There are many national and a handful of international crop monitoring and forecasting systems. All maintain websites and publish bulletins, which summarise their analyses for decision makers at the national level and beyond. A collection of national bulletins is available from the WAMIS website of WMO, as well as from a number of national agrometeorological services. Note, incidentally, that the bulletins published in WAMIS, the Severe Weather Information Centre, many sources listed by INSAM (for instance this link) … not to forget the national on-line press, all constitute a first hand global source of information on local crop growing conditions. (First published: 20130210 / Last Updated: 20130508)
The international systems are publicly funded and operated by governments or intergovernmental organisations (EC-JRC MARS, China CropWatch, FAO-GIEWS, USAID-FEWNET, USDA-FAS) or run commercially (here is a small sample: MDA/CropCast daily, weekly, special; EARS; GeoSys).
All crop monitoring systems use some mix of ground and satellite data; their methods are mostly similar as far as sufficient information is available to outsiders to form an opinion… In general, they are best at home (because of better domestic ground data than their competitors). In general too, the monitoring products vary little among sources: all compare NDVI or some other indices with the previous season and/or the local average, current rainfall with averages, slightly modified FAO Water Satisfaction indices etc.
This post does not refer to the “exotic” national terms listed in the box below: it focuses on some very common non-standard terms, or common terms that are used in a non-standard way. We were all, in one way or another, trained “on-the-job”. We picked up the terms from different disciplines and never really took the trouble to verify their “exact” meaning. Then there is the imitation syndrome: we follow the use of the major players, and the major player(s) often just don’t bother about terminology or being “provincial”. There are also some nice idiosyncrasies… as for instance at the JRC, where data are “ingested” and “assimilated”, where “cumulated rainfall” stands for other peoples’ “accumulated rainfall” and “official statistics” simply denote agricultural “statistics”, with a slight derogatory connotation (for some good and some not-so-good reasons). As WBF reminded me the other day, there is not more than a couple of hundreds of experts worldwide… so that it should be no surprise that they developed their own dialect.
But enough introduction… Here comes the list, in random order!
Acreage: this word is used very frequently in crop monitoring bulletins and websites to mean “cropped area”. Readers should be aware that an acre is a “customary” and obsolete unit. According to Wikipedia, the acre is a unit of area used in the imperial and U.S. customary systems. An acre is about 40% of a hectare – slightly smaller than an American football field. The acre is no longer used except in a few countries, including the United States, Burma, and to some extent Canada. The acre is the American mu1. The International System of Units (SI) recommends the square m² and its multiples: the are (10 m x 10 m or 100 m²), and the hect–are. Hecto is the prefix2 for 100, so that a hectare is 10000 m². The meaning of “acreage” (meaning number of acres) is thus best conveyed by “hectarage” or more simply by “area” or “cropped area.”
Phenology is the science of describing development (the morphological3 changes that occur after a plant germinates: germination itself, two leaves stage, flower initiation…). The word is also used for insect development, for instance. The words “crop phase”, “crop stage”, “phenological phase”, “phenological stage”, “phenophase” and “phenometry” all refer to development. There are many common and scientific words based on “pheno-”, which means “appearance”, as in “phenomenon” (a thing that appears), phenotype (the way a plant or animal looks, as opposed to the genotype, its genetic base). Note that “harvest” is not a phenological phase, but a farm operation that takes place after the phenophase of maturity has been reached. Very often, environmental conditions permitting, harvest takes place more than one month after maturity
Growth describes the biomass accumulation during development. Unlike “development”, it is a quantitative measure. Growth is related to other terms such as primary production, net primary production (NPP) and harvest index (HI).
