DSpace Collection:http://hdl.handle.net/10174/11292024-03-28T14:32:04Z2024-03-28T14:32:04ZPerformance of machine learning algorithms for forest species classification using WorldView-3 data in the Southern Alentejo region, PortugalCoelho, Ana MargaridaSousa, AdéliaGonçalves, Ana Cristinahttp://hdl.handle.net/10174/358912024-01-08T12:00:40Z2023-08-31T23:00:00ZTitle: Performance of machine learning algorithms for forest species classification using WorldView-3 data in the Southern Alentejo region, Portugal
Authors: Coelho, Ana Margarida; Sousa, Adélia; Gonçalves, Ana Cristina
Abstract: Recent advances in remote sensing technologies and the increased availability of high spatial resolution satellite data allow the acquisition of detailed spatial information. These data have been used for monitoring the Earth's surface, namely monitoring land use land cover, quantifying biomass and carbon, and evaluating the protection and conservation
of forest areas. O WorldView-3 is a high spatial resolution satellite (0.50m) with 8 multispectral bands (visible and
infrared) which allows obtaining detailed data from the Earth's surface.
This study aims to map the forest occupation by specie with two WoldView-3 images, and to evaluate the
performance of machine learning classifiers (maximum likelihood, support vector machine and random forest) in two
regions of Alentejo, south of Portugal. The main forest species are Quercus suber in one region and Quercus
rotundifolia in another. The procedures performed were multiresolution image segmentation and object-oriented classification based on 4 bands (blue, green, red and near infrared). As auxiliary data, vegetation indices (NDVI and SAVI) and principal components were calculated.
In the object-oriented classification process, the three classifiers were tested. The support vector machine classifier was the one that presented the best accuracy (kappa and overall accuracy), for both images, allowing to obtain good results in the identification of forest species. In the image dominated by Quercus suber, the values of kappa and overall
accuracy were 90% and 95%, and for the image where Quercus rotundifolia predominated, 90% and 96% respectively.
The methodology applied to the high spatial resolution satellite data showed very good results in the identification and mapping of main forest species. Higher precision values stand out for the image where the Quercus rotundifolia predominates, where there is less spectral variation, namely fewer land use classes, thus reducing errors between classes that may be spectrally similar.2023-08-31T23:00:00ZTechnological approach to evaluate the livestock trampling effect on soil compactionSerrano, JoãoShahidian, ShakibMarques da Silva, JoséPaniagua, LuísMoral, Franciscohttp://hdl.handle.net/10174/356722023-11-22T09:51:13Z2023-07-01T23:00:00ZTitle: Technological approach to evaluate the livestock trampling effect on soil compaction
Authors: Serrano, João; Shahidian, Shakib; Marques da Silva, José; Paniagua, Luís; Moral, Francisco
Editors: Stafford, John
Abstract: The economic and environmental sustainability of extensive livestock production
systems requires the optimization of soil management, pasture production and animal
grazing. All these aspects are interdependent and linked to soil compaction. This study
aims: (i) to assess the spatial variation of the compaction profile; (ii) to evaluate the
effect of animal trampling on soil compaction; and (iii) to demonstrate the utility of
various technological tools in monitoring indicators of soil characteristics (Cone Index,
CI), of pasture vegetative vigor (Normalized Difference Vegetation Index, NDVI) and
of cows’ grazing zones (Global Positioning Systems, GPS collars). The compaction
resulting from animal trampling was significant outside tree canopy (OTC) in the four
evaluated dates and in the three soil layers considered (0-0.10 m; 0.10-0.20 m; 0.20-
0.30 m). These results suggest that this could be a dynamic process, with recovery
cycles in the face of grazing management, seasonal fluctuations in soil moisture or
spatial variation of specific soil characteristics. The NDVI showed potential for
monitoring the effect of livestock trampling during the peak spring production phase,
with greater vigor in areas with less animal trampling. These results open good
perspectives to support the decision making processes and respond to the challenge of a
holistic and sustainable management of the Montado Mediterranean ecosystem.2023-07-01T23:00:00ZLong-term evaluation of the Grassmaster II probe used to estimate productivity of dryland pasturesSerrano, JoãoShahidian, ShakibMarques da Silva, Joséhttp://hdl.handle.net/10174/356702023-11-22T09:50:01Z2023-07-01T23:00:00ZTitle: Long-term evaluation of the Grassmaster II probe used to estimate productivity of dryland pastures
Authors: Serrano, João; Shahidian, Shakib; Marques da Silva, José
Editors: Stafford, John
Abstract: The estimation of pasture productivity is a tool of great interest for the management of
animal grazing. The standard method of assessing pasture mass requires great effort and
expense to collect enough samples to accurately represent a pasture. This work presents
the results of a long-term study to calibrate a Grassmaster II capacitance probe to
estimate pasture productivity in two phases: (i) the calibration phase (2007-2018) which
included measurements in 1411 sampling points in three parcels; and (ii) the validation
phase (February and March 2019) which included measurements in 48 sampling points
in four parcels. A regression analysis was performed between the capacitance (CMR)
measured by the probe and values of pasture green matter and dry matter (respectively,
GM and DM, in kg ha-1
). The results showed significant correlations between GM and
CMR and between DM and CMR, especially in the early stages of pasture growth cycle.
