PRECISION AGRICULTURE INITIATIVE FOR TEXAS HIGH PLAINS

2001 ANNUAL REPORT

Texas Agricultural Experiment Station and Texas Cooperative Extension

Reporting Period:  January 1, 2001 – December 31, 2001

A.        Summary of Progress:  (Summarize progress on each specific objective listed above)  

1. Utilize GPS-referenced information on chemical and physical soil properties and soil anomalies; irrigation methods and intensities; plant tissue properties; aerial imaging; and yield mapping technology to diagnose water-related production problems and guide peanut production practices for economically and environmentally sustainable yields of high quality peanuts.

Irrigation Application Research

The work was conducted at the Western Peanut Growers Research Farm in Gaines County, Texas.   Irrigation studies were conducted in the fifth through seventh spans of the east irrigation circle.  Flavorunner 458 peanuts were planted on May 16-17, 2001; they were dug October 18-20.  Four sample locations from each of 7 treatments were hand harvested October 27-29; the area was machine harvested November 5-6, 2001.   Nitrogen fertilizer was applied at a rate of 20 lb/acre three times during the season, for a total of 60 lb/acre.  Base irrigation rate targeting 75% evapotranspiration replacement was 24.45 inches; in-season precipitation received was 4.8 inches.   Reference crop evapotranspiration was estimated at 46.2 inches; peanut evapotranspiration was estimated between 23.9 inches and 24.84 inches, depending upon crop coefficient curve used.   Total irrigation and precipitation exceeded target values by up to 40%, assuming very high irrigation application efficiency. 

Irrigation Application Rates

Irrigation application rates targeting 50%, 75%, 100% and 125% evapotranspiration replacement were applied through LEPA irrigation through the 2000 and 2001 cropping seasons.  Standard LEPA practice included application by drag hoses in alternate furrows, circular planting pattern to match traffic of the center pivot irrigation system, and furrow dikes (to the extent practical) to improve in-furrow water application uniformity.    Yield was determined through plot sampling (2 rows by 20 ft. for each treatment and replication block) and with a yield mapping combine.   Samples were graded in the laboratory to determine whether different irrigation treatments effected differences in product quality.

Surprisingly we did not find statistically significant differences in yield or grade responses to different irrigation application rates during the 2000 or 2001 crop seasons.  This was attributed in part to high spatial variability (soil differences) within the year 2000 study area, which may have masked effects of the different irrigation rates.   A distribution uniformity test performed early in the 2001 season indicated variability in application depth within the irrigation rate study area; hence a more detailed analysis is warranted.  Field slope in the year 2001 study area was more pronounced, generating concern that the higher irrigation application rates were washing out furrow dikes more severely, and hence the application rates (to the field surface) were somewhat different from the effective irrigation (into the soil profile).  More detailed spatial and slope-specific analyses are underway to investigate these issues further.   One of the goals in this research is to determine most suitable irrigation methods for given conditions; this includes testing the field slope limits of LEPA applicability.

Table 1.  Mean harvested yield, hand sample and yield mapping combine, 2001.

Irrigation rate, Target %ET	50% ET	75% ET	100% ET	125% ET
Hand harvested Yield (lb/ac)	3,267 (a)*	3,639 (a)	3,467 (a)	3,376 (a)
Yield Monitor Yield (lb/ac)	3,558 (a)	3,335 (a)	3,152 (a)	3,518 (a)
Grade	71.2 (a)	73.8 (a)	72.3 (a)	73.0 (a)
Estimated Actual %ET Replacement	63 %	94%	125%	150%

* Note: values in each row followed by the same letter are not significantly different at 0.05

 probability.

     Target ET replacements were exceeded by approximately 25%. Late season increase in water

application rate intended to protect other studies in the field contributed to the excess.