NDVI, the crop monitoring jack-of-all-trades is often used to determine phenology, but also crop “condition” or even soil moisture (Rojas et al 2011). It is very difficult to say if NDVI measures growth or development. The practice of NDVI accumulation assumes that NDVI measures growth. But NDVI is also somehow related to phenology. This is because as a plant grows, it not only accumulates biomass, but is also differentiates into organs (roots, leaves, flowers), so that there is a connection between NDVI and phenology too, but it is mostly an indirect one. There are, in fact, situations when phenology and growth are disconnected. This happens frequently in West Africa when photosensitive varieties of millet are grown, and planted too late, which can happen for a variety of reasons, for instance replanting because of the late onset of the rainy season. In that case, the “flowering signal4” is received by a plant when it is still very young, and it starts flowering when the biomass is low. Another example is when planting density is very low by design or by accident. The opposite situation, high biomass but slow development also occurs, for instance when low temperature negatively impacts development, a common occurrence in the East Africa, highlands. Finally, as is very visible in Zimbabwe, natural vegetation (miombo) can be photosensitive, so that vegetation starts greening ahead of rainfall…
Yield, production, productivity: “Yield” (Tons/Ha) measures the “economic biomass” (grain, fibre, sugar etc.) produced over a over a unit area during a cropping season, or a calendar year or even a marketing year. Production (Tons) expresses the amount of “economic biomass” in the same conditions but over some administrative unit, for instance a province or a country. “Yield” and “production” pose no particular problem, but “productivity” is often used incorrectly. Productivity is not an agronomic term, and therefore there is some confusion in its usage among agronomists. In ecology and economics, it is the “rate of production”, which is to say the production standardised by unit time and unit area (or production unit in economics). Based on this, productivity per unit time (year, or cropping season) and unit area (Ha) is the same thing as yield, or average annual yield if the time unit is a year5 or a marketing season. In most cases, however, productivity is used more specifically for “instantaneous productivity” (in the same way as we talk of “instantaneous velocity”, i.e. the velocity of a very short time interval, the one that is shown by a speedometer). Instantaneous productivity would be the “yield” of a leaf during a short time interval, for instance a second6. The concept is, therefore, used more frequently in crop physiology (e.g. photosynthesis studies), and in crop modelling than in agronomy. “Productivity” is also used to express the production potential, again mostly in the ecological literature, although sometimes agronomists apply it to crops7. Altogether, unless the precise meaning is defined, it is better to avoid using “productivity.”
Index8, indicator, variable, parameter, factor, element, descriptor, metric: the most basic term is “variable”. It designates the measure or the state something that varies. Air temperature is a variable, and so is leaf chlorophyll content and light absorption by a maize canopy. A variable can be quantitative or qualitative. An index is usually based on one or more variables combined in such a way as to be particularly meaningful for a given purpose. For instance, Body Mass Index (or BMI9) is often used in by nutritionists and experts in food security to identify underweight and obese persons, when BMI exceeds certain low or high thresholds. Another index is NDVI10, a function of satellite observed reflectances in the red and near infra-red channels, and frequently used to asses the presence and amount of living green biomass. “Indicator” is basically a statistical concept. It is a variable or index that is used as a proxy (indirect, often impressionistic measure) for some more “fuzzy” (i.e. mostly rather complex) variable such as “risk of war”, “level of development” etc. There is some confusion between “indicators” and “indices.” Both became very popular after the 1992 UN Conference on Environment and development. The current practice seems to be that “indicators” are more basic and close to variables in terms of data processing, while “indices” (such as UNDP’s Human Development Index, HDI11) are highly synthetic descriptors based on several variables and indicators. In practice, many terms that are in common use (e.g. NDVI) are named wrongly: NDVI is an indicator, not an index.
Then we have “parameter” which is very often confused with a “variable”. Wordings such as “environmental parameter” or “rainfall is a weather parameter” are usually incorrect. A parameter is best thought of as a tuning button which controls a model. Admittedly, the term “model” is not very precise, but this is rather convenient in the present case. If I have a model (conceptual, mental representation or computer simulation tool) that can be tuned to conform to reality, the tuning occurs by modifying the parameters. There can be some overlap between “parameters” and “variables”, but rainfall and NDVI are not normally “parameters” unless they are used as tuning buttons in some model.
Meteorologists and climatologists also use the word “element” for weather variables. This is, actually, rather interesting as they consider that weather and the climate (defined as average weather) of a location are made up by inter-related components, the weather (climate) elements. Rainfall and temperature, and solar radiation are thus weather “elements”19. They are also weather variables. Note, however, that clouds are listed among weather elements, although they would not be called “variables”. Cloudiness would be, though. “Weather element” is not an exact synomym of “weather variable” because, according to context, “element” can designate an atmospheric phenomenon and the measure that expresses its magnitude, i.e. “precipitation” (element) and “rain in mm” (variable), or “cloud” (element) and “cloudiness in oktas” (variable). However, in climatological and meteorological practice, and unlike “parameter” and “variable”, “element” and “variable” can mostly be used interchangeably.
