ABSTRACT Nutrient prediction models applied to tree species from Brazilian Caatinga can be a crucial tool in understanding this biome. The study aimed to fit a mixed model to predict nitrogen (N), phosphorus (P), and potassium (K) content in tree species native to the Caatinga biome located in Floresta municipality, Pernambuco State – PE, Brazil. The following species were considered the area’s most important and evaluated in the present study: Poincianella bracteosa (Tul.) L.P.Queiroz, Mimosa ophtalmocentra Mart. ex Benth, Aspidosperma pyrifolium Mart, Cnidoscolus quercifolius (Mull. Arg.) Pax. & Hoffm, and Anadenanthera colubrina var. cebil (Griseb.) Altschul. Four trees, representing the average circumference in each diameter class, were harvested for NPK quantification. The Spurr model was evaluated for NPK prediction, and species inclusion as a random effect was significant (p > 0.05) in all models. The Spurr model with fixed and random effects presented better statistics than fixed-effect models in all parameters for all nutrients. Generated NPK predicting equations can be a handy tool to understand the impact of wood extraction over Caatinga’s biogeochemical cycles and guide forest management strategies in semi-arid regions of the world.

RESUMO Modelos de predição de nutrientes aplicados a espécies arbóreas da Caatinga brasileira podem ser uma ferramenta crucial para a compreensão do bioma. O estudo teve como objetivo ajustar um modelo misto para prever os teores de nitrogênio (N), fósforo (P) e potássio (K) em espécies arbóreas nativas do bioma Caatinga localizadas no município de Floresta, Pernambuco – PE, Brasil. As seguintes espécies foram as mais importantes da área e avaliadas no presente estudo: Poincianella bracteosa (Tul.) L.P.Queiroz, Mimosa ophtalmocentra Mart. ex Benth, Aspidosperma pyrifolium Mart, Cnidoscolus quercifolius (Mull. Arg.) Pax. & Hoffm e Anadenanthera colubrina var. cebil (Griseb.) Altschul. Quatro árvores, representando a circunferência média em cada classe de diâmetro, foram colhidas para quantificação de NPK. O modelo Spurr foi avaliado para predição de NPK e a inclusão de espécies como efeito aleatório foi significativa (p > 0,05) em todos os modelos. O modelo de Spurr com efeitos fixos e aleatórios apresentou estatísticas melhores que os modelos de efeito fixo em todos os parâmetros para todos os nutrientes. As equações de previsão de NPK geradas podem ser uma ferramenta útil para entender o impacto da extração de madeira sobre os ciclos biogeoquímicos da Caatinga e orientar estratégias de manejo florestal em regiões semiáridas do mundo.

Resumo Estudos que visam quantificar e analisar a dinâmica do crescimento da vegetação, fornecem informações para o manejo e conservação das florestas ao longo do tempo. Sendo assim, o objetivo principal deste trabalho foi analisar a dinâmica e o crescimento em área basal, em diferentes intervalos (2008-2016; 2008-2017 e 2008-2019), em duas áreas com diferentes históricos de perturbação e uso, no município de Floresta - PE. Foram instaladas 40 parcelas permanentes de 20 m x 20 m, com coleta de dados do ano de 2008 até 2019, em duas áreas, A1: menos conservada e A2: mais conservada, em que foram medidos os indivíduos e fustes com circunferência a 1,30 m do solo ≥ 6 cm. Foi verificado o número de indivíduos e fustes por espécies, calculado a área basal nos diferentes intervalos, taxas de ingressos, mortalidade, crescimento bruto e líquido, incluindo e excluindo ingressos. Foi constatado decréscimo em número de indivíduos e fustes ao longo dos anos para ambas as áreas, no entanto, houve aumento na área basal da A1, 2008: 3,1 m2.ha-1; 2016: 3,8 m2.ha-1; 2017: 3,9 m2.ha-1 e 2019: 4,6 m2.ha-1, já na A2 ocorreu o inverso, com área basal inicial de 8,3 m2.ha-1; 6,4 m2.ha-1; 6,4 m2.ha-1; e 7,0 m2.ha-1, respectivamente, verificou-se ainda que nos anos de menos chuvas a área basal teve valor médio menor, o inverso é verdadeiro, maior volume precipitado significa aumento de área basal. Apesar da A1 ter se sobressaído, as altas taxas de mortalidade de fustes e indivíduos em todos os intervalos estudados fez com que a área basal diminuísse gradativamente, o crescimento líquido foi afetado, além disso nos anos de menores precipitações pluviométricas houve valores menores da área basal, evidenciando a dependência climática, impactando na dinâmica florestal.

