Crop Science 42:1943-1949 (2002)
© 2002 Crop Science Society of America
CROP ECOLOGY, MANAGEMENT & QUALITY
Using Satellite and Field Data with Crop Growth Modeling to Monitor and Estimate Corn Yield in Mexico
Alma Delia Báez-González*,a,
Pei-yu Chenb,
Mario Tiscareño-Lópeza and
Raghavan Srinivasanb
a Laboratorio Nacional de Predicción de Cosechas y Monitoreo Climático, Campo Experimental de Pabellón, INIFAP, Km. 32.5 Carr. Zac-Ags., Ap. Postal #20, Pabellón de Arteaga, Aguascalientes, Mexico
b Spatial Science Lab., Dep. of Forest Science, Texas A&M Univ., College Station, TX 77843, USA
* Corresponding author abaez{at}pabellon.inifap.conacyt.mx
 |
ABSTRACT
|
|---|
The large-scale monitoring and estimation of crop yield is essential for food security in Mexico. This study developed and validated a method of monitoring and estimating corn (Zea mays L.) yield by means of satellite and ground-based data. In autumn–winter 1999 and spring–summer 2000, eight locations under irrigated and nonirrigated conditions in corn valleys of Mexico were localized by Global Positioning Systems (GPS) and were sampled every 15 d. Photosynthetic active radiation (PAR), leaf area index (LAI), crop development stage (DVS), planting dates, and grain yield data were gathered from the field. The normalized difference vegetation index (NDVI) was derived from NOAA-Advanced Very High Resolution Radiometer (AVHRR) images. A growth model was developed to integrate satellite and ground data. Net primary productivity (NPP) was estimated using PAR and NDVI. Dry weight increase (kg ha-1 d-1) was determined considering NPP and the partitioning factor. Results indicated that the model accounts for 89% of the variability in yields under irrigated conditions and 76% under nonirrigated conditions. The methodology seems advantageous in large-scale monitoring and assessment of corn yield.
Abbreviations: AVHRR, advanced very high resolution radiometer • DAS, days after sowing • DVS, development stage of the crop • GPS, global positioning system • HRPT, high resolution picture transmission • INIFAP, agricultural and forestry national research institute of Mexico • LAI, leaf area index • MVC, maximum value composite • NDVI, normalized difference vegetation index • NIR, near-infrared channel • NPP, net primary productivity • PAR, photosynthetic active radiation • SD, Standard deviation • TOA, top-of-the atmosphere reflectance • VCI, vegetation coefficient index • VIS, visible channel
|
INTRODUCTION
|
|---|
ACCURATE and up-to-date reports of regional and national crop conditions and yield are often difficult to obtain in Mexico because large areas are involved and traditional methods of yield estimation such as visual assessment of farm plots and survey of grain stores are resource and time consuming. To remedy this situation, the National Laboratory for Crop Prediction, which is part of the Agricultural and Forestry National Research Institute of Mexico (INIFAP), in collaboration with the Spatial Sciences Laboratory of the Texas A&M University and the USDA-ARS, has embarked on a project to develop approaches for predicting, monitoring, and assessing crop yield at national scale using remote sensing technology and modeling.
Satellite imagery has been used to assess crop yield at regional level (Prince, 1991; Fueller, 1998; Sannier et al., 1998; Seiler et al., 1998). It has been efficient in measuring crop parameters such as photosynthetic rate, photosynthetic active biomass, photosynthetic size of canopy, LAI, and NPP (Sellers, 1985, 1987; Weigand and Richardson, 1984, 1990; Chen et al., in press; Ochi et al., 2000).
The relationship between satellite information and crop parameters is based on vegetative indices, e.g., the NDVI, which provides an important source of information on vegetation function as well as land-use cover (Tan and Shih, 1997; Fang et al., 1998; Jiang and Islam, 1999; Ochi and Murai, 1999). The NDVI derived from satellite-image data has been strongly linked to vegetation condition and plant biomass on the land surface (Tan and Shih, 1997; Fang et al., 1998; Jiang and Islam, 1999; Ochi and Murai, 1999). Values for NDVI range from -1.0 to 1.0. Larger NDVI values indicate that the land surface is covered with dense healthy vegetation, while negative values indicate the presence of clouds, snow, water, or a bright nonvegetated surface (Yin and Williams, 1997). A typical NDVI temporal profile for healthy green vegetation rises as plant cover increases in spring, reaches a peak or plateau during summer, and declines with plant senescence in fall.