Readers interested in “indicator science” (yes, there is some science involved!) will probably like the document by Bakkes et al, 1994.
A “factor” is a variable that affects the outcome of a process: sunshine is a dominant factor in the “determination of rice yields” (the process) in countries where water is not limiting. NDVI, on the other hand, is a mere descriptor of rice condition, but it does not influence crop condition: it is not a factor. The word “descriptor” is sometimes useful when none of the other words actually applies. For instance, we can describe a NDVI profile in terms of number of peaks, time when peaks occur, width of peaks and many more. The number of peaks, time and width can conveniently be referred to as “descriptors.” USDA and USDA/FAS are particularly fond of “metric”, which is basically a quantitative descriptor. It is actually a rather useful words when the trickier “variable” or “parameter” are to be avoided. I wonder if “acreage” can consistently be called a “metric”?
Crop condition, “good” crop, “poor crop”: Let us start by mentioning that the expression “crop condition” is normally not used in the plural. The vigour and good phytosanitary situation of maize can be expressed as “the good condition of maize”. I can say “crop conditions” if I refer to several crops, but even in this case the singular will usually do. Crop condition is frequently described as “good”, “poor”, “average” etc. The words themselves do not make any difficulty… but the perception of what a “good” crop actually is is open to interpretation and, interestingly, depends a lot on methodology. The analyst decides that a crop is “poor” when it performs poorly in comparison with a reference considered to be “normal” or “average” (these two terms are not equivalent, another pitfall!) The yardstick is not actual yield but some proxy, especially NDVI or the FAO water (requirements) satisfaction index (WSI or also WRSI; Senay 2004). Since there is no linear relation between yield (or “condition”) and the index, the comparison can only be impressionistic. In the case of WSI, the index is assumed to measure the degree of crop water satisfaction. Clearly, a value of 100% is expected for a crop to perform satisfactorily. If the reference value is 60% (as in many semi-arid areas) and the current season scores 70%, the crop is “better than the reference”, but it is nevertheless “poor”… because it is “normal” for the crop to be “poor” in that specific location. There is a permanent confusion between “normal” is “good”, “normal” is “average” and “average” is “normal”. It is, therefore, good practice to avoid the qualifiers “good” and “poor” altogether and to use some wording which means “above the yardstick”, “close to the yardstick” and “below the yardstick.” A final sentence to explain why “normal” and “average” are not equivalent: In the present context we do not refer “normal” using the statistical meaning of the word (normal=Gaussian), but rather to conditions that are expected most of the times. Especially in semi-arid areas, the frequency distribution of rainfall exhibits a positive skew and, as a result, the distribution of crop yields is skewed as well (Figure 2). A positive skew means that the frequency diagramme is stretched out to high values; it occurs when there are many low values of a variable (say: rainfall) and only occasionally high values. For instance, we may have recorded the following 10 rainfall values (mm): 0, 0, 0, 0, 100, 0, 0, 0, 0, 0. The average is 10 mm, but the expected or “normal” value is 0 mm. Rainfall will be below average 9 years out of 10.
Maize, corn: Originally, “corn” means “cereal seed, grain”; the word is not unlike “acre” in the sense that it refers to local use. “Corn” is maize in the US, wheat in England and oats in Scotland12, not to mention that anglophone West Africa has “Guinea corn” (sorghum), etc. Particularly in an international context, the best practice is to use the technical names, i.e. maize, wheat, sorghum, oats, barley etc. Grain trade statistics usually distinguish wheat, rice and “coarse grains”, i.e. all the other cereal crops, thus including maize, barley, rye, oats, teff etc. In French, “coarse grains” are known as “céréales secondaires” (secondary cereals), which is a nice example how confusing terminology can be, since maize has overtaken wheat and rice as the most produced cereal worldwide!
Winter crop, summer crop, autumn crop: although winter crops are in the field during winter, the important factor is not the season but the fact that they are exposed to a cold period; winter crops do not grow during winter, they just survive it. Many winter crops actually need the cold period to trigger flowering in spring. The process is known as vernalisation and without vernalisation, winter crops produce no seed. The phenomenon also exists in many temperate fruit trees and explains, for instance why there are plums but no cherries in Madagascar: winters are not cold enough13. In 1987, I bought some seeds of a Bhutanese turnip (locally known as “petso li”) on the market in Thimpu and planted them in my garden in Rome during the summer of 1987. They never germinated. After an unusually cold winter (1988-89), more than one year later, they started growing everywhere! Turnip belongs to the cabbage family (brassicaceae or crucifers) in which vernalisation is frequenyly required. In temperate countries, cabbage is often grown as a winter crop14.
