Question 1: Discuss three possible outcomes of Precision Agriculture (PA) with regard to crop yields. For each outcome, state a logical time frame.
Precision Agriculture (PA) that define by United States Department of Agriculture or better known as USDA as a management system that is information and technology based, is site specific and uses one or more of the following sources of data: soils, crops, nutrients, pests, moisture or yield, for optimum profitability, sustainability and protection of the environment. Like any system that is designed to provide improvement, there is a cycle of events which should take place to monitor its effectiveness and helps to understand some of the definition.
Thus, 3 possible outcomes of PA with regard to crop yields are as follows,
1. Higher yield, similar cost.
The most challenging task by farmers and planters today, with due respect to the increase human population are the increase demand of food sources with limited area available to be planted. Thus, among the strategies are to increase production with same level of agriculture input such as total acreage of land and same level of fertiliser as well as pest and disease control practises. To adapt, planters are looking for a better management practises and tools whereby PM are being exploited in such a way that yield will be increased, waste will be reduced and mitigate the economic and security risks faced by agriculture industry.
The aim of PA is looking at the increased efficiencies of agriculture inputs after taking into an accounts of understanding and the implementation that will be in line with natural variability exist in the field. Further, yield is not a goal but to better manage and distribute inputs on site-specific basis to better optimised long term cost and benefits.
Survey conducted in a paddy field shows that even the same level of inputs such as seeds, fertiliser and pesticide at the same location or place may not produce high yield due to differences in soil fertility, water management and human factor. For example, with the implementation of PA, soil fertility maps may be produced and decision can be made on the amount of fertiliser apply with priority made on the less fertile soil area. More fertilizer will be applied in less fertile and less at fertile zone. However, total fertilizer consumption will be the same if looking on big scale of land available.
2. Same yield, lower cost
Increase application of agriculture inputs such as fertilizer may not always end up with increase in yield, but simply hold them constant whilst reducing input costs. PA enable planters to reap increasing profits through better management and the application of more appropriate or reduce chemical treatments also helps to preserve the environment.
PA such as remote sensing is used to detect soil fertility, crops health and provide tools to detect pest and diseases (P&D) as early as possible whereby early measures can be implemented before major outbreak and planters may experiencing lowering losses and profits will be optimised. Applying the same level of pesticide across the entire field may no longer the best policy however, information obtained from PA by using Global Positioning System (GPS), Remote Sensing, Variable Rate Application (VBA), Neural Network and Decision Support System are utilised to decide the best solution.
For example, the level infestation of bagworm can be detected at early stage and major outbreak can be avoided when the application of pesticide can be conducted during instar 1 to 4 of the life cycle of bagworm. The data obtained may be a tool to decide on action to be taken in the field and ensure at the right rate, right time, right place and right type of chemical used. Thus, early detection may promote less pesticide used in the plantation. The yield will be ensured with the proper management of P&D in the field.
3. Higher yield, Lower Cost
The application of PA, may be leveraged on the combination use of sensors, robots, GPS, mapping tools and data-analytic software to customised plant care without jeopardising labour costs. Images on plants individually produced by stationary or robot-mounted sensors and camera equipped drones wirelessly may provide information on stem size, leaf shape and soil moisture content. The information may be used to monitor on plant health and enable planters to decide to plant or harvest crops. Further, PA can improve time management, optimising water, chemical and produce healthier crops with result of better and higher yields.
In a related development, seed producers are applying technology to improve plant “phenotyping.” By following individual plants over time and analysing which ones flourish in different conditions, companies can correlate the plants’ response to their environments with their genomics. That information, in turn, allows the companies to produce seed varieties that will thrive in specific soil and weather conditions. Advanced phenotyping may also help to generate crops with enhanced nutrition.
A survey on soybean growers is US, reveal that
- Growers report an average savings of about 15% on several crop inputs such as seed, fertilizer and chemicals.