Cloud contamination that appears in virtually every AVHRR scene decreases NDVI values; therefore, daily NDVI images in a continuous time series do not always depict vegetation conditions accurately during the growing season. To minimize the effects of cloud contamination, the MVC procedure is used (Holben, 1986). However, it has been observed that 10-d composites are not cloud free and 2- or 3-wk NDVI composites are still cloud contaminated. It was thus essential in this study to develop an alternative to solve this problem. The solution is to retain high temporal resolution by detecting and removing cloudy pixels from daily AVHRR scenes, and then creating 10- or 15-d NDVI composites from cloud-free data only.
Mathematical modeling is a research strategy which contributes to understanding a real system in a more practical and economical way (de Wit, 1969; Penning de Vries, 1977). Models have the advantage of enabling researchers to evaluate a wide array of alternatives and to assemble processes in an integrated package (Branson et al., 1981; Báez-González, 1992; Báez-González and Jones, 1995a, b; Tiscareño-López et al., 1999).
Previous studies have shown that satellite data complement the performance of crop growth models and improve model estimates of crop yield (Kanemasu et al., 1985; Maas, 1988a,b; Jaggard and Clark, 1990, p. 201–206). Maas (1988a) described several techniques for using remotely sensed information with agricultural crop growth models: satellite data may be used as model input to evaluate one or more driving variables. They may also be used to update model simulations. Other techniques are reinitialization and reparameterization, which may be used to adjust model initial conditions and parameter values to achieve agreement between simulated and observed growth. In this study, a simple growth model was developed to simulate corn by means of remotely sensed observations and ground-based data as driving variables.
The objective of this study was to develop and validate a methodology of monitoring and estimating corn yield under irrigated and nonirrigated conditions in near real time by means of satellite information and ground-based data with crop growth modeling.
|
MATERIALS AND METHODS
|
|---|
Field Data
Eight locations (>300 has each) in important corn valleys of Mexico were monitored during the growing season of autumn–winter 1999 and spring–summer 2000 (Fig. 1)
. The locations, which were in farmers' fields, had 90% or greater homogenous corn cover. Four locations were irrigated while four were nonirrigated. At least four sampling sites of 10 m2 each were permanently established in each of the eight locations. GPS was used to locate the sampling sites, and the planting dates are reported in Tables 1 and 2
.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 1. Geographic location of the corn field areas monitored during the growing season of autumn–winter 1999 and spring–summer 2000 in Mexico.
|
|
Every 15 d during the growing season, the following crop information was gathered: PAR, LAI, and DVS. A linear PAR/LAI ceptometer was used to sample PAR and LAI in each site. The decimal code for the description of the development stage of the crop was used (Tottman et al., 1979). At the end of the growing season, grain yield (kg ha-1) was measured in the field by destructive methods on the same day that the farmers harvested the fields. At each site, two rows (5 m each) of corn were harvested and a sample of grain yield was collected. Grain weight was reported at 140 g kg-1 of moisture.
It was not possible to collect weather data in real time in the study areas. Soil water content was not measured; however, every 15 d visual observations of the crop condition (moisture stress, pest, etc.) were reported.
Satellite Data
AVHRR High Resolution Picture Transmission (HRPT) data were downloaded daily from the NOAA-14 satellite to the receiving station at the Backland Research Center in College Station, TX, USA. Each AVHRR scan line contains 2048 pixels with a resolution of 1.1 km, and every AVHRR scene extends from the U.S.-Canadian border to southern Mexico.
Raw AVHRR data were acquired to level 1b format. Metadata extracted from the AVHRR header file includes orbit number, earth location in geographical coordinates, data acquisition time, and other related information. The metadata were used for geocoding and thermal data calibration. The level 1b data were processed by PCI image processing software (PCI Geomatics Group, 1998). The preprocessing for this study included radiometric calibration and geometric correction for all 5 channels; conversion to reflectance and atmospheric correction (Rahman and Dedieu, 1994) for channels 1 (visible channel [VIS]) and 2 (near-infrared channel [NIR]); cloud removal; and computation of satellite zenith, solar zenith, and relative azimuth angles.
The AVHRR data were calibrated as reflectance factor (also known as albedo) in percentage for the VIS and NIR channels, and as brightness temperature in Kelvin degree for thermal channels 3 to 5. The reflectance of channels 1 and 2 was named top-of-the-atmosphere (TOA) reflectance before atmosphere correction, and surface reflectance after atmosphere correction.