Irrigation Application Methods

Throughout the 2000 and 2001 cropping seasons two LEPA methods (drag hoses and bubbler-mode nozzles) and two LESA methods (low drift spray and wobbler-type nozzles) were used to apply water at a base target irrigation rate of 75% crop evapotranspiration replacement.   Yield was determined through plot sampling (2 rows by 20 ft. for each treatment and replication block) and with a yield mapping combine.  Samples were graded in the laboratory to determine whether different irrigation treatments effected differences in product quality.

In the 2000 cropping season, yield results from LEPA irrigation application treatments were significantly better than those from LESA irrigation treatments.   The LEPA treatments (drag hoses and bubbler nozzles) were not different from each other.  The LESA treatments (low drift spray and wobbler spray) were not different from each other.   The yield advantage of LEPA over LESA in 2000 was consistent with results from deficit irrigation studies conducted at another site in Dawson County over the period 1995-2000.   This yield advantage of LEPA over LESA irrigation methods observed in the 2000 cropping season was not observed in 2001.   In 2001, yield was not significantly different between or among LEPA and LESA irrigated peanuts. 

Sample peanut grades from the spray irrigated peanuts were somewhat better than the LEPA irrigated peanuts.   (Grades from LESA irrigated peanut samples were 75.5 to 76.2; grades from LEPA irrigated peanut samples were 71.5 to 73.8.)  These differences, however, did not result in a significantly different crop value per acre basis (mean crop value was approximately $1,143/ac using quota loan rate prices).

In considering the 2000 and 2001 cropping seasons, there are a few differences that may explain the mixed experimental results.   In 2000, a late season foliar necrosis was observed to occur rapidly and severely in the spray-irrigated (LESA) peanuts, but was essentially non-existent in the LEPA irrigated peanuts.  The problem apparently affected overall yield, and obviously affected harvest losses.  Investigators are still considering water quality and potential fungal effects or interactions to explain this occurrence.  In the 2001 season, the plants showed little or no sign of this foliar damage.  

Table 2.  Mean harvested yield, hand sample and yield mapping combine, 2001.

Irrigation Method	Spray	Wobbler	LEPA	Bubbler
Hand harvested Yield (lb/ac)	4,029 (a)*	3,893 (a)	3,639 (a)	3,485 (a)
Yield Monitor Yield (lb/ac)	3,892 (a)	3,490 (a)	3,335 (a)	3,674 (a)
Grade	76.2 (a)	75.5 (a)	73.8 (a,b)	71.5 (b)
Estimated Actual %ET Replacement	94%	94%	94%	94%

* Note: values in each row followed by the same letter are not significantly different at

0.05 probability.

During the 2001 season, there was little or no precipitation, and the research team was especially concerned that limited lateral water movement from wet furrows across the planted rows may have been inadequate to support good conditions for pegging and pod development.  Base irrigation treatments on the circle outside the irrigation study area were converted to spray irrigation to support soil fertility and plant breeding research; irrigation study treatments were left intact to preserve the study.   Irrigation rates were increased near the end of the season, also to accommodate other research in the field.  This resulted in a total irrigation rates that exceeded the original targeted rates.   Investigators plan to enhance the study in year 2002 to investigate whether there is a yield and/or quality benefit to converting between LEPA and LESA within the season to balance the water efficiency benefits of LEPA with the apparent better pegging conditions supported by LESA irrigation.  Further, we plan to include an additional variety of peanuts in the study to determine whether there is a variety by environment interaction.

Comparative LEPA Performance in Different Crop Years

            The 2001 crop year was extremely dry with only 0.05" of rainfall from June 29 to August 22 at AG-CARES. A comparison of 2001 rainfall patterns with the 1956-2000 average patterns for Dawson County is shown in Table 3. The table also shows comparisons for approximate peanut crop growth phases: (1) Pre-watering April; (2) Vegetative phase May - June; and (3) Fruiting phase July - Mid-September, although the historic rainfall amounts include July - September and is, therefore, perhaps an over-estimate of the duration of active pod development.

