04. Module II (Biomass and yield calculation) - un-fao/gaezv5 GitHub Wiki
Introduction
The main purpose of Module II is the calculation of agro-climatical potential biomass and yield for a wide range of land utilization types (LUTs) under various input/management levels for rain-fed and irrigated conditions.
Module II consists of two steps:
- Calculation of crop biomass and yield potentials considering only prevailing radiation and temperature conditions, and
- Computation of yield losses due to water stress during the crop growth cycle. The estimation is based on rain-fed crop water balances for different levels of soil water holding capacity. Yield estimation for irrigation conditions assumes that no crop water deficits will occur during the crop growth cycle.
The activities and information flow of Module II are shown in the figure below.
Figure: Information flow of Module II
Land Utilization Types (LUTs)
Differences in crop types and production systems are empirically characterized by the concept of Land Utilization Types (LUTs). A LUT comprises technical specifications for crop production within a given socioeconomic setting. Attributes specific to a LUT include agronomic information, type of the main produce, water supply type, typical cultivation practices, utilization of produce, and associated crop residues and by-products. The GAEZ v5 framework distinguishes more than 300 crop/LUT and management combinations, which are separately assessed for rain-fed and irrigated conditions. These LUTs are grouped into more than 60 different food, feed, fiber, and bio-energy crops (see Appendix 4-1 with the list of crops). The calculated yield of each crop/LUT is affected by water source (rain-fed, irrigated) and by the assumed intensity of inputs and management. In GAEZ, three generic levels of input/management are defined: low, intermediate, and high input level.
Low level inputs
Under a low level of inputs (traditional management assumption), the farming system is largely subsistence based. Production relies on the use of traditional cultivars (if improved cultivars are used, they are treated in the same way as local cultivars), labor intensive techniques, and no application of plant nutrients, no use of chemicals for pest and disease control and minimum conservation measures. Fallows are required to maintain soil fertility.
Intermediate level inputs
Under an intermediate level of input (improved management assumption), the farming system is partly market oriented. Production for subsistence plus commercial sale is a management objective. Production is based on improved varieties, on manual labor with hand tools and/or animal traction and some mechanization, is medium labor intensive, applies some nutrients/fertilizer and chemical pest disease and weed control, and uses adequate fallows and some conservation measures.
High level inputs
Under a high level of input (advanced management assumption), the farming system is mainly market oriented. Commercial production is a management objective. Production is based on improved or high yielding varieties, is fully mechanized where possible with low labor intensity and uses optimum applications of nutrients and chemical pest, disease and weed control. In GAEZ, this variety in management and input levels is translated into yield differences by assigning different parameters for LUTs depending on the input/management level, e.g. such as harvest index and maximum leaf area index. LUTs are parameterized to reflect environmental and eco-physiological requirements for growth and development of different crop types. Numerical values of crop parameters used in the simulations differ depending on the assumed input/management level to which LUTs are subjected.
Thermal suitability screening of LUTs
As initial criteria to screen the suitability of grid-cells for the possible presence of individual LUTs, GAEZ tests the match of prevailing thermal conditions with the LUT’s temperature requirements. There are several steps applied to evaluate the extent to which thermal and relative humidity conditions during the crop cycle fit the respective LUT requirements: (i) Thermal (latitudinal) climatic conditions; (ii) permafrost conditions; (iii) length of temperature growing period (LGPt5); (iv) length of frost free period (LGPt10); (v) temperature sums (Tsumt); (vi) temperature profiles; (vii) vernalization conditions; (viii) diurnal temperature ranges (for selected tropical perennials), and (ix) relative humidity conditions (especially for selected tropical perennials). LUT specific requirements are individually matched with temperature regimes (and relative humidity) prevailing in individual grid-cells. Matching is tested for the full range of possible starting dates and resulting in optimum match, sub-optimum match and not suitable conditions. The “optimum and suboptimum match categories” are considered for further biomass and yield calculations.