Abstract Studies that aim to quantify and analyze the dynamics of vegetation growth provide information for the management and conservation of forests over time. Therefore, the main objective of this work was to analyze the dynamics and growth in basal area, at different intervals (2008-2016; 2008-2017 and 2008-2019), in two areas with different disturbance and use histories, in the municipality of Forest - PE. Forty permanent plots measuring 20 mx 20 m were installed, with data collection from the year 2008 to 2019, in two areas, A1: less conserved and A2: more conserved, in which individuals and shafts with a circumference of 1.30 were measured m from the ground ≥ 6 cm. The number of individuals and stems per species was verified, calculating the basal area at different intervals, rates of entry, mortality, gross and net growth, including and excluding entries. A decrease in the number of individuals and stems over the years was observed for both areas, however, there was an increase in the basal area of the A1, 2008: 3.1 m2.1.6ha-1; 2016: 3.8 m2.1.6ha-1; 2017: 3.9 m2.1.6ha-1 and 2019: 4.6 m2.1.6ha-1, while in (A2) the reverse occurred, with an initial basal area of 8.3 m2.1.6ha-1 ; 6.4 m2.1.6ha-1; 6.4 m2.1.6ha-1; and 7.0 m2.1.6ha-1, respectively, it was also found that in years with less rain the basal area had a lower mean value, the inverse is true, higher precipitated volume means an increase in basal area. Although A1 stood out, the high mortality rates of stems and individuals in all studied intervals caused the basal area to gradually decrease, the net growth was affected. basal, showing climate dependence, impacting forest dynamics.

Resumen La clasificación de la madera aserrada es una de las actividades más subjetivas de la fase final del proceso de aserrado de trozas en los aserríos. La presente investigación tuvo como objetivo proponer herramientas de auxilio en la clasificación de madera aserrada de coníferas. El estudio se realizó en el aserrío Combate de Tenerías de la Empresa Forestal Integral (EFI) Macurije, Pinar del Río, Cuba. Los datos utilizados proceden de 259 tablas de Pinus caribaea var. caribaea clasificadas siguiendo las exigencias (24 variables) y clases establecidas por el clasificador de madera aserrada de coníferas utilizado en Cuba. Se procedió al ajuste de un modelo de Regresión Logística Multinomial Ordinal (RLMO) y al entrenamiento de Redes Neuronales Artificiales (RNAs). Los parámetros del modelo RLMO fueron estimados usando el método de máxima verosimilitud optimizada con el algoritmo de Newton-Raphson. El entrenamiento de las RNAs se realizó con el algoritmo Broyden-Fletcher-Goldfarb-Shanno (BFGS). El problema de multicolinearidad, casi siempre presente en modelaciones con numerosas variables predictivas, fue abordado con el análisis factorial, utilizando los factores retenidos como entradas de los modelos. Con base en el porcentaje de aciertos en la clasificación, la RNA RBF 24-8-4 fue superior a las ecuaciones de regresión logística ordinal.

Abstract Lumber classification is one of the most subjective activities of the final phase of log sawing process in sawmills. The objective of this research was to propose tools that assist in conifers lumber grading. The research was carried out at the sawmill Combate de Tenerías of Macurije integrated forest company, Pinar del Río, Cuba. The data used comes from 259 lumber pieces of Pinus caribaea var. caribaea classified following the requirements (24 variables) and classes established by the conifers lumber grader used in Cuba. We proceeded to fit a Multinomial and Ordinal Logistic Regression (MOLR) model and train Artificial Neural Networks (ANNs). The parameters of the MOLR model were estimated using the maximum likelihood method optimized with the Newton Raphson algorithm. ANNs were trained with Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm. The multicollinearity problem, usually present in modeling with numerous predictive variables, was addressed with factor analysis, using the factors retained as inputs to the models. Based on the percentage of correct classification, the ANN RBF 24-8-4 was superior to the ordinal logistic regression equations.