Cloud-contaminated pixels were assigned a specific value to differentiate them from clear-sky pixels. Normalized Difference Vegetation Index values were derived as (NIR - VIS)/(NIR+VIS) on the basis of surface reflectance. The NDVI values ranged from -1.0 and +1.0. The MVC procedure (Holben, 1986) for the removal of contaminated pixels was applied for creating 10- and 15-d NDVI composites for irrigated and nonirrigated areas, respectively. Each NDVI composite represents the maximum NDVI value of the same locations in 10 and 15 d. Holben (1986) suggested a limit for solar zenith angle of 80. Hence, pixels having a solar zenith angle greater than 80 were not used for NDVI composites in this study.
Simulation of Corn Growth
In this study, the system approach (Spedding, 1979) was applied to integrate information and mimic the behavior of the crop. The corn crop was considered the system under study. The simulation model was the main tool used to mimic the behavior of the corn crop and to integrate satellite and ground-based data. A schematic representation of the crop growth model used in this study is presented in Fig. 2
. It shows the link between a series of processes influencing the growth and development of the crop.
Assessment of Net Primary Productivity
The NPP is a fraction of the incoming solar energy stored into organic dry matter. The NPP was calculated following the approach of Goward and Huemmrich (1992) and Ruimy et al. (1994):
 | [1] |
where NPP = net primary productivity, NDVI = normalized difference vegetation index, and PAR = photosynthetic active radiation.
The PAR is defined as the radiation in the 400- to 700-nm waveband (Hay and Walker, 1989). It represents the portion of the solar spectrum that plants use for photosynthesis. Reported PAR and the corresponding NDVI values for the sampling period were used in Eq. [1] to calculate NPP value for the irrigated and nonirrigated sites.
Development Stage, Growth, and Partitioning of Dry Matter
Development stage is defined as progress from germination to maturity and major phases can be distinguished by well-defined events such as flower initiation and anthesis (Ong and Squire, 1984).Growth implies the conversion of primary photosynthesis into structural plant material (Hay and Walker, 1989). The total growth of the crop (kg ha-1 d-1) is partitioned among the plant organs according to partitioning coefficients adapted from Penning de Vries et al. (1989) and introduced as forcing functions; their values change with the developmental stage of the crop reported from field observation. For the purpose of this study, leaves and stems were considered as one plant organ. The simulated and measured grain yields of each site were compared and results were analyzed by regression (Steel and Torrie, 1980).
|
RESULTS AND DISCUSSION
|
|---|
The simulation of grain yield under irrigated conditions showed a mean error of prediction (simulated minus measured) of 0.05 Mg ha-1 and the growth model accounts for 89% of the variability in measured yields (Table 3
and Fig. 3)
. In the case of yield under nonirrigated conditions, the mean error of prediction was 0.32 Mg ha-1 and the model accounts for 76% of the variability in measured yields (Table 4
and Fig. 4)
.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 3. Measured and simulated corn grain yield in irrigated sites in Mexico during autumn–winter 1999–2000. The solid line is the fitted regression line and the dashed line is the 1:1 fit.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 4. Measured and simulated corn grain yield in nonirrigated sites in Mexico during spring–summer 1999–2000. The solid line is the fitted regression line and the dashed line is the 1:1 fit.
|
|
Similar results were reported by Ochi et al. (2000) who mention a correlation factor of 0.91 between the global NPP and the world cereal production. The results are also consistent with the correlation coefficient reported by Seiler et al (1998), of 0.7 between corn yield and vegetation coefficient index (VCI).
Tables 5 and 6
present the LAI measured at silking stage of the crop under irrigated and nonirrigated conditions. It is important to mention that the crops at nonirrigated sites presented a wide range of management variability (i.e, different fertilizer application, seed rate, etc.). Nevertheless, the analysis shows that the approach of assessing grain yield by estimating NPP based on NDVI and PAR measured in the field is a method that can be applied to any management condition.
View this table:
[in this window]
[in a new window]
|
Table 5. Leaf area index (LAI) of corn at silking stage in irrigated sites in Mexico during autumn–winter 1999–2000.