Thermal climate
In Module II, the GAEZ procedures first check whether a LUT is considered suitable to be cultivated in the thermal regime prevailing in a grid-cell. The procedure assesses the compatibility of the LUT requirements in terms of overall temperature provision, climatic seasonality and seasonal day-length to enable the screening for respectively long-day, day neutral and short days crop LUTs. The screening of crop/LUTs about prevailing thermal climate results in a “yes/no” filter for further calculations to be performed for an LUT in individual grid-cells.
Permafrost
Areas classified as continuous or discontinuous permafrost are considered not suitable. Gelic soils, indicating permafrost, that occur outside the reference continuous and discontinuous permafrost zones are further appraised in the agro-edaphic suitability assessment.
Temperature growing period
The period during the year when temperatures are conducive to crop growth and development is represented by the temperature growing period, which is defined as the period during the year with mean daily temperature above 5⁰C, also referred to as LGPt5. Growth cycle lengths of crop/LUTs are matched with LGPt5. The result of the matching provides optimum match when the growth cycle can generously be accommodated within LGPt5. Otherwise the match is considered sub-optimum or not suitable. Hibernating crops survive low temperatures, e.g. during a winter season, by entering a dormancy period. GAEZ considers five hibernating crop species: winter wheat, winter barley, winter rye, winter rape, winter onion. These are the only annual crop/LUTs considered for cultivation at daily average temperatures <5⁰C. A dormancy period occurs when Ta ranges between 5⁰C and the crop-specific critical low temperature for cold-break. If the dormancy period is longer than 200 days, or daily average temperatures drop below critical thresholds (see below), the winter LUT is not suitable. For the effect of snow cover on lowering temperature thresholds for cold break, see details in Fischer et al., 2002.
Frost free period
Difference in sensitivity of crop/LUTs for early and late frost is accounted for through the matching of crop/LUT growth cycles with prevailing frost free periods. The frost free period is approximated by the period during the year when mean daily temperatures are above 10⁰C (LGPt10). Depending on the sensitivity of a specific crop/LUT the matching of growth cycle length with the available frost free period provides optimum match, sub-optimum match or not suitable conditions.
Accumulated temperature sum
Individual crop/LUT heat unit requirements are matched with temperature sums accumulated during the crop/LUT growth cycle duration (Tsumc). The Tsumc is defined as the sum of mean daily temperatures calculated from a base temperature of 0⁰C. The match of the crop LUT heat unit requirements with the prevailing temperature sum is optimum, when the requirements are falling within the specified optimum Tsumc range, sub-optimum when falling in Tsumc range conditions and not suitable when prevailing Tsumcs are too high or too low. Optimum and sub-optimum Tsumc ranges are presented for all crops/LUTs in the Appendix 4-6.
Temperature profile
The temperature profile requirements are crop/LUT-specific rules that specify conditions for crop cycle duration in terms of classes of mean daily temperatures. These classes in 5⁰C intervals are defined separately for days with increasing or decreasing temperature trends (Fischer, G., H. van Velthuizen, 2002). Updated temperature profile requirements data sets, for respectively optimum conditions and for sub-optimum conditions, have been specified for use in GAEZ v5 (see Appendix 4-3). Potential crop calendars of each LUT are tested for the match of crop/LUT temperature profile requirements and grid-cell temperature profiles, while considering growth cycle starting days within the length of the growing period for rain-fed conditions, and separately within the year for irrigated conditions. For all feasible crop calendars within the LGP (rain-fed) or within the year (irrigated) the temperature profile conditions are tested against optimum and sub-optimum crop temperature profile requirements and in each case an “optimum”, “sub-optimum” or “not suitable” match is established.
Vernalization
Some crops require a vernalization period (i.e., days with cold temperatures) for performing specific phenological development phases such as flowering. The production of flowers and grains, which directly influences crop yield, is dependent on the extent and intensity of exposure to periods with cold temperature. This cold temperature requirement is measured in vernalization days. In GAEZ, there are four hibernating crops that need to fulfill vernalization requirements in order to produce: winter wheat, winter barley, winter rye and winter rape. Details are provided in Appendix 4-4.