Abstract This study aimed to evaluate decomposition and release of C and of N from the litter of forest Caatinga species in the humification phase, as well as to estimate the residence time of C and N. Litter decomposition of the species was evaluated in 2014/2015 in the municipality of Floresta, Pernambuco, Brazil, using a method of the litter bags at 270; 300; 330; and 360 day. Litter from species of the Fabaceae family presented different decomposition groups, suggesting that decomposition was more influenced by the C/N ratio than N concentration only. The N release was 6.25% greater than the release of C. C will remain in the litter for more time, while N will be almost totally released in approximately 1.14 years. The species litter decomposition were shared, suggesting that the nutrients cycling is fundamental in the preservation of the Caatinga, mainly due to the longer residence time of C.

Resumo O banco de sementes no solo é um dos mecanismos de regeneração natural que contribui para o surgimento de espécies em áreas florestais ou degradadas, no entanto, a composição de espécies pode ser influenciada pelo histórico de uso da terra, pela matriz em que o fragmento está inserido, entre outros. O objetivo dessa pesquisa foi avaliar a composição florística, abundância e ecologia morfofuncional de plântulas de espécies arbóreas encontradas no banco de sementes no solo em fragmento de Floresta Tropical. Foram retiradas amostras de solos e levadas para o viveiro florestal e cobertas por tela sombrite 70%. As plântulas emergidas foram contadas, identificadas e classificadas quanto à síndrome de dispersão, grupo sucessional e ecologia morfofuncional dos cotilédones. Emergiram 3.441 plântulas, em 14 famílias e 28 morfoespécies. Predominaram Miconia prasina e Cecropia pachystachya com 22,20 e 17,78%, respectivamente; a síndrome zoocórica com 64%; houve predominância de espécies secundárias iniciais, seguida de pioneiras. 78% das plântulas são do tipo PEF (fanero-epígeo-foliáceo). Embora o remanescente esteja inserido em uma matriz agrícola possui uma média riqueza de espécies arbóreas no banco de sementes, sendo a maioria das espécies classificadas como fanero-epígeo-foliáceo, demonstrando a dinâmica e capacidade de autorregeneração da área estudada.

Abstract The seed bank in the soil is one of the mechanisms of natural regeneration that contributes to the emergence of species in forest or degraded areas, however, the composition of species can be influenced by the history of land use, by the matrix in which the fragment is inserted, among others. The objective of this research was to evaluate the floristic composition, abundance and morphofunctional ecology of seedlings arboreal species found in the soil seed bank in tropical forest fragment. Soil samples were taken and taken to the forest nursery and covered by shading screens 70%. The emerged seedlings were counted, identified and classified in terms of dispersion syndrome, successional group and morphological and functional ecology of cotyledons. 3,441 seedlings emerged in 14 families and 28 morphospecies. Miconia prasina and Cecropia pachystachya predominated with 22.20 and 17.78%, respectively; zoochoric syndrome with 64%; there were predominance of early secondary species, followed by pioneers. 78% of the seedlings are of the PEF type (phanero-epigeal-foliaceous). Although the remnant is inserted in an agricultural matrix, it has an average tree species richness in the seed bank, most species being classified as phanero-epigeal-foliaceous, demonstrating the dynamics and self-regeneration capacity of the studied area.

Abstract This work was carried out with the objective of evaluating, technically and economically, timber extraction in plantations of Pinus caribaea var. caribaea of Macurije forest company located in the province of Pinar del Río, Cuba. For this, time and movement studies were carried out using "productivity" and "unit cost" of wood extraction as evaluation criteria. The skidders and oxen performance evaluation, in different harvest systems and slope classes, was performed through univariate and multivariate (cost and/or productivity) factorial experiments (3 × 3 × 2), following a randomized block design (B1 = 21 years and B2 = 32 years), with the factors "machinery" and "slope" with three and "harvest systems" with two levels. The feasibility and usefulness of the multivariate approach in the evaluation of timber extraction were evidenced; timber extraction is more viable with the tree length harvest systems in stands of 32 years. The interaction machinery-slope had a significant influence on the variation of the whole cost-productivity. The John Deere 548E skidder, for presenting lower unit cost and higher productivity than the J-80A, is recommended for semi-mechanized timber extraction. The most viable is the yoke of oxen because of its greater profitability. However, due to its low productivity, it can only be used in combination with other machines or during periods of decelerated production. Its exclusive use is impracticable since it would imply significantly increasing its quantity to compensate for its low productivity.