- Savings on inputs often pays for the technology within a year for a large cropping operation and two to three years for smaller operations.
- Growers are increasingly using precision tools to conduct their own comparisons on their own ground.
- Findings show that most growers, particularly those with more than 500 total acres, are using several precision farming technologies.
- The larger the acreage increase, the more likely the farmer is to use multiple precision farming technologies.
For example, seed is an input where costs have risen dramatically. Avoiding overlap with new technology not only reduces total input costs, but it also improves yields on acres that used to have poor production because of too much seed. In any given season, growers often find themselves ordering extra seed to cover their acres, but growers who use automatic clutches often return the extra seed, using exactly as much as their acres call for. The savings go directly into their pocket. While growers who do not use precision technology believe that they can save money other ways, like buying at a discount, the combination of wise buying and precision application can save even more.
Question 2: Remote Sensing (RS) is a key enabler of precision agriculture (PA). Discuss how remote sensing can be used for crop stress detection as a function of time (temporal variability).
Remote sensing can be describe as a technology to acquire and extract the information about the earth surface or any phenomenon without making any physical contact with the object. Common understanding of remote sensing is a satellite technology that capture the picture of the earth including earth surface, atmosphere and ocean. Nowadays, remote sensing is widely used in many applications such as geology, oceanography, cartography, land survey, military and also agriculture. Remote sensing can be divided into two category, active and passive remote sensing. Passive remote sensing is the sensor capture the reflectance of sunlight from the objects. Example of remote sensing satellite such as LandSat-TM, SPOT-5, IKONOS and the latest is WorldView-IV with high-spatial resolution imagery. Meanwhile, active remote sensing required the sensor to transmit the signal to the object, and emit the reflected signal or back-scatter from the object. Example of active remote sensing system such as Radar-Sat and LiDAR.
Remote Sensing used for Crop Stress Detection.
There are four characteristics of remotely sense data, spatial resolution, spectral resolution, temporal and radiometric resolution. Spatial resolution represents by value of pixel size of the imagery. Spectral resolution represents by number of bands of the imagery. Temporal resolution represents the duration of satellite to visit the same location at certain time. Meanwhile, radiometric resolution is representing the value or size of information that contain in the satellite imagery. For agriculture application, spectral and spatial resolution is very important to fulfil the requirement of analysis. Using multi-spectral or maybe hyper-spectral remote sensing data, we can identify the vegetation or crop behaviour. The reflectance capture by sensor come in several type of wavelength. Common sensor or normal camera only captured the visible wavelength. However, some of satellite sensor or camera can capture infra-red wavelength and consider as multispectral sensor.
With multi-spectral remote sensing data, we can analyse the crop throudh vegetation indices method. Mostly used vegetation indices is Normalize Difference Vegetation Index (NDVI). In order to analyse the crop using vegetation indices, the sensor must consist of visible wavelength (Red, Green and Blue) and Near Infra-Red (NIR) wavelength. Near Infra-Red wavelength is very sensitive with chlorophyll, the lower value of reflectance or Digital Number (DN) from the object represent the lower chlorophyll content. In general, low chlorophyll content represents the stress crop. Due to limitation of multi-spectral satellite imagery availability cause of low temporal resolution and cloud contamination in the imagery, crop stress report from satellite imagery is not really significant and applicable compare to daily agriculture practice.
For example, using high-resolution Worldview-II satellite imagery for crop stress study at paddy field. This satellite platform carries multispectral sensor consist of 8 different wavelength (band). However, the crop stress study can only be made when the satellite imagery is available (free cloud cover) at the study area with most recent time. Mostly, the crop stress study happens for the previous time, not in real time. Using drone technology with multi-spectral sensor, crop stress monitoring can be emphasis in daily agriculture practice in almost real time and crop rehabilitation is more efficient.