Diurnal temperature range and relative humidity conditions
For several tropical perennial crops such as coconut, cacao and oil palm, diurnal temperature ranges and/or relative humidity levels affect crop growth and yield. For these perennials requirements vis-à-vis optimum, sub-optimum and not suitable diurnal temperature ranges as well as permissible ranges of average, maximum or minimum relative humidity have been defined.
Combining temperature related constraints
When optimum conditions for crop cultivation are not met for all requirements, the degree of sub-optimality is derived by quantifying for each tested requirement a constraint factor fc1k, k=1, …, K, based on the distance of the calculated indicator from respectively the thresholds for ‘optimum’, ‘sub-optimum’ and ‘not suitable’ levels. At the threshold defining sub-optimum conditions it is assumed that crop growth and yield are reduced by 25%, whereas no reduction is applied for values exceeding the threshold for optimum conditions. When the calculated constraint value falls in between the optimum and sub-optimum thresholds, a constraint factor is assigned by linear interpolation. When the constraint value lies between sub-optimum and not suitable thresholds, then again a linear function is applied to calculate the constraint factor. Details of tested constraints for each crop are given in Appendix 4-3. For instance for sugar cane, optimal conditions require a temperature sum of at least 6750°C, for sub-optimal conditions of 6000°C, and sugar cane is regarded as not suitable when the temperature sum is less than 5700°C. For a calculated temperature sum of 6250°C in a specific grid cell, the prevailing conditions fall in between the specified optimal and sub-optimal thresholds. In this case the resulting constraint factor will have a value of 0.833 = 0.75 + 0.25 x ((6250–6000)/(6750–6000)), i.e., a reduction by 16.7%. For an even lower annual temperature sum of only 5800°C, the constraint factor for sugar cane would be 0.250 = 0.75 x (5800–5700)/(6000–5700), i.e., a reduction by 75%.
The “most limiting” evaluated related constraint factor is then used to reduce potential yields calculated in Module II. For this yield adjustment a reduction factor fc1 is calculated over all constraints:
$$ fc_1 = \min_{k=1,\ldots,K} { fc_{1k} } $$
which represents the minimum, i.e., the most severe of the individual temperature (and relative humidity) related reduction factors.
Biomass and yield calculation
In this section the calculation procedures of constraint-free biomass and yield (i.e., carbon accumulation driven mainly by prevailing radiation and temperature regimes in a grid-cell) are explained. The procedures used are based on the ecophysiological model developed at FAO by A.H. Kassam (Kassam, 1977; Kassam et al., 1983, 1991a).
The constraint-free crop yields calculated in the AEZ biomass model reflect yield potentials with regard to temperature and radiation regimes prevailing in the respective grid-cells. The model requires the following crop characteristics: (a) Average length of growth cycle (days from emergence to full maturity); (b) minimum and maximum length of growth cycle; (c) minimum temperature requirements for emergence; (d) maximum rate of photosynthesis, (e) respiration rates for leguminous and non-leguminous crops as a function of temperature during the growth cycle; (f) length of yield formation period; (g) leaf area index (LAI) at maximum growth rate; (h) harvest index; (i) crop adaptability group, and (j) sensitivity of crop growth cycle length to heat provision. Appendix 4-5 presents details of the calculation procedures and Appendix 4-6 provides the model parameters.
The results of the biomass and yield calculation depend on the timing of the crop growth cycle (crop calendar). Maximum biomass and yields are separately calculated for irrigated and rain-fed conditions, as follows:
Irrigation: For each day within the window of time when crop temperature and radiation requirements are met optimally or at least sub-optimally, the period resulting in the highest biomass and yield is selected to set the crop calendar of the respective crop/LUT for a particular grid-cell.