Summer crops, (usually spring-planted) do not require vernalisation, or they like warm weather. They can also be planted in summer and harvested in autumn. Many tropical crops (such as the Ethiopian teff) grow well in temperate countries during summer, because heat supply is sufficient. Maize, cotton and many vegetables are typical summer crops because they cannot grow in cold temperatures.
The important point is that all crops that are not winter crops are summer crops, regardless whether they are planted in spring, summer or early autumn. If the intention is to stress when a crop is planted or harvested, then it is best to call them “summer-planted wheat” or “autumn-harvested sesame.”
Rate, gradient, trend: A rate is a dynamic measure, usually of something changing over time (e.g. crop growth rate) while a gradient refers to a variable (usually) changing over space (e.g. NDVI or temperature gradient, or crop price gradient from food surplus to food deficit areas). The tradition is to name the low end of the gradient first, i.e. a north-south obesity gradient normally means that people in the South are heavier than those in the North. “Trend”, like “rate” in most cases refers to time, as in the technology trend of Moroccan wheat yield has been 0.050 T/Ha for the last 12 years: on average, wheat yield increased 50 Kg by year, due to new varieties, improved fertiliser use, mechanization, better management etc. In the harder sciences and in crop modelling and crop physiology , “rate” usually refers to short time intervals (as in that sunny afternoon the rate of photosynthesis of the potted maize plant on my balcony reached 13.2 micrograms of carbon dioxide absorbed per second). Trend is mostly applied to more macroscopic and slower processes. As usual, the distinction between “rate” and “trend” goes through many shades and the concepts partially overlap.
Ratio, fraction, proportion, percentage: “Fraction”, “ratio” and “proportion” describe more static situations or, at least do not draw the attention to dynamics of change: The “sunshine fraction” is obtained by dividing the number of hours of bright sunshine in a day by the length of daytime hours, usually around 12 (depending on season and latitude). A “sunshine fraction” of 0.7 indicates that the sun shines 70% of the time, while we have only diffuse radiation during the remaining 30%. “Ratio” is basically the same as “fraction” but tradition confines it to more technical usage, for instance “sex ratio” to indicate the relative number baby boys and baby girls, for instance 107 girls to 100 boys at birth. Ratio is used most of the time when we compare only two categories, as in boys and girls. “Proportion” is used to indicate the relative share of several categories. For instance, the proportion of the areas under wheat, maize, soybean and fallow20 is 30%, 40%, 20% and 10% . The proportion can also be indicated as simple number, provided the sum covers all categories: For instance, the proportion of the areas under maize, wheat, soybean and fallow is 4:7:3:5.
Grain crops: grain crops are cereals, i.e. plants of the poaceae (grass) family. Common speech frequently uses “grain” for “seed” and since all 15 flowering plants produce seeds, it is not uncommon to come across sentences mentioning “grains such as maize and soybean…”. The expression is not correct. Soybean is no grain crop. The standard crop categories include cereals (or grain crops), oilcrops (including soybeans and groundnuts), roots and tubers (white potatoes, sweet potatoes, cassava, yams), vegetables (cabbage), pulses (the legumes that are not oilcrops, such as dry beans, vetches, haricot beans), tree nuts and a couple of others. I have some doubts about crops such as buckwheat. Botanically, it belongs to the polygonaceae family, but otherwise it’s cultivation and use is that of a cereal. I would be tempted to include it among the coarse grains16. Refer to FAO (2010) for the list of crop categories. The categories are basically categories of use. There are other possible categories according to other criteria, for instance “cash crops” and “industrial crops”. A “cash crop” is not a type of crop, it is a crop grown for cash. Maize can be a cash crop and durum wheat (the ingredient of Italian style “pasta”) is an industrial crop. As to staple crop (or simply “staple”), it is the crop that is most frequently consumed as food, the main source of energy. There are also many subtle and interesting differences among languages between “nutrition”, “feeding”, “eating”, “food”, “cooked food”, “cuisine”, “food consumption”, “food ingestion” etc.