            As the season progressed, research personnel became concerned by late-July that pod development was not progressing normally. By early August concern intensified; and on August 14, we dropped drag hoses and socks off of all drops except for one 24-row area, which remained LEPA, and applied two passes of 0.4" of irrigation water totaling 0.8" in an effort to wet the pod development zone. Drag hoses and socks were then reinstalled and LEPA irrigation resumed on the entire field.

            Peanut yields were disappointing in 2001, averaging only 3,622 lb/ac in the areas used for these comparisons. A comparison of PYMS data from the 24-row area that remained LEPA was made with an adjacent 24-row area that received the August 14 LESA irrigation. The peanuts that received the LESA applications significantly out-yielded (3,775 lb/ac) those that remained LEPA (3,469 lb/ac).

______________________________________________________________________________

Table 3. Comparison of 2001 Rainfall with 1956-2000 averages by month and approximate crop development period.    

______________________________________________________________________________

Month                          2001    1956-2000                   Crop Period                 2001    1956-2000

April                             0.09"      1.25"                          Pre-Water                    0.09"      1.25" 

May                             1.60"      2.36"

June                             1.43"      2.39"                          Vegetative                    3.03"      4.75"

July                              0.00"      2.13"

August                          0.67"      1.68"

September                    1.37"      2.93"                          Fruiting             2.04"      6.74"

______________________________________________________________________________

            Despite past experience at AG-CARES indicating that LEPA was superior or equal to LESA, LEPA did not perform as well in 2001. Although adverse weather conditions complicated harvest, LEPA may not have performed well in 2000 crop year, either. Although the differences were not statistically significant, LESA plots yielded an average of 1,952 lb/ac while LEPA averaged 1,629.

            We have examined aspects of the 2001 crop year in an attempt to understand these results. Several possibilities exist: (1) 2001 was a very dry year; (2) 2001 was a very hot year; and (3) FlavorRunner 458 was used in 2000 and 2001, while Florunner was used in 1995-1997 and Tamrun 88 was used in 1998-1999.

Low Rainfall

            Although 2001 was a dry year, 1998 was drier than 2001 (see Table 4). And in 1998, LEPA out-yielded LESA (see Table 5). Also, in many of the past years, the degree to which LEPA out-yielded LESA increased at lower irrigation application rates. It does not, therefore, seem likely that failure to meet the deficit between water use and that supplied by rainfall was a major independent factor in the poorer performance of LEPA irrigation in 2001.

______________________________________________________________________________

Table 4. Comparison of 1998 Rainfall with 1956-2000 averages by month and approximate crop development period.

______________________________________________________________________________

Month                          1998    1956-2000                   Crop Period                 1998    1956-2000

April                             0.07"      1.25"                          Pre-Water                    0.07"      1.25" 

May                             0.31"      2.36"

June                             1.84"      2.39"                          Vegetative                    2.15"      4.75"

July                              0.56"      2.13"

August                          1.47"      1.68"

September                    0.64"      2.93"                          Fruiting             2.67"      6.74"

______________________________________________________________________________

 

 

______________________________________________________________________________

Table 5. Effect of Cultural Practices on Peanut Yields.  AG-CARES. Lamesa, Texas 1998.

______________________________________________________________________________

Treatment                            Yield (lb/ac)                                                         

______________________________________________________________________________

LEPA 75 ET                            6436 a¹

SPRAY 75 ET                         4516 b

______________________________________________________________________________

¹ Treatment means within the same section followed by the same letter are not statistically significant at the 5% probability level.

______________________________________________________________________________

 

 

High Temperatures

            The 2001 crop year was warmer than the average for 1995-2001 (Fig. 1). It was not, however, warmer than the 1998 crop year (Fig. 2). So, it does not seem likely that warmer than usual temperatures explain the failure of LEPA to perform well at AG-CARES in 2001.