Resumen El presente trabajo se realizó con el objetivo de evaluar, técnica y económicamente la extracción de madera en plantaciones de Pinus caribaea var. caribaea de la empresa forestal Macurije, ubicada en la provincia de Pinar del Río, Cuba. Para ello, se llevaron a cabo estudios de tiempos y movimientos teniendo como criterios de evaluación "productividad" y "costo unitario" de extracción de la madera. La evaluación del desempeño de los tractores articulados (skidders) y de las yuntas de bueyes en diferentes sistemas de aprovechamiento y clases de pendiente se realizó por medio de experimentos factoriales (3 × 3 × 2) univariados y multivariados (costo unitario e/o productividad), siguiendo un diseño al azar en bloques (B1 = 21 años y B2 = 32 años), con los factores "maquinaria" y "pendiente" con tres niveles y "sistema de aprovechamiento" con dos niveles. Se evidenciaron la viabilidad y la utilidad del enfoque multivariado en la evaluación de la extracción de madera; es más viable extraer la madera con el sistema de tronco entero en plantaciones de 32 años. La interacción maquinaria-pendiente tuvo influencia significativa en la variación del conjunto costo-productividad. El skidder John Deere 548E, por presentar menor costo unitario y mayor productividad que el J-80A, es el recomendado para la extracción semimecanizada de la madera. El más viable es la yunta de bueyes por su mayor rentabilidad. Sin embargo, por su baja productividad, solo podrá ser utilizado en combinación con otras máquinas o en periodos de producción desacelerada. Su uso exclusivo no es práctico ya que implicaría aumentar significativamente su cantidad para compensar su baja productividad.

Resumo Procurou-se estabelecer a relação entre o Índice de área foliar no campo (IAFc) de plantios clonais de Eucalyptus saligna Smith e três diferentes Índices de Vegetação (IV) obtidos de uma imagem Landsat 8/OLI: Índice de Vegetação da Diferença Normalizada (Normalized Difference Vegetation Index, em inglês - NDVI), Índice da Razão Simples (Simple Ratio Index - SRI) e Índice de Vegetação Ajustado para o Solo (Soil Adjusted Vegetation Index - SAVI), com o objetivo de selecionar o melhor estimador do IAF por sensoriamento remoto (IAFSR), obtendo assim a espacialização IAF nos talhões. O IAFc foi obtido utilizando o equipamento LAI-2000 e seu comportamento foi analisado em diferentes idades. Os índices de vegetação foram obtidos por meio de aritmética das bandas 4 e 5 do sensor. A análise de regressão linear simples foi utilizada para ajustar o modelo de estimativa de IAFSR (IAFSRi= β0 + β1 . IVi + εi), sendo os critérios de escolha as estatísticas de R2aj%, Syx% e análise de resíduos. Os resultados mostraram que o índice que melhor estimou o IAFSR foi o SRI (IAFSR=-5,6159 + 0,9716 . SRI), com R2aj%=67,0 e Syx%=12,5. Todos os modelos ajustados mostraram tendência em sobrestimar o IAF em valores inferiores a dois e subestimar o IAF a partir de valores superiores a 3,5 (NDVI e SRI) e superiores a três para SAVI. As equações obtidas para as diferentes idades não produziram melhora nas estimativas de IAFSR.

Abstract The relationship between the ground Leaf area index (LAIg) from clonal plantations of Eucalyptus saligna Smith and three different vegetation indices (VI): Normalized Difference Vegetation Index (NDVI), Simple Ratio Index (SRI) and Soil Adjusted Vegetation Index (SAVI) was evaluated in order to select the best IV to estimate the IAF by remote sensing (LAIRS), and obtaining the spatial distribution of LAI in the stands. LAIg was measured using LAI-2000 and its behavior was examined at different ages. The vegetation indices were obtained from a Landsat 8/OLI through the arithmetic of the bands 4 and 5. The linear regression analysis was used to adjust the model LAIRS (LAIRSi= β0 + β1 . IVi + εi), and the criteria for selecting the best equation were the statistics R2aj%, Syx% and residual analysis. The results showed that the best vegetation index to estimate IAFSR was SRI (LAIRS =-5.6159 + 0.9716 . SRI ), resulting R2adj%=67.0 and Syx=12.5%. The results of all adjusted models tended towards overestimation of LAI in values lower than two and underestimation in values above 3.5 (NDVI e SRI) and above three for SAVI. The equations for different ages produced no improvement in LAIRS.

