geneva.

Slavov, N., G. Georgiev and V. Alexandrov, 1994. Simulation Model of Maize Productivity with Irrigation Account. Proceedings of the 17th European Conference on Irrigation and Drainage, Modification of Irrigation Schedule of Crops due to Scarcity of Water, Vol. 1, pp. 329-335.

Slavov, N., G. Georgiev and V. Alexandrov, 1996. Management of Cereal Crop Production using Mathematical Models of the Decision Support System for Agrotechnology Transfer DSSAT. Proceedings of the International Conference on Stable Agriculture during Transient Conditions of Market Economy, Plovdiv (In Bulgarian).

Stancalie, G., 2004. Possibilities offered by the remote sensing and GIS techniques for the monitoring and evaluation of agricultural drought in Romania. Proceedings of the Technical

33

Workshop on Drought Preparedness in the Balkans within the context of UNCCD. UNCCD Secretariat, Bon, Germany, (CD)

Steiner, J.L., A. Howell and A.D.Schneider, 1991. Lysimetric evaluation of daily potential evapotranspiration models for grain sorghum. Agronomy Journal, 83: 240-247.

Svoboda, Mark and Michael Hayes, 2004. Drought Assessment: The U.S. Drought Monitor and Beyond. Proceedings of the workshop on drought monitoring in agriculture. Austria

Szalai, S., 2004. Drought monitoring in the early warning system Topp, G. C., Davis, J. L. and Annan, A. P., 1982: Electromagnetic determination of soil water

contents using TDR: I. Applications to wetting fronts and steep gradients. Soil Science Society of America Journal, 46, pp. 672–678.

Tsuji, G., Uehara, G., Balas, S. (eds.), 1994. DSSAT version 3, Vol. 1-3, IBSNAT, University of Hawaii, Honolulu.

Tsuji, G., Hoogenboom, G., Thornton, P., 1998. Understanding Options for Agricultural Production. Kluwer

van der Keur, P, S. Hansen, K. Schelde and A. Thomsen, 2001. Modification of DAISY SVAT model for potential use of remotely sensed data. Agricultural and Forest Meteorology, 106: 215–231.

Vijendra K. Boken, Arthur P. Cracknell and Ronald L. Heathcote (eds.), 2005. Monitoring and Predicting Agricultural Drought: A Global Study” ISBN13: 9780195162349ISBN10: 019516234X hardback, 496 pages

Visvalingam, M. and Tandy, J. D., 1972: The neutron method for measuring soil moisture content: a review. Journal of Soil Science, 23, pp. 499–511.

Wagner, W., 2004. Operational Soil Moisture Monitoring with the EPS/Metop ASCAT for Hydrometeorological Applications, Proceedings of the workshop on drought monitoring in agriculture. Austria

White, D.H., 1999. Opportunities and roles for models to aid the management of agricultural and other natural resources within the variable climate of Austria. Environment International, 25(6/7): 907-912.

Wilhite, D.A., 1993. Drought Assessment, Management and Planning: Theory and Case Studies.Kluwer Academic Publishers, USA, 293 pp.

Wilhite, D.A., 2000. Drought as a natural hazard: Concepts and definitions. Drought: A Global Assessment. London, Routledge, UK.

Wilhite, D., 2004. Planning for Drought: Moving from Crisis to Risk Management Proceedings of the workshop on drought monitoring in agriculture. Austria

Wilhite, D.A. and M.H. Glantz., 1985. Understanding the drought phenomenon: The role of definitions. Water Int., 10, 111-120.

World Meteorological Organization (WMO), 1975. Drought and Agriculture. WMO Technical Note No.138. Report of the CagM WG on the Assessment of Drought. WMO, Geneva.

World Meteorological Organization (WMO), 1993. Drought and Desertification. Report on the Eleventh Session of the Commission for Climatology. Havana 1993, WMO/TD-No 605.

World Meteorological Organization (WMO), 1995. The Global Climate System Review. No. 819, 150 pp.

World Meteorological Organization (WMO), 1997. Climate, Drought and Desertification. Geneva, WMO No. 869, 12 p.

34