Variety x Environment Interactions

            The introduction of the high O/L variety, FlavorRunner 458, into the LEPA irrigation system is an obvious difference between the 2001 and 2000 crop years and previous years at AG-CARES. Although we did not have Florunner or Tamrun 88 at AG-CARES in 2001 to compare with the FlavorRunner 458, there is enough circumstantial evidence that there may be a variety x environment interaction in water or heat stress damage for us to include its study in future research proposals.

Effect of Irrigation Method on Temperatures and Relative Humidity in the Crop Canopy

            During the period from August 15 through September 25, we measured the temperature and relative humidity (RH) in the canopy of peanut rows irrigated by LEPA and LESA methods. To simplify the discussion, we will only include drag hoses with socks as LEPA and low drift spray nozzles as LESA. Measurements were made using Hobo H8 Pro RH/Temperature Loggers mounted inside a small rain shield, installed in the center of the respective peanut row with the plant canopy surrounding the measurement equipment. Data were studied at hourly intervals.

            Throughout the measurement period, air temperatures in the plant canopy fluctuated above and below those measured at the weather station. Although there were numerous exceptions, temperatures within the LESA irrigated canopies tended to be lower than those in LEPA. From 8/15 to 8/25, average temperatures were 0.76 degrees F cooler in LESA canopies than in LEPA. From 8/26 to 9/4, average temperatures were 0.27 degrees F cooler in LESA canopies than in LEPA. From 9/5 to 9/14, average temperatures were 0.18 degrees F cooler in LESA canopies than in LEPA. From 9/15 to 9/25, average temperatures were 0.02 degrees F cooler in LESA canopies than in LEPA.

            Throughout the measurement period, RH in the plant canopy were generally higher than those measured at the weather station. Although there were some exceptions, RH within the LESA irrigated canopies tended to be higher than those in LEPA. From 8/15 to 8/25, average RH was 0.36 lower in LESA canopies than in LEPA. From 8/26 to 9/4, average RH was 1.86 higher in LESA canopies than in LEPA. From 9/5 to 9/14, average RH was 1.66 higher in LESA canopies than in LEPA. From 9/15 to 9/25, average RH was 3.03 higher in LESA canopies than in LEPA.

            Reduced temperatures and increased RH should encourage flower fertilization and pod set in peanuts. Analyses are continuing to determine whether or not durations of differences in favorable conditions were long enough to affect pod development.

Effect of Irrigation Method on Pod Development Zone Soil Moisture

            Soil samples were collected to a depth of two inches at nine positions across the bed and adjacent “wet” and “dry” furrows. Sample # 1 was collected in the center of the furrow beneath the drop; Sample #’s 2, 3, and 4 progressively higher on the bed toward the center of the row (seed drill); Sample # 5 directly in the center of the row; Sample #’s 6, 7, and 8 progressively lower on the bed toward the center of the “dry” furrow; Sample # 9 directly in the center of the “dry” furrow. See Figure 3. Wet soil samples were weighed, oven dried, reweighed, and gravimetric soil moisture contents calculated. The depths to which a soil probe could be pushed into the soil were also measured at similar locations across the bed and adjacent furrows. Soil samples and probe depths were taken at multiple sites throughout the respective irrigation application method area.

Figure 3. Soil Sampling positions in relationship to Irrigation Application Nozzle.

            On the average, LEPA was wetter than LESA at position #1 by 0.86 % water (by weight) and at the #2 position by 0.67 %. The top 2 inches of soil was slightly drier with LEPA than with LESA at the sampling positions farther away from the “wet” furrow, with differences of 0.89, 0.83, and 1.24 % water, at Sample #’s 7, 8, and 9, respectively. Sample # 7 is on the side of the bed, # 8 is near where the side of the bed meets the bottom of the “dry” furrow, and # 9 is at the center of the “dry” furrow. There was less difference between LEPA and LESA profiles at some sampling dates than at others. At some sampling dates, there was little difference between LEPA and LESA in the portion of the bed that was likely to produce mature pods—Sample position #’s 3 through 7.