Rain-fed: Within the window of days with optimum or sub-optimum temperature conditions, and starting within the duration of the moisture growing period, the crop calendar resulting in the highest expected (water-limited) yield is selected to represent maximum biomass and yield for rain-fed conditions of the respective crop/LUT in a particular grid-cell.
In other words, for each crop type and grid-cell the starting and ending dates of the crop growth cycle are determined optimally to obtain best crop yields, separately for rain-fed and irrigated conditions. This procedure also entails adaptation of crop calendars (‘smart farmer’) in simulations with year-by-year historical weather conditions, or under climate distortions applied in accordance with various climate change scenarios.
Net biomass and yields for most LUTs in GAEZ are expressed in kilograms of dry matter (DM) per hectare with the exception of some oil crops (yield expressed as oil), sugar crops (yield expressed as sugar) and cotton (yield expressed as lint). In the case of forage crops and grasses the yields are expressed as 10 kg DM per hectare. This includes Alfalfa, Miscanthus, Switchgrass, Reed canary grass, napier grass, Grass legumes and Grasses.
Water limited biomass production and yields
Under rain-fed conditions, water stress may occur during different stages of the crop development reducing biomass production and the yields achieved. In GAEZ v5, water requirements of each LUT are calculated daily and are considered in the calculation of LUT-specific water balance and actual evapotranspiration in a grid-cell. A water-stress yield-reduction factor (fc2) is calculated and applied to the net biomass (Bn) and calculated potential yield (Yp).
Crop water requirement
The total water requirement of a crop without any water stress is assumed to be the crop-specific potential evapotranspiration (ETm). ETm is calculated in proportion to reference potential evapotranspiration (ETo), as in Module I, multiplied by crop and crop-stage specific parameters ‘Kc’. The values of Kc for different stages of crop development (figure below) are given as input parameters (Allen et al., 1998).
Figure: Schematic representation of Kc values for different crop development stages
- D1: initial phase: from planting to 10% ground cover (from planting/germination/emergence to establishment);
- D2a: early crop development stage;
- D2b: late crop development stage;
- D3a: early mid-season stage (flowering);
- D3b: late mid-season stage (reproductive stage), and
- D4: late season stage (start maturation to full maturity).
Input parameters define the relative length of each crop stage as a percentage of total cycle length (GC). Further three input parameters define the crop coefficients for water requirements (Kc, fractional) as follows: Kc1 being the reference crop coefficient for the initial stage, Kc3 being the reference crop coefficient for the mid development stage, and Kc5 being the reference Kc crop coefficient applying at the end of late season stage. In addition, an average Kc parameter representative for the entire growth cycle (KcT) can be specified to calculate an overall crop water requirement. Procedures described in Allen et al. (1998) are applied in each grid cell to adapt reference Kc coefficients to local conditions in terms of rainfall distribution, average relative humidity and wind during each crop development stage.
The value of Kc for a particular day j of the year is defined by:
$$ Kc_j = \begin{cases} Kc_1, & j \in D_1 \\[5pt] Kc_1 + \dfrac{(j - d_1)(Kc_2 - Kc_1)}{d_2}, & j \in D_{2a} + D_{2b} \\[5pt] Kc_2, & j \in D_3a + D_3b \\[5pt] Kc_2 + \dfrac{(j - (d_1 + d_2 + d_3))(Kc_3 - Kc_2)}{d_4}, & j \in D_4 \end{cases} $$
where d1, d2, d3 and d4 denote the length (number of days) of the respective major crop development stages. Parameters used for simulations in GAEZ v5 are listed in Appendix 4-2.
Yield reduction due to water deficits
Yield reduction in response to water deficits is calculated as a function of the relationship between actual crop evapotranspiration (∑ETa, mm/day) and maximum crop evapotranspiration (∑ETm, mm/day), both accumulated within each crop development stage. Daily ETa is calculated from the water balance as described also in Module I, with the difference of being LUT-specific in Module II. Also, in Module II, the value of the soil water depletion fraction (p) varies depending on the crop and the level of potential evapotranspiration ETo.