Training, capacity building, education17 etc: Capacity building can be defined as the development of the ability of a group of people to perform functions, solve problems and set and achieve objectives in a sustainable manner. Note two things: the emphasis is on people and on sustainability. Equipment is not included. To “train” is to make proficient by instruction and practice, as in some art, profession or work. The focus is more on “operational capacity” than on sustainability. Education puts emphasis on the broad understanding of connections between issues and makes sure the people develop the necessary skills and intelligence. Contrary to “training”, “education”also assumes that the person is able to imagine new solutions. “Advocacy” and “awareness rising” belong to the realm of information, i.e. just making sure a decision maker or the public will keep in mind several option when confronted with the solution of a problem. Finally, “extension” (as in “agricultural extension”) and “outreach” ( an American word that’s getting popular) cover the dissemination of technical knowledge to practitioners, i.e. farmers.
Forecasting, prediction, projection: to the best of my knowledge and experience, the two first terms are almost equivalent and used interchangeably in agronomy and food security, as far as their meaning is
concerned. “Projection”, however, is used more often by economists and climate change experts than by agronomists, remote sensers etc. Table 1 shows that the frequency of occurrence of “prediction” and “forecasting” is equivalent, while “projection” is less frequent.
Monitoring, assessment, evaluation, estimation: “Monitoring” is the qualitative description of crop situation, including information on planting date, condition, comparison between years and places etc. Monitoring is done “for the record” and it is mostly based on combined quantitative and qualitative information. “Assessment” and the equivalent term “evaluation” imply that the situation is assigned a quantitative value or, at least, a rank, “valued”. “Estimation” is not unlike “evaluation” but contains the additional meaning of both “judgement” and “approximation.” “Estimation” is more qualitative than “evaluation”. Table 2 shows the relative frequency of the listed terms. “Monitoring” is the most neutral and also the most commonly used term. Documents translated from other languages (French, German, Chinese) often use the word “control” as a near synomym of the words above. This is best avoided as “control” has a much stronger meaning in English; the word means “exercise control, dominate18.” This acceptation my somehow apply to agricultural policies and fertiliser use, but only remotely to weather and crop condition.
Note 1 The mu is a traditional Chinese measure or cropland. With an accuracy better than 1%, there are 6 mu (?) in an acre. Both the mu and its length measure counterpart the li are well known by crosswords fans.
Note 2 See http://en.wikipedia.org/wiki/International_System_of_Units for other prefixes.
Note 3 Morphology refers to the “form”, the aspect, the look. In biology, morphology includes both the external look of organisms or organs and the internal shape and structure, which is covered by anatomy. Leaf anatomy refers to the internal structure of leaves, while leaf morphology, includes both external and internal features. There is also a word to describe only external features (eidonomy), but it is rarely used.
Note 4 In practice: a specific day length.
Note 5 If cropping intensity is high, it can be applied to the all the crops combined, although this is not very meaningful if different crops are grown in succession!
Note 6 The agronomists of the Wageningen tradition (who all are the scientific heirs of CT de Wit, the inventor of crop modelling) often use a an imaginary leaf of 1 Ha. See, for instance de Wit et al 1978, van Keulen & Wolf 1986 and Gommes 1998.
Note 7 For agronomists “potential yield” (and the resulting “potential production”) is the maximum possible yield that can be achieved in a given location. The absolute maximum depends on available solar radiation and crop type, and there are standard methods to estimate it (such as the one by Monteith; Monteith 1972, Section 3.1.1 in Gommes 1999; the approach is outlined here in this blog ). However, the absolute potential can rarely be achieved, as its realization also depends on the amount of available water (climatic potential or climatic productivity, “irrigated potential” or “irrigated productivity”) as well as on nutrient availability, including fertiliser. Needless to say, the variety of conditions gives rise to an equal variety of “productivities”.
Note 8 Plural is normally “indices” or “indexes” (US English)
Note 9 See http://en.wikipedia.org/wiki/Body_mass_index#Categories. BMI is the weight of a person (strictly: mass, in Kg) divided by her height (m) squared. BMI is thus expressed in Kg m-2.
Note 10 A dimensionless index; http://en.wikipedia.org/wiki/NDVI
Note 11 http://en.wikipedia.org/wiki/HDI_index
Note 12 http://dictionary.reference.com/browse/corn
Note 13 It also explains why one can find plums and bananas growing side by side in Bhutan or the highlands in southern Morocco. If someone is aware of cherries and bananas in the same field, please let me know.