ABSTRACT The objective of this study was to obtain regression equations and artificial neural networks (ANNs) for prediction and prognosis of the yield of Pinus caribaea var. caribaea Barrett & Golfari. The data used for modeling comes from measuring the variables diameter at breast height (DBH) and total height (Ht) in 550 temporary plots and 14 circular permanent plots with 500 m2 in Pinus caribaea var. caribaea plantations, aged between 3 and 41 years old. In growth prediction, the results indicated Schumacher model as the best fit to the data. On prognosis, the modified Buckman system was better than Clutter’s. ANNs presented a similar performance to the Buckman model in volume prognosis, however these were superior for basal area prognosis.

ABSTRACT The aim of this study was to analyze the dynamics of the community and shrub-woody individuals in a dry forest in the Brazilian semiarid region after 29 years of vegetation suppression. Individuals with a circumference at 1.30 m above the ground (CBH) ≥ 6 cm were counted and their heights and CBH were measured in 40 permanent plots on three monitoring occasions (2011, 2013 and 2015). In the period between 2011 and 2015 the area presented a below average precipitation for the last 30 years. Phytosociological parameters and diversity were estimated. The density of individuals decreased from 2011 to 2015 (p < 0.01) and the basal area was similar. During the monitoring period, Poincianella bracteosa stood out in density, frequency and dominance. Phytosociological changes were more evident for an interval of four years. After 29 years of vegetation suppression, the community diversity is within the values for preserved semiarid (caatinga) vegetation. A long period of drought was the main disturbance factor affecting vegetation dynamics.

ABSTRACT The aim of this study was to determine the carbon content (C) in stem, leaves and thin and thick branches of eight species in an area of Caatinga, municipality of Floresta, Pernambuco, Brazil. To do so, dry biomass samples of at least 15 trees distributed into five diametric classes collected from the study area and fractionated in laboratory were used for each of the eight evaluated species. Carbon content was determined using CHNS/O elemental analyzer. Average carbon content of 46.4% is recommended for Anadenanthera colubrina, Aspidosperma pyrifolium, Bauhinia cheilantha, Cnidoscolus quercifolius, Mimosa ophthalmocentra and Mimosa tenuiflora, without need for specific values for each compartment. Croton heliotropiifolius and Poincianella bracteosa require distinct values per compartment for individuals in order to avoid trends in carbon stock estimates.

Abstract In this paper, procedures are proposed for determining the technical, economic and technical-economic rotations in forest plantations. The classical criteria of maximum productivity were determined: annual average increase and maximum land expected value. For the determination of the rotations, in addition to using the analytical tools (by adjusting forest growth models) and graphical tools conventionally used, mathematical programming models were proposed: model R-01 based on integer linear programming for the determination of the mono-criteria rotations and the models R-02 and R-03 based on weighted integer goal programming and lexicographic integer goal programming, respectively, for multi-criteria rotation determination. The validity of the proposed models was verified through the consistency of the technical, economic and technical-economic rotation ages determined by diverse site index in Pinus caribaea Morelet var. caribaea Barr. & Golf. plantations. The technically optimal rotations varied between 31 years and 35 years according to the site index and the economically optimal rotations between 18 years and 21 years. It was noticed that the technical-economic rotations are highly influenced by the criterion of greater importance in both approaches. The results found for Pinus caribaea var. caribaea with these approaches indicated intermediate rotations between the technical and the economic ones, these varying between 23 years and 25 years.