The sensitivity of each crop to water stress is expressed by the value of the water stress coefficient (Ky, fractional), a LUT-specific parameter which changes with crop development stage (Doorenbos and Kassam, 1979; Doorenbos and Pruitt, 1977) There are Ky values specified for each crop development stage as follows:
- Ky1: yield response factor initial phase;
- Ky2a: yield response factor early crop development stage;
- Ky2b: yield response factor late crop development stage;
- Ky3a: yield response factor early mid-season stage (flowering);
- Ky3b: yield response factor late mid-season stage (reproductive stage);
- Ky4: yield response factor late season stage (start maturation to full maturity), and
- KyT: yield response factor total growth cycle.
GAEZ uses both the crop stage specific coefficients and estimated water deficits and the overall value of KyT to calculate a water-stress yield reduction factor (fc2).
$$ fc_2^T = 1 - Ky^T \times \left(1 - \frac{\sum_{j=1}^{TCL} ETa_j}{\sum_{j=1}^{TCL} ETm_j} \right) $$
$$ TETa_j = \sum_{k \in D_j} ETa_k, \quad TETm_j = \sum_{k \in D_j} ETm_k, \quad j = 1, \dots, 4 $$
$$ fc_2^{CS} = \min_{j=1, \dots, 4} \left( 1 - ky_j \left(1 - \frac{TETa_j}{TETm_j} \right) \right) $$
and
$$ fc_2 = \min(fc_2^{CS}, fc_2^T) $$
where TETaj and TETmj are respectively total actual evapotranspiration and total potential evapotranspiration for days during crop stage Dj. Factor fc2 is the minimum of factor fcT2, representing the effect of overall water deficit, and the factor fccs2 represents the most severe impact of crop-stage specific water stress. Water limited yield (Yw) is then calculated as potential yield (Yp) multiplied by the water-stress reduction factor fc2:
$$ Y_w = Y_p \times fc_2 $$
Adjustment of LAI and HI for perennial crops
Perennial crops have limited opportunity to express their genetic potential to expand canopy (i.e., develop leaf area index, LAI) and to complete formation of yield components (e.g. fill grains) if the period for growth, here termed effective growth cycle length (CYCeff, days) is too short in a given location. These two aspects of perennial crops are captured in GAEZ by adjustment factors for LAI (fpLAI) and for harvest index (fpHI), which are adjusted if the length of the effective growth cycle falls below a crop-specific critical threshold. Under rain-fed conditions the CYCeff is limited by the number of growing period days in a year. Under irrigation conditions temperature growing periods LGPt10 or LGPt5 are used depending on the crop:
Under rain-fed conditions:
$$ CYC_{\text{eff}} = (LGP,\ CYL_{\text{max}}) $$
Under irrigated conditions:
$$ CYC_{\mathrm{eff}} = \left(LGP_{t=t_0},\ CYL_{\mathrm{max}}\right) $$
where t0 is set to 5°C or 10°C depending on the crop. Then the adjustment factors for HI and LAI are computed as follows:
$$ fP_{HI} = \frac{CYC_{eff} - \alpha_{HI}}{\beta_{HI}} $$
$$ fP_{LAI} = \frac{CYC_{eff} - \alpha_{LAI}}{\beta_{LAI}} $$
with CYCeff as defined above.
For example, when simulating rain-fed sugar cane, the maximum annual growth cycle is set in the parameter file to a value of CYCmax = 330 days. In a location with only 240 growing period days the effective growth cycle would be set to CYCeff = 240 days and using the parameter values in the table below would result in fPHI = (240–120)/180 = 0.667 and fPLAI = (240–70)/200 = 0.850.