Question 3: Discuss the key differences in precision agriculture implementation between annual cropping system and perennial cropping system (specify your choice of crops)
Precision agriculture (PA) can be defined as management of spatial and temporal variability in fields using information and communications technologies. Temporal changes within or between years have been addressed in good agricultural practise by means of laboratory analyses of example spots, while spatial patterns of plant growth, which have also been known for a long time, have been quantified in large scale with the assistance of PA. PA is, therefore, also referred to as site-specific management. This approach considers a management system for farms that aims to increase yield or sustainability. PA can assist farmers, because it permits precise and optimized use of inputs adapted to the apparent plant status, consequently leading to reduced costs and environmental impact. Because the practise provides record trail, enhanced traceability of farm activities can be obtained that consumers and administration increasingly require.
Precision horticulture targets individual trees or zones of tree blocks adaptively to its apparent status that shall trim down environmental footprint of fruit and vegetables production through enhanced resource efficiency and improved production performance. In horticulture, quality analysis of the product is more important than in any other crop. The field size is frequently smaller compared to arable production. The planting density is lower and even single plants may be treated individually adapted to the spatial or temporal pattern. The plant architecture is more complex with planting systems of single rows and missing trees in rows may occur
Horticultural crops are divided into annual and perennial crops. In the latter, the planting system remains stable over years, while morphological adaptation of canopy and root develops according to the environment. Temporal data over more than one season are important, since historical plant data potentially provide valuable information on the status of endogenous growth factors, e.g., the status of phyto-hormone sand assimilates. Horticultural products are the result of many manual operations and hand harvesting. In perennial fruit trees, even additional production measures are requested, e.g., thinning of flowers and fruits, pruning. In orchards, structures for irrigation, hail net or frost protection are limiting the use of methods for soil mapping, e.g., for electromagnetic measurements, which are disturbed by iron installations.
Annual cropping
Applications in mechanically harvested vegetables have also been presented: Pelletier and Upadhyaya (1999) developed a yield monitor for processed tomato using load cells under the conveying chains of the machine. Hofstee and Molema (2002) presented vision system for potato yield mapping. A colour line scan camera above the conveyor belt captured 2D pictures of the potatoes. Correlation between potato size and weight was established and used for estimation of potato flow in the machine. Yield estimated by the sensor compared to yield weighed on the platforms showed good precision between 3.5 and 4.6%. Yield mapping systems for potatoes based on load cells have shown similar good results of approximately 5% measuring uncertainty (Rawlins et al., 1995).
Most horticultural crops are not mechanically harvested and therefore many customised approaches for specific horticultural crops have been tested for yield mapping. In Florida citrus plantations, Schueller et. al. (1999) used a system to weigh palette bins where oranges were collected. Each worker got picking bags to collect fruits picked manually. After filling, bags were emptied in nearby tubs or pallet bins placed between trees (Whitney et al., 1999). Bins were removed by hydraulic lift, which used load cells for weighing, and GPS to record the position of the bin. It was assumed that each bin represented yield of surrounding trees.
A reasonable assumption since workers would empty their bags into the nearest bin. Yield was estimated by dividing weight by area covered by each bin. Position and yield were used to prepare yield maps. Spatial variability of yield was observed in a 3.6 ha orange orchard. Results were confirmed in Mediterranean growing regions in grapefruit (Peeters et al., 2015).
For apple orchards, Aggelopoulou et. al. (2010) mapped yield, where apples were handpicked and placed in 20-kg plastic bins along rows of spindle-formed trees. Each bin was weighed and geo-referenced using DGPS. The bins, corresponding to 5 or 10 trees, were grouped to represent their yield. The estimation of yield of each tree was not possible due to spindle formation, where branches of adjacent trees were coinciding. The system facilitated workers, who picked fruits continuously, and yield mapping did not interfere with their work. The same procedure for yield mapping wasalso performed for pears in a small field of less than 1 ha by Vatsanidou et. al. (2015).