Resumen En el presente trabajo se proponen procedimientos de determinación de los turnos técnicos, económicos y técnico-económicos en plantaciones forestales. Se procedió a la determinación de los criterios clásicos de máxima productividad: incremento medio anual y máximo valor esperado de la tierra. Para la determinación de los turnos, además de utilizar las herramientas analíticas (por ajuste de modelos de crecimiento forestal) y gráficas convencionalmente utilizadas, se propusieron modelos de programación matemática: el modelo R-01 basado en la programación lineal entera, para la determinación de los turnos monocriterios, y los modelos R-02 y R-03 basados en la programación por metas ponderada entera y en la programación por metas lexicográficas entera, respectivamente, para la determinación del turno con múltiples objetivos. La validez de los modelos propuestos fue verificada a través de la consistencia y realismo biológico de las edades de rotación técnicas, económicas y técnico-económicas determinadas por índice de sitio en plantaciones de Pinus caribaea Morelet var. caribaea Barr. & Golf. de la empresa Macurije. Los turnos técnicamente óptimos variaron entre 31 años y 35 años según la calidad del sitio y los turnos económicamente óptimos entre 18 años y 21 años. Se percibió que los turnos técnico-económicos son altamente influenciados por el criterio de mayor importancia, tanto en el enfoque de metas ponderadas como en el de metas lexicográficas. Los resultados encontrados para el Pinus caribaea var. caribaea con esos enfoques indicaron turnos intermedios entre los técnicos y los económicos, variando estos entre 23 años y 25 años.

RESUMO Mecanismos de regeneração natural são fundamentais para o restabelecimento das funções ecológicas de fragmentos florestais e de áreas degradadas. O presente trabalho teve como objetivo avaliar a chuva de sementes e o estabelecimento de plântulas de espécies arbóreas em Floresta Atlântica, localizada na região Nordeste do Brasil. Para amostragem da chuva de sementes foram alocados 60 coletores (área de 0,196 m2), ficando equidistantes, aproximadamente, de 50 m. As coletas foram realizadas com intervalo de 30 dias por um período de 12 meses. Para avaliar o estabelecimento de plântulas foram alocadas subparcelas ao lado dos coletores, totalizando 60 subparcelas de 0,25 m2 cada. Os indivíduos arbóreos identificados pelos diásporos coletados na chuva de sementes e pelas plântulas estabelecidas no fragmento foram contados e classificados quanto à síndrome de dispersão e grupo sucessional. A chuva de sementes foi representada por 60 morfoespécies e 20 famílias botânicas. O grupo ecológico e a síndrome de dispersão predominante foram de espécies secundárias iniciais e zoocórica, respectivamente. Em relação às plântulas estabelecidas foram contadas 122 plântulas, sendo Tapirira guianensis L. com maior número de plântulas. A chuva de sementes apresenta potencial para manutenção da dinâmica florestal por apresentar alta riqueza de espécies que compõe a atual vegetação, sendo composta, predominantemente, por espécies que produzem frutos abundantes com dispersão zoocórica que servem de alimento para fauna. Enquanto poucas sementes germinam para formar o banco de plântulas, a maioria está contribuindo para formação do banco de sementes.

ABSTRACT Mechanisms of natural regeneration are fundamental for ecological function restoration in forest fragments and degraded areas. The present work aimed to evaluate the seed rain and the establishment of seedlings of arboreal species in the Atlantic Forest, located in the northeast of Brazil. For sampling of the seed rain, 60 collectors were allocated (area of 0.196 m2), being equidistant, approximately, 50 m. The samples were collected at intervals of 30 days for a period of 12 months. To evaluate the establishment of seedlings, subplots were allocated next to the collectors, totaling 60 subplots of 0.25 m2 each. The arboreal individuals identified by the diaspores collected in the seed rain and by the seedlings established in the fragment were counted and classified for the dispersion syndrome and successional group. The seed rain was represented by 60 morphospecies and 20 botanical families. The ecological group and the predominant dispersion syndrome were of early secondary and zoocoric species, respectively. In relation to the established seedlings, 122 seedlings were counted, being Tapirira guianensis L. with higher number of seedlings. The rainfall of seeds presents potential for the maintenance of the forest dynamics due to the high species richness that compose the current vegetation. Being predominantly composed of species that produce abundant fruits with zoocoric dispersion that serve as food for fauna. While few seeds germinate to form the seedling bank, most of them are contributing to seed bank formation.