Table: Parameterization used to correct harvest index (HI) and leaf area index (LAI) for sub-optimum length of the effective growth cycle (CYCeff)
| Crop | fPLAI | fPHI |
|---|---|---|
| αLAI. | βLAI. | αHi |
| Cassava | 0 | 240 |
| Sugar cane | 70 | 200 |
| Banana | 0 | 300 |
| Oil palm | 0 | 330 |
| Yellow yam | 0 | 270 |
| Cocoyam | 0 | 270 |
| Citrus | 0 | 180 |
| Cocoa | 0 | 270 |
| Tea | 0 | 270 |
| Coffee (arabica) | 0 | 240 |
| Coffee (robusta) | 0 | 270 |
| Coconut | 0 | 240 |
| Tea | 0 | 300 |
| Alfalfa | 0 | 180 |
| Miscanthus | 0 | 210 |
| Switchgrass | 0 | 180 |
| Reed canary grass | 0 | 150 |
| Napier grass | 0 | 180 |
| Para Rubber | 0 | 330 |
For each of the two variables two separate parameters are used to calculate the adjustment factors for HI and LAI of perennials. These parameters relate to critical values of the length of the effective growth cycle, below which a yield reducing adjustment is applied or no yield is obtained. Also, note that a perennial crop may be considered not suitable for levels of CYCeff well above αHi or αLAI. The effective growth cycle accounts for the days in the year when perennial crops are effectively growing. Under rain-fed conditions CYCeff cannot exceed the number of growing period days determined for a grid cell and therefore the period of vigorous growth may be limited by temperature, rainfall and soil moisture availability. It also excludes any period of dormancy or resting of perennial crops. The parameterization for perennial crops used in GAEZ v5 is given in Table 4-1. The adjusted harvest index HIadj and leaf area index LAIadj for perennial crops are then calculated as:
$$ HI_{adj} = HI_{max} \times fP_{HI} $$
$$ LAI_{adj} = LAI_{max} \times fP_{LAI} $$
where HImax and LAImax denote the reference parameter values specified for growing conditions where the full growth cycle is possible.
Crop calendar
The crop calendar (i.e., sowing and harvesting dates) for a given LUT and grid-cell is determined by identifying within the permissible window of time the sowing date that leads to the highest attainable yield. GAEZ tests all possible LUTs/sowing dates within each grid-cell, separately for rain-fed and irrigated conditions. For each LUT, the total crop cycle expected for the ‘average climate’ (40-year time period from 1981–2000 or 2001–2020) is given in days as an input parameter. For the average base climate, an accumulated temperature sum (TS5) is calculated for each crop LUT. This crop-specific value of TS5 is assumed to represent for a location the specific crop cycle requirement of the LUT. When simulating individual years, the crop cycle is adjusted until the specific TS5 is reached, as calculated for average climate conditions, e.g. is shortened in years warmer than normal.
Figure: Optimum crop calendar (FAO and IIASA 2012)
For each grid cell and LUT the algorithm determines the highest attainable yield, which then defines the respective outcome for that location.
For rain-fed production GAEZ calculates potential crop yields by shifting computed calendars within the permissible part of the LGP and selects the start date of the crop when yield is the highest. This optimum crop calendar for rain-fed conditions is reflecting, for a crop/LUT, the optimum combination of radiation regime, temperature regime and soil moisture availability, as shown in the figure above. For irrigated production GAEZ tests all possibilities of crop yield performance in LGPt5 (i.e., in the period during the year when Ta >5⁰C) and selects the period with highest attainable yields, thus driven mainly by radiation and temperature regime. The calendar search in GAEZ is flexible and alternatively could also use a selection criterion which would account for the trade-off between additional irrigation water use and additional yield generated.
Description of Module II outputs
The output of Module II records for each grid-cell and LUT the relevant results of the biomass calculation, including potential yields, yield-reducing factors, accumulated temperatures, actual crop evapotranspiration, water deficits and crop calendar information. The process generates thousands of maps which are named using a 4-character crop acronym and a 3-character map type acronym. The tables below list the types of information mapped by Modules II and III, and the corresponding four-character crop acronyms.