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Leaf area index (LAI) of corn at silking stage in nonirrigated sites in Mexico during spring–summer 2000.
|
|
It was possible to assess grain yield with an accuracy of 0.1 to 0.5 Mg ha-1 of difference between simulated and measured yields when the field data (PAR, LAI, DVS) were gathered at the initial stage of plant development. As illustrated in Tables 7 and 8
, the lowest difference between simulated and measured grain yield was shown mostly in sites where the crop was at the vegetative stage of development at the start of the monitoring period; this is true of both irrigated and nonirrigated sites. One reason for this high degree of accuracy is explained by Serrano et al. (2000) who mention that NDVI is sensitive for low LAI (<3), which makes it suitable for assessing crop growth at the initial stages.
View this table:
[in this window]
[in a new window]
|
Table 7. Simulation of NPP and corn grain yield in irrigated sites in Navolato and Culiacan using days after sowing (DAS), photosynthetic active radiation (PAR), leaf area index (LAI) and development stage of the crop (DVS) from the field and NDVI from satellite images. Comparison of simulated and measured grain yield.
|
|
View this table:
[in this window]
[in a new window]
|
Table 8. Simulation of NPP and corn grain yield in nonirrigated sites at Nayarit and Chiapas by means of days after sowing (DAS), photosynthetic active radiation (PAR), leaf area index (LAI), and development stage of the crop (DVS) from the field and NDVI from satellite images. Comparison of simulated and measured grain yield.
|
|
On the other hand, an over- or underestimation of approximately 1.0 Mg ha-1 of grain yield was observed mostly in irrigated (Table 7) and nonirrigated sites (Table 8) where data were gathered at the reproductive stage of the crop. One possible reason for this simulation error is the partitioning coefficients used by the model when the plant reached the reproductive stage of development. During this stage, the NPP calculated by the model is assigned fully to the panicle of the plant. Therefore, there is a need to improve the calibration of the coefficients.
In this study, composite images were used to reduce cloudy effects. However, in some locations, such as Nayarit and Jalisco, which are nonirrigated (Table 8), it was difficult to have NDVI values for some sites because of the cloudy conditions which sometimes lasted for more than 15 d. It has been reported by Holben (1986) and Gutman (1991) that the atmospheric moisture has a strong influence on NDVI values becaue of the different effects of water vapor on bands 1 and 2. Justice et al. (1991) have likewise mentioned that the NDVI values are reduced by the atmospheric moisture, affecting the accuracy of existing NDVI versus biomass relationships for the Sahelian region. According to Diallo et al. (1991), the effects of increased atmospheric interference by haze and water vapor could change the satellite measurements of NDVI.
The 3x3 window method applied in remote sensing (Chen et al., in press) was used to get the missing values for Nayarit and Jalisco; however, because of the wide extent of the cloudy area, it was not possible to get any value. Hence, the previous NDVI value was used to replace the missing one. This may have also affected the accuracy of yield assessment at those two locations.
On the whole, the results of the study show that it is possible to monitor crop growth and assess grain yield on a large scale through the integration of satellite imagery, field data, and growth modeling. The degree of accuracy varies according to the atmospheric interference and crop management conditions.
The methodology can be improved by assessing PAR by the use of imagery data, making field work unnecessary. In this study, field data was essential for validating the approach. To expand the potential of the methodology, it is necessary to validate total dry matter production simulated by the model.
|
ACKNOWLEDGMENTS
|
|---|
We thank Jaime Macias, Jose Ramón Manjarrez, Jose Luis Mendoza, Aurelio Palacios, Valerio Palacios, Carlos Tinoco, and Roberto Valdivia for technical and field assistance, and Elvira Aranda Tabobo for assistance in the preparation of the manuscript.
Received for publication August 21, 2001.
|
REFERENCES
|
|---|
- Báez-González, A.D. 1992. A simulation study of sorghum and pearl millet forage crops in semi-arid Mexico. Ph.D. diss. Univ. of Reading, Reading, England.
- Báez-González, A.D., and J.G.W. Jones. 1995a. Models of sorghum and pearl millet to predict forage dry matter production in semi-arid Mexico. I. Simulation models. Agric. Syst. 47:133–145.
- Báez-González, A.D., and J.G.W. Jones. 1995b. Models of sorghum and pearl millet to predict forage dry matter production in semi-arid Mexico. II. Regression equations. Agric. Syst. 47:133–145.
- Branson, F.A., G.F. Gifford, K.G. Renard, and R.F. Hadley. 1981. Rangeland hydrology. Range Science Series. No. 1. Soc. of Range Management, Denver, CO.
- Chen, P.Y., R. Srinivasan, G. Fedosejevs, and R. Kiniry. Monitoring crop growth in Texas using weekly NDVI composites with/without cloud detection derived from NOAA–14 AVHRR data. Int. J. Remote Sens. (In press).