Objetivou-se avaliar características morfológicas de clones de Panicum maximum durante o período de rebrota após a interrupção da irrigação. Os clones de Panicum foram coletados em Sergipe e em três locais de Pernambuco ("São Bento do Una", "Itambé" e "São João da Coroa Grande"), além das cultivares 'Mombaça', 'Tanzânia' e 'Massai'. Os tratamentos consistiram de sete clones submetidos a 7, 14, 21 ou 28 dias de suspensão da irrigação. O delineamento utilizado foi o inteiramente casualizado em arranjo fatorial 7 × 4 com quatro repetições. Foram avaliados o número de perfilhos, altura da planta, comprimento da lâmina foliar, número de folhas verdes por perfilho e número de folhas mortas por perfilho. O fator clone afetou significativamente o número de perfilhos por vaso e o número de folhas mortas por perfilho (P≤0,05). O clone de Sergipe produziu mais perfilhos (8.39) que os demais após 28 dias de interrupção da irrigação, enquanto a cultivar 'Mombaça' apresentou mais folhas mortas (5.39) após 7 dias sem irrigação. Em geral, a interrupção da irrigação por 28 dias resultou em plantas mais altas com lâminas foliares maiores e com número superior de folhas verdes por perfilho.

This study aimed to evaluate the morphological characteristics of Panicum maximum clones during the regrowth period following cessation of irrigation. Panicum clones were collected in Sergipe and in three Pernambuco locations (‘São Bento do Una’, ‘Itambé’ and ‘São João da Coroa Grande’), in addition to ‘Mombaça’, ‘Tanzânia’ and ‘Massai’ cultivars. The treatments consisted of seven clones, which were subjected to either 7, 14, 21 or 28 days of water suspension. The experimental design was completely randomized in a factorial 7 × 4 design, with four replicates. The morphological characteristics evaluated included the number of tillers, plant height, leaf blade length, number of green leaves per tiller and number of dead leaves per tiller. The clone factor significantly affected the number of tillers per pot and the number of dead leaves per tiller (P ≤ 0.05). The clone from Sergipe produced a greater number of tillers (8.39) than the other clones following 28 days of water suspension, while the ‘Mombaça’ cultivar had a greater number of dead leaves (5.39) after 7 days of water suspension. In general, the 28-day cessation of irrigation period resulted in taller plants with larger leaf blades and a greater number of green leaves per tiller.

RESUMO Garcinia gardneriana é uma espécie florestal nativa do Bioma Mata Atlântica, com potencial para aplicação industrial, medicinal, ornamental e madeireira, cujas sementes têm dificuldades para germinar. Assim, o objetivo do presente trabalho foi avaliar diferentes métodos para superar a dormência de sementes de Garcinia gardneriana de modo eficaz e com baixo custo para promover a germinação rápida e uniforme. Foram testados os seguintes tratamentos: testemunha - sementes intactas (T1); sementes sem tegumento (T2); sementes sem tegumento, seguido de imersão em solução de ácido giberélico (GA3) a 500 mg.L-1 por 24 horas (T3); sementes sem tegumento, seguido de imersão em solução de nitrato de potássio (KNO3) a 0,2% por 24 horas (T4); semente sem tegumento, seguido de estratificação a 10°C por 120 horas (T5); sementes com tegumento e imersas em solução de KNO3 a 0,2% por 24 horas (T6); sementes submetidas à estratificação a 10°C durante 120 horas (T7); sementes escarificadas com lixa para massa n°80 no lado oposto ao hilo (T8). Para a superação da dormência foram avaliadas as seguintes variáveis: emergência total, primeira contagem de emergência, índice de velocidade de emergência (IVE), tempo médio de emergência, comprimento da parte aérea e raiz primária e massa seca da parte aérea e do sistema radicular das plântulas. A retirada do tegumento favorece a emergência de plântulas mais vigorosas oriundas de sementes de Garcinia gardneriana, por isso, apresentam dormência tegumentar (exógena). A imersão das sementes em solução de ácido giberélico pode acelerar a velocidade de emergência, mas não influencia o tempo médio de emergência. Os tratamentos mais eficazes para superação da dormência das sementes de Garcinia gardneriana foram a remoção do tegumento, seguido de imersão em solução de ácido giberélico (GA3) a 500 mg.L-1 por 24 horas ou apenas a remoção do tegumento.