Table: Mapped output produced by Module II/III analysis
| Type | Description | Unit |
|---|---|---|
| cbd | LUT crop cycle starting date | Day-of-year |
| cyl | Cycle length of selected crop/LUT | Days |
| eta | Actual crop evapotranspiration from precipitation (i.e., excluding irrigation) | mm |
| fc0 | Combined temperature, soil moisture and agro-climatic constraint factor | Scalar |
| fc1 | Yield reduction factor due to temperature profile evaluation | Scalar |
| fc2 | Yield reduction factor due to soil moisture deficits during LUT growth cycle | Scalar |
| fc3** | Yield reduction factor due to agro-climatic constraints evaluation | Scalar |
| idx | Sequence number of LUT selected to define grid cell crop results | Class |
| Tsc | Accumulated temperature during LUT crop cycle | ∑°C |
| wde | LUT water deficit/net irrigation requirement during crop cycle | mm |
| yld | Agro-climatic potential yield | Kg/ha* |
Table: Crop name acronyms used in GAEZ v5 file names of Module II/III mapped outputs
| Acronym | Crop name | Acronym | Crop name |
|---|---|---|---|
| alfa | Alfalfa | bana | Banana |
| barl | Barley (the better of sbrl and wbrl) | bckw | Buckwheat |
| bean | Phaseolous bean | bhsg | Biomass highland sorghum |
| blsg | Biomass lowland sorghum | bsrg | Biomass sorghum (best blsg, bhsg and btsg) |
| btsg | Biomass temperate sorghum | cabb | Cabbage |
| carr | Carrot | casv | Cassava |
| chck | Chickpea | citr | Citrus |
| cocc | Cacao (comum) | coch | Cacao (hybrid) |
| cocn | Coconut | coco | Cacao (the better of comum and hybrid) |
| cofa | Coffee arabica | coff | Coffee (the better of arabica and robusta) |
| cofr | Coffee robusta | cott | Cotton |
| cowp | Cowpea | cyam | Cocoyam |
| dpea | Dry peas | flax | Flax fibre |
| fmlt | Foxtail millet | gras | Pasture grasses |
| grlg | Pasture legumes | grnd | Groundnut |
| gram | Gram | gyam | Greater yam |
| hmze | Highland maize (tropics) | hsrg | Highland sorghum (tropics) |
| jatr | Jatropha | lmze | Lowland maize |
| lsrg | Lowland sorghum | maiz | Maize (best of lmze, hmze and tmze) |
| misc | Miscanthus | mllt | Millet (better of fmlt and pmlt) |
| mzsi | Silage maize | napr | Napier grass |
| oats | Oat | oilp | Oil palm |
| oliv | Olive | onio | Onion |
| pigp | Pigeon pea | pmlt | Pearl millet |
| prub | Para-rubber | rape | Rapeseed |
| rcgr | Reed canary grass | ricd | Dryland rice |
| ricw | Wetland rice | ryes | Rye (the better of srye and wrye) |
| sbrl | Spring barley | sorg | Sorghum (best of lsrg, hsrg and tsrg) |
| soyb | Soybean | spot | Sweet potato |
| srye | Spring rye | sugb | Sugar beet |
| sugc | Sugar cane | sunf | Sunflower |
| swhe | Spring wheat | swgr | Switchgrass |
| teas | Tea (best of China, Assam and hybrid types) | tmze | Temperate/sub-tropical maize |
| toba | Tobacco | toma | Tomato |
| tsrg | Temperate/sub-tropical sorghum | wbrl | Winter, sub-tropical and tropical highland barley |
| whea | Wheat (the better of swhe and wwhe) | wpot | White potato |
| wrye | Winter rye | wwhe | Winter, sub-tropical and tropical highland wheat |
| wyam | White yam | yams | Yam (best of wyam, gyam, yyam and cyam) |
| yyam | Yellow yam |
See the GAEZ v5 dataset on agro-climatic potential yield (kg DW/ha) for different crops and scenarios on the FAO Agro-Informatic Data Catalog (link).