            With this information, we are gaining a more precise understanding of impacts of different peanut irrigation application techniques on the microclimate within the peanut canopy and the pod development zone of the soil, as well as the depth of moisture in the soil. Although not specifically site specific at this point, we will add more site specificity in the future with multiple, GPS-reference sampling sites.

General Progress

Base information

            We completed analyses of all half-acre grid soil samples collected in the winter of  2000 for chemical properties for depths of 0-6", 6-12", 12-24", and 24-36" wherever initial sampling depths allowed. We also completed mechanical analysis of soil samples to determine percent sand, silt, and clay at 0-6", 6-12", 12-24", and 24-36" depths. Maps were completed using geographic information software (GIS) for all soil chemical and physical properties. 

            In December, 2000, Greg Sokora, Gordon Paden, and Ron Crumley (USDA, Natural Resources Conservation Service) performed a detailed GPS-referenced elevation analysis of the entire WPGRF with our assistance. This data will allow us to characterize slopes throughout the research farm. The slope data will be especially useful in evaluating site-specific irrigation performance throughout the field. Initially we prepared graphic presentations of this data to illustrate overall elevations. We recently developed the skills necessary to calculate site-specific slopes and to correlate slope down-the-row with crop performance at selected sites in the field. Much work remains to totally exploit this information. Figure 4 is a graphic representation of an elevation map constructed from the elevation data. In initial analyses of 2001 LEPA irrigation areas, there was no relationship between slope down-the-row and yield. These analyses will be continued for 2000 yield maps at WPGRF and 1998-2001 yield maps at AG-CARES.

Peanut Yield Mapping System

            The ability to produce a yield map of the field is a very important tool in site-specific or precision agriculture approaches. Detailed yield maps can then be linked by GPS coordinates to other maps, such as those for soil chemical and physical properties, soil depth, surface slope, remote sensing data (such as, aerial photography), weeds, diseases, nematodes, insects, and varying cultural practices. In 1998, we were able to install and test a peanut yield mapping system (PYMS) developed by engineers and scientists at the University of Georgia (UGA) Coastal Plains Research Station at Tifton, GA.  UGA cooperators included Dr. George Vellidis, Calvin Perry, and Jeffrey Durrence.  With their cooperation, the Texas Agricultural Experiment Station acquired a four-row peanut combine.  The UGA collaborators transported that combine to their shop where they modified it to accept the mapping instrumentation.  The combine was then shipped to Lamesa where we installed load cells, wiring, and other instrumentation with the assistance of personnel from Nix Implement Company.

            We used the PYMS in 1998-2001 at AG-CARES, 1999 in producers' fields, and 2000 and 2001 at WPGRF. Information from yield maps shown in Figures 5, 6, and 7 were used in several of this year's research in this report. Although the system has not been commercialized to date, it has been rugged and accurate in four years of field use in West Texas, as well as by its developers in the Southeast. If commercialized and sold for a reasonable price, PYMS will be of great value in supplying information for farmers' management decisions.

2.                  Adapt and utilize variable rate technology to measure and manage effects of water inputs, yield mapping harvesters, and other instruments in precision agriculture systems.

            We continue to collaborate with scientists at Halfway and the Uvalde center who are doing actual design and testing of variable rate water application technology. Much of our research information about site specific responses to various irrigation application methods and rates will be valuable in putting their creations into practical use. The Peanut Yield Mapping System has been discussed widely in other sections of this report and those of other peanut researchers. We have collaborated with scientists who are directly using hyperspectral technology and will work with them

3.         Evaluate effectiveness of site-specific water management in peanut production systems and transfer useful information to growers, consultants, and industry clientele via meetings, popular and scientific publications, and web sites.

            Progress under this objective has been fairly well covered in other discussions in this document. We have passed along information to various audiences as soon as we were comfortable enough with the accuracy of our observations to do so.