ABSTRACT Garcinia gardneriana is a native tree species of the Atlantic Forest biome with commercial potential, industrial application, medicinal, ornamental and timber, which seeds have difficulties on germination. In this sense, the objective of the present study was to evaluate different methods to overcome dormancy efficiently and at low cost to promote rapid and uniform seed germinations of this species. In order to study the influence of different methods to Garcina gadneriana overcome dormancy, the following treatments were tested: control, intact seeds without any treatment (T1); seeds without seed coat (T2); seeds without seed coat, followed by immersion in a solution of gibberellic acid (GA3) 500 mg.L-1 for 24 hours (T3); seeds without seed coat, followed by immersion in a solution of potassium nitrate (KNO3) at 0,2% for 24 hours (T4); seeds without seed coat, followed by laminating at 10°C for 120 hours (T5); seeds with seed coat and soaked in solution of potassium nitrate (KNO3) at 0,2% for 24 hours (T6); stratifying the seeds subjected to 10°C for 120 hours (T7); seeds scarified with sandpaper to mass nº 80 opposite the hilum (T8). For dormancy overcome were evaluated: total emergency, first emergency count, emergency speed index, mean emergency time, length and dry weight of seedlings root and seedlings root. The removal of the seed coat favors the emergence of more vigorous seedlings derived from Garcinia gardneriana seeds, so exhibit dormancy coats (exogenous). The seed soaking in gibberellic acid solution can accelerate the emergence speed index but does not influence the mean emergency time. It is recommended as treatments to dormancy overcome of Garcinia gardneriana, the seeds without seed coat, followed by immersion in a solution of gibberellic acid (GA3) 500 mg.L-1 for 24 hours or only seeds without seed coat to be a more practical treatment.

ABSTRACT The present study was carried out with the objective of obtaining regression equations and Artificial Neural Networks (ANNs) for the prognosis of Pinus caribaea var. caribaea survival in Macurije Forest Company, province of Pinar del Río - Cuba. The data used in the modeling comes from the measurement of the variables age (years) and survival (density) in circular permanent plots of 500 m² established in P. caribaea var. caribaea plantations. The study was divided into three stages: i) Adjustment of survival traditional regression models; ii) Training of ANNs for survival prognosis, including categorical variables «site» and «Basic Units of Forest Production»; iii) Comparison of regression equations performance with those of ANNs in survival prognosis. The best models and ANNs were selected based on: adjusted determination coefficient - R2 aj (%), square root of the mean square error - RMSE (%) and residue distribution analysis. The evaluation of the models goodness of fit also included the verification of the assumptions of normality, homocedasticity and absence of serial autocorrelation in the residues by Kolmogorov-Smirnov, White and Durbin-Watson tests, respectively. The model of Pienaar and Shiver, in 1981 turned out to be the best fit in survival prognosis. The ANN MLP 13-10-1 was the one with the best generalization capacity and presented a performance similar to that of Pienaar and Shiver equation.

RESUMEN El presente estudio fue realizado con el objetivo de obtener ecuaciones de regresión y Redes Neuronales Artificiales (RNAs) para la proyección de la supervivencia de Pinus caribaea var. caribaea en la empresa forestal Macurije, provincia de Pinar del Río, Cuba. Los datos utilizados en las modelaciones provienen de la medición de las variables edad (años) y supervivencia (densidad) en parcelas permanentes circulares de 500 m², establecidas en plantaciones de P. caribaea var. caribaea. El estudio se dividió en tres etapas: 1) Ajuste de modelos de regresión tradicionales para proyección de supervivencia; 2) Entrenamientos de RNAs para la proyección de supervivencia, incluyendo variables categóricas «sitio» y «Unidades Básicas de Producción Forestal»; 3) Comparación de los desempeños de las ecuaciones de regresión con los de las RNAs en la proyección de la supervivencia. Los mejores modelos y RNAs fueron seleccionados y basados en: coeficiente de determinación ajustado -R2 aj (%), raíz cuadrada del error medio cuadrático - RMSE (%) y análisis de distribución de residuos. La evaluación de la bondad de ajuste de los modelos también incluyó la verificación de los supuestos de normalidad, homocedasticidad y ausencia de autocorrelación serial en los residuos por las pruebas de Kolmogorov-Smirnov, White y Durbin-Watson, respectivamente. El modelo de Pienaar y Shiver en 1981 resultó ser el de mejor ajuste en la proyección de la supervivencia. La RNA de arquitectura MLP 13-10-1 fue la de mejor capacidad de generalización y presentó un desempeño similar al de la ecuación obtenida del ajuste del modelo de Pienaar y Shiver.