The analysis of the data grouped by classes of pasture moisture content (PMC) shows
higher correlation coefficients for PMC content > 80% (r = 0.775; p <0.01; RMSE =
4806 kg ha-1
and CVRMSE=28.1% for GM; r = 0.750; p <0.01; RMSE = 763 kg ha-1
and
CVRMSE=29.7% for DM), with a clear tendency for the accuracy to decrease when the
pasture vegetative cycle advances and, consequently, the PMC decreases. The
validation of calibration equations when PMC > 80% showed a good approximation
between GM or DM measured and GM or DM predicted (r=0.908; p<0.01; RMSE =
4293 kg ha-1
; CVRMSE=24.4% for GM; r = 0.904; p <0.01; RMSE = 590 kg ha-1
and
CVRMSE=20.4% for DM). It can be concluded that (i) the capacitance probe is an
expedient tool that can enable the farm manager to estimate pasture productivity with
acceptable accuracy and support the decision-making process in the management of
dryland pastures; (ii) the more favorable period for the use of this probe in dryland
pastures in a Mediterranean climate, such as the Portuguese Alentejo coincides with the
end of winter and beginning of spring (February-March), corresponding to PMC > 80%.2023-07-01T23:00:00ZVineyard management zones: A case study based on soil apparent electrical conductivity surveySerrano, JoãoMau, VascoRodrigues, RodrigoPaixão, L.Shahidian, ShakibMarques da Silva, JoséPaniagua, LuísMoral, F.http://hdl.handle.net/10174/356672023-11-22T09:48:12Z2023-09-03T23:00:00ZTitle: Vineyard management zones: A case study based on soil apparent electrical conductivity survey
Authors: Serrano, João; Mau, Vasco; Rodrigues, Rodrigo; Paixão, L.; Shahidian, Shakib; Marques da Silva, José; Paniagua, Luís; Moral, F.
Editors: Pérez-Ruiz, Manuel; Cegarra, Gregorio; Urrestarazu, Luiz; Diaz de la Torre, Isabel; Lizana, Antonio; Salas, Antonio; Ortega, Manuel; Martos, Rocio; Fernandez, Alejandro
Abstract: In the current context of increasing costs of production factors, it is essential to optimize the management of available resources, seeking to incorporate technologies that improve knowledge of the variables involved in the agronomic production process. In this study, carried out in a 3.3 ha vineyard located near Évora, in the South of Portugal, a contact sensor (“Veris 2000 XA”) was used to map soil apparent electrical conductivity (ECa). A precision altimetric survey of the field was also carried out with a global navigation satellite system receiver (GNSS). The results of these surveys were submitted to geostatistical treatments that allowed the definition of three management zones (MZ; less, intermediate and more productive potential). The validation of these MZ was carried out by laboratory analysis of soil samples (texture, pH, organic matter ˗ OM, moisture content, phosphorous, potassium, exchange bases and cation exchange capacity ˗ CEC). All these soil parameters proved the validity of the MZ (of less, intermediate and more productive potential) defined from the ECa and altimetric survey. This validation attests to the interest of expeditious technological tools for monitoring ECa as a fundamental step in implementing smart agronomic decision-making processes.2023-09-03T23:00:00Z