- Diallo, O., A. Diouf, N.P. Hannan, A. Ndiaye, and Y. Prévost. 1991. AVHRR monitoring of savanna primary production in Senegal, West Africa: 1987–1988. Int. J. Remote Sens. 12:1259–1279.
- Fang, H., B. Wu, H. Liu, and X. Huang. 1998. Using NOAA AVHRR and Landsat TM to estimate rice area year-by-year. Int. J. Remote Sens. 19:521–525.
- Fueller, D.O. 1998. Trends in NDVI series and their relation to rangeland and crop production in Senegal, 1987–1993. Int. J. Remote Sens. 19:2013–2018.
- Goward, N.S., and K.F. Huemmrich. 1992. Vegetation canopy PAR absorbance and the normalized difference vegetation index: An assessment using the SAIL Model. Remote Sens. Environ. 39:119–140.
- Gutman, G.G. 1991. Vegetation indices from AVHRR: an update and future prospects. Remote Sens. Environ. 35:121–136.
- Hay, K.M.R., and A.J. Walker. 1989. An introduction to the physiology of crop yield. Longman Scientific & Technical. Essex, England.
- Holben, B.N. 1986. Characteristics of maximum-values composite images from temporal AVHRR data. Int. J. Remote Sens. 7:1417–1434.
- Jaggard, K.W., and C.J.A. Clark. 1990. Remote sensing to predict the yield of sugar beet in England. Application of remote sensing in agriculture. Butterworth-Heinemann, Woburn, MA.
- Jiang, L., and S. Islam. 1999. A methodology for estimation of surface evapotranspiration over large areas using remote sensing observations. Geophys. Res. Lett. 26:2773–2776.
- Justice, C.O., T.F. Eck, D. Tandré, and B.N. Holben. 1991. The effect of water vapour on the normalized difference vegetation index derived for the Sahelian region from NOAA AVHRR data. Int. J. Remote Sens. 12:1165–1187.
- Kanemasu, E.T., G. Asrar, and M. Yoshida. 1985. Remote sensing techniques for assessing water deficits and modeling crop response. HortScience 20:1043–1046.[ISI]
- Maas, S.J. 1988a. Use of remotely-sensed information in agricultural crop growth models. Ecol. Modell. 41:247–268.
- Maas, S.J. 1988b. Using satellite data to improve model estimates of crop yield. Agron. J. 80:655–662.[Abstract/Free Full Text]
- Ochi, S., and S. Murai. 1999. Analysis of relationship between NPP and population carrying capacity for major river basins in Asia. p. 258–262. In Proceedings of 9th SEIKEN Forum "Global Environment Monitoring from Space."
- Ochi, S., R. Shibasaki, and S. Murai. 2000. Modeling and assessment of NPP/crop productivity in Asia by GIS combined with remote sensing data. 2–3 Nov. 2000. International Rice Research Institute (IRRI) Los Baños, Philippines.
- Ong, C.K. and G.R. Squire. 1984. Response to temperature in a stand of pearl millet (Pennisetum typhoides S.&H.). 7. Final number of spikelets and grains. J. Exp. Bot. 35: 1233–1240.[Abstract/Free Full Text]
- PCI Geomatics Group. 1998. Introduction to using PCI software, Version 6.3. Canada: PCI Enterprises, Richmond Hill, ON, Canada.
- Penning de Vries, F.W.T. 1977. Evaluation of simulation models in agriculture and biology: Conclusions of a Workshop. Agric. Syst. 2:99–105.
- Penning de Vries, F.W.T., D.M. Jansen, H.F.M. Gen Berge, and A. Bakema. 1989 Simulation of ecophysiological process of growth in several annual crops. Simulation Monographs. PUDOC, Wageningen, the Netherlands.
- Prince, S.D. 1991. Satellite remote sensing of primary production: comparison of results for Sahelian grassland 1981–1988. Int. J. Remote Sens. 12:1301–1311.
- Rahman, H., and G. Dedieu. 1994. SMAC: A simplified method for the atmospheric correction of satellite measurements in the solar spectrum. Int. J. Remote Sens. 15:123–143.
- Ruimy, A., B. Saugier, and G. Dedien 1994. Methodology for the estimation of terrestrial net primary production from remote sensed data. J. Geophys. Res. 99:5263–5283.