Note 14 Another typical feature in many winter crops is “cold hardening” or “hardening”. This describes the process by which a winter crop becomes resistant to low temperatures. If cold sets in gradually, plants will develop a resistance to it. Too low temperature affecting a non-hardened crop may kill it. This is also whyc snow is important, because it protects crops from lethal freezing temperature.
Note 15 Almost all. Some do sometimes reproduce vegetatively.
Note 16 For coarse grains, see under “maize, corn”
Note 17 I am not an expert on education/training related terminology. This section is largely an adaptation and cut-and-paste job based on definitions from http://www.ask.com/question, http://dictionary.reference.com
Note 18 http://dictionary.reference.com/browse/control?s=t
Note 19 I have often used the expression “climate complex” or even “agri-environmental complex” to express the fact that weather variables are inter-related (correlated) and that they do not vary randomly. The same applies to weather and other environmental variables. For instance, there is a link between cloudiness and rainfall, between maximum temperature and solar radiation, minimum temperature and air moisture… See this and this link, or Sombroek and Gommes, 1996.
Note 20. About “fallow”. The word is used either as adjective (the land is fallow), noun (this fallow was ploughed by my neighbour but never seeded) or verb (I fallow this land every three years). “Fallowed land” is unusual wording, and “fallowed cropland” is even more unusual because cropland is mostly understood as “cropped land”. In fact, “fallow arable land” would be the most appropriate wording for “fallowed cropland.”
- Bakkes JA, van den Born GJ, Helder JC, Swart RJ, Hope CW, Parker JDE 1994 An Overview of Environmental Indicators: State of the art and perspectives. National Institute of Public Health and Environmental Protection Bilthoven, The Netherlands in Co-Operation with the University Of Cambridge United Kingdom. Study commissioned by UNEP 72 pp.
- de Wit CT, Goudriaan J, van Laar HH , Penning de Vries FWT, Rabbinge R, van Keulen H, Louwerse W, Sibma L, de Jonge C 1978 Simulation of assimilation, respiration and transpiration of crops. Pudoc, Wageningen, 141 pp.
- GEO 2011 Terms of Reference for the Data and Indicator Working Group. http://www.unep.org/GEO/pdfs/geo5/D&IWG_ToR_FINAL.pdf
- FAO 2010 Classification of crops. Downloadable from here: http://www.fao.org/fileadmin/templates/ess/documents/world_census_of_agriculture/appendix3_r7.pdf
- Gommes R 1999 Roving Seminar on crop-yield weather modelling; lecture notes and exercises. WMO. Geneva, 153 pp. ftp://ext-ftp.fao.org/SD/Reserved/Agromet/Documents/agro003a.pdf.
- Gommes R, El Hairech T, Rosillon D, Balaghi R, Kanamaru H 2009 Impact of Climate Change on agricultural yields in Morocco. World Bank-Morocco study on the impact of climate change on the agricultural sector. 105 pp. Downloadable from ftp://ext-ftp.fao.org/SD/Reserved/Agromet /WB_FAO_morocco_CC_yield_impact/report/
- Monteith JL 1972 Solar radiation and productivity in tropical ecosystem. Journal of Applied Ecology 9, 747-766.
- OECD 2001 Environmental indicators for Agriculture. Methods and results, Vol III. http://www.oecd.org/greengrowth/sustainableagriculture/40680869.pdf
- Rojas O, Rembold F, Delincé J, Léo O 2011 Using the NDVI as auxiliary data for rapid quality assessment of rainfall estimates in Africa, International Journal of Remote Sensing 32(12): 3249-3265
- Senay G 2004. Crop Water Requirement Satisfaction Index (WRSI), Model Description. http://iridl.ldeo.columbia.edu/documentation/usgs/adds/wrsi/WRSI_readme.pdf
- Sombroek, W.S. and R. Gommes, 1996. The climate change-agriculture conundrum, pp. 1-14 in Bazzaz, F., and W. Sombroek 1996 (Eds.), Global Climate Change and Agricultural Production, FAO and John Wiley, 345 pp. A web version is available from the website of FAO.
- van Keulen H, Wolf J (eds) 1986 Modelling of agricultural production: weather, soils and crops. Simulation monographs, Pudoc, Wageningen, 478 pp.