For palm trees, Mazloumzadeh et al. (2010) created yield maps as follows: a few days before harvesting the dates, locations of trees were surveyed and plotted as x-y co-ordinates, fixed at the south-western corner of the grove. Numbers were allocated to all trees located in the grove and, during harvesting, yield of each tree was recorded. In plum, hand-picking was carried out in bins that were transported to the laboratory for single fruit analyses. Spatial pattern of yield and soil ECa was found in an orchard of 180 trees capturing 0.37 ha. Results pointed to low correlation of elevation, soil ECa and generative plant growth (Käthner and Zude-Sasse, 2015). Konopatzki et al. (2015) mapped yield in pear orchard of 5 ha size. They performed selective (n=3) harvests of 36 trees and recorded fruit mass, length and diameter, and soil properties. Results showing high variability of yield with coefficient of variation = 77%, and generally low correlations with soil properties. Perry et al. (2010) carried out yield mapping of pears by weighing total fruit mass picked per tree. They found that yield was strongly spatially clustered, suggesting possible management by zones.
Pozdnyakova et al. (2005) analysed spatial variability of yield in a cranberry plantation. They used 0.3 x 0.3 m frames to measure the number of fruits before harvesting. Using mean berry mass, they estimated the yield. High spatial variability was also observed here.
References
Aggelopoulou, K.D., Wulfsohn, D., Fountas, S., Gemtos, T.A., Nanos, G.D., and Blackmore, S. (2010). Spatial variation in yield and quality in a small apple orchard. Precision Agriculture 11, 538–556. http://dx.doi.org/10.1007/s11119-009-9146-9.
Hofstee, J.W., and Molema, G.J. (2002). Machine vision based yield mapping of potatoes. Paper No. 02-1200 (St. Joseph, MI, USA: ASAE). Perennial Cropping
Mazloumzadeh, S.M., Shamsi, M., and Nezamabadi-pour, H. (2010). Fuzzy logic to classify date palm trees based on some physical properties related to precision agriculture. Precision Agriculture 11, 258–273. http://dx.doi.org/10.1007/s11119-009-9132-2.
Peeters, A., Ben-Gal, A., Gebbers, R., Hetzroni, A., Zude, M., et al. (2015). Getis-Ord’s hot- and cold-spot statistics as a basis for multivariate spatial clustering of tree-based data. Computers and Electronics in Agriculture 111, 140–150. http://dx.doi. org/10.1016/j.compag.2014.12.011.
Pozdnyakova, L., Giménez, D., and Oudemans, P.V. (2005). Spatial analysis of cranberry yield at three scales. Agronomy Journal 97, 49–57. http://dx.doi.org/10.2134/agronj2005.0049.
Rawlins, S.L., Campbell, G.S., Campbell, R.H., and Hess, J.R. (1995). Yield mapping of potato. In Proceedings of Site-Specific Management for Agricultural Systems, P.C. Robert, R.H. Rust, and W.E. Larson, eds. (Madison, WI, USA: ASA, CSA, SSSA). pp. 59–68.
Schueller, J.K., Whitney, J.D., Wheaton, T.A., Miller, W.M., and Turner, A.E. (1999). Low-cost automatic yield mapping in hand-harvested citrus. Computers and Electronics in Agriculture 23, 145–153. http://dx.doi.org/10.1016/S0168-1699 (99)00028-9.
Vatsanidou, A., Fountas, S., Nanos, G., and Gemtos, T. (2014). Variable Rate Application of nitrogen fertilizer in a commercial pear orchard. Fork to Farm: International Journal of Innovative Research and Practice 1(1).
Whitney, J.D., Miller, W.M., Wheaton, T.A., Salyoni, M., and Schueller, J.K. (1999). Precision farming applications in Florida citrus. Applied Engineering in Agriculture 15, 399–403. http://dx.doi.org/10.13031/2013.5795.
0 komen:
Post a Comment