- Sannier, C.A.D., J.C. Taylor, W. Du Plessis, and K. Campbell. 1998. Real-time vegetation monitoring with NOAA-AVHRR in Southern Africa for wildlife management and food security assessment. Int. J. Remote Sens. 19:621–639.
- Seiler, R.A., F. Kogan, and J. Sullivan. 1998. AVHRR-based vegetation and temperature condition indices for drought detection in Argentina. Adv. Space Res. 21:481–484.
- Sellers, P.J. 1985. Canopy reflectance, photosynthesis, and transpiration. Int. J. Remote Sens. 6:1335–1372.
- Sellers, P.J. 1987. Canopy reflectance, photosynthesis and transpiration. II. The role of biophysics in the linearity of their interdependence. Remote Sens. Environ. 21:143–183.
- Serrano, L., I. Filella, and J. Peñuelas. 2000. Remote sensing of biomass and yield of winter wheat under different nitrogen supplies. Crop Sci. 40:723–731.[Abstract/Free Full Text]
- Spedding, C.R.W. 1979. An introduction to agricultural systems. Applied Science Publishers, UK.
- Steel, R.G.D., and J.H. Torrie. 1980. Principles and procedures of statistics: A biometrical approach. 2nd ed. McGraw-Hill, New York.
- Tan, C.H. and S.F. Shih. 1997. Using NOAA satellite thermal infrared data for evapotranspiration estimation in South Florida. Soil and Crop Science Society of Florida. Proceedings, Volume 56.
- Tiscareño-López, M., A.D. Báez-González, M. Velásquez-Valle, K.N. Potter, J.J. Stone, M. Tapia-Vargas, and R. Claveran-Alonso. 1999. Agricultural research for watershed restoration in Mexico. J. Soil Water Conserv. 54:686–692.
- Tottman, D.R., R.J. Makepeace, H. Broad. 1979. An explanation of the decimal code for the growth of cereals, with illustration. Ann. Appl Biol. 93:221–34.[ISI]
- Wiegand, C.L. and A.J. Richardson. 1984. Leaf area, light interception, and yield estimates from spectral components analysis. Agron. J. 76:543–548.[Abstract/Free Full Text]
- Wiegand, C.L., and A.J. Richardson. 1990. Use of spectral vegetation indices to infer leaf area, evapotranspiration and yield: I. Rationale. Agron. J. 82:623–629.
- de Wit, C.T. 1969. Dynamic concepts in biology. In J.G.W. Jones (ed.) The use of models in agricultural and biological research. Agricultural Research Council and Grassland Research Institute. Hurley, Maidenhead, Berkshire, U.K.
- Yin, Z., and T.H.L. Williams. 1997. Obtaining spatial and temporal vegetation data from Landsat MSS and AVHRR/NOAA satellite images for a hydrologic model. Eng. Remote Sens. 63:69–77.
This article has been cited by other articles:
|
|
|
|
 
D. Inman, R. Khosla, R. Reich, and D. G. Westfall
Normalized Difference Vegetation Index and Soil Color-Based Management Zones in Irrigated Maize
Agron. J.,
January 11, 2008;
100(1):
60 - 66.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Z. Shi, G. R. Ruecker, M. Mueller, C. Conrad, N. Ibragimov, J. P. A. Lamers, C. Martius, G. Strunz, S. Dech, and P. L. G. Vlek
Modeling of Cotton Yields in the Amu Darya River Floodplains of Uzbekistan Integrating Multitemporal Remote Sensing and Minimum Field Data
Agron. J.,
September 11, 2007;
99(5):
1317 - 1326.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Ko, S. J. Maas, S. Mauget, G. Piccinni, and D. Wanjura
Modeling Water-Stressed Cotton Growth Using Within-Season Remote Sensing Data
Agron. J.,
October 31, 2006;
98(6):
1600 - 1609.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. D. Baez-Gonzalez, J. R. Kiniry, S. J. Maas, M. L. Tiscareno, J. Macias C., J. L. Mendoza, C. W. Richardson, J. Salinas G., and J. R. Manjarrez
Large-Area Maize Yield Forecasting Using Leaf Area Index Based Yield Model
Agron. J.,
March 1, 2005;
97(2):
418 - 425.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. B. Lobell, J. I. Ortiz-Monasterio, G. P. Asner, R. L. Naylor, and W. P. Falcon
Combining Field Surveys, Remote Sensing, and Regression Trees to Understand Yield Variations in an Irrigated Wheat Landscape
Agron. J.,
January 1, 2005;
97(1):
241